The formation of the Pentateuch : bridging the academic cultures of Europe, Israel, and North America 9783161538834, 3161538838

Citation preview

Forschungen zum Alten Testament Edited by Konrad Schmid (Zürich) · Mark S. Smith (Princeton) Hermann Spieckermann (Göttingen)

111

The Formation of the Pentateuch Bridging the Academic Cultures of Europe, Israel, and North America Edited by

Jan C. Gertz, Bernard M. Levinson, Dalit Rom-Shiloni, and Konrad Schmid

Mohr Siebeck

Jan C. Gertz is Professor of Hebrew Bible/Old Testament at the Ruprecht-Karls-Universität Heidelberg. Bernard M. Levinson is Professor of Jewish Studies and Hebrew Bible and of Law at the University of Minnesota in Minneapolis. Dalit Rom-Shiloni is Associate Professor of Hebrew Bible at the Department of Biblical Studies, Tel Aviv University. Konrad Schmid is Professor of Hebrew Bible/Old Testament at the University of Zurich.

e-ISBN PDF 978-3-16-153884-1 ISBN 978-3-16-153883-4 ISSN 0940-4155 (Forschungen zum Alten Testament) Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliographie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2016  by Mohr Siebeck Tübingen, Germany. www.mohr.de This book may not be reproduced, in whole or in part, in any form (beyond that permitted by copyright law) without the publisher’s written permission. This applies particularly to reproductions, translations, microfilms and storage and processing in electronic s­ ystems. The book was printed by Gulde Druck in Tübingen on non-aging paper and bound by Buchbinderei Spinner in Ottersweier. Printed in Germany.

Table of Contents Jan Christian Gertz, Bernard M. Levinson, Dalit Rom-Shiloni, Konrad Schmid Convergence and Divergence in Pentateuchal Theory – The Genesis and Goals of This Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Part One Empirical Perspectives on the Composition of the Pentateuch Jan Christian Gertz Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Christopher A. Rollston Inscriptional Evidence for the Writing of the Earliest Texts of the Bible – Intellectual Infrastructure in Tenth- and Ninth-Century Israel, Judah, and the Southern Levant . . . . . . . . . . . . . . . . . . . . . . . . 15 David P. Wright The Covenant Code Appendix (Exodus 23:20–33), Neo-Assyrian Sources, and Implications for Pentateuchal Study . . . . . . . . . . . . . . 47 David M. Carr Data to Inform Ongoing Debates about the Formation of the Pentateuch – From Documented Cases of Transmission History to a Survey of Rabbinic Exegesis . . . . . . . . . . . . . . . . . . . . . . . . . 87 Molly M. Zahn Innerbiblical Exegesis – The View from beyond the Bible . . . . . . . . . 107 Armin Lange From Many to One – Some Thoughts on the Hebrew Textual History of the Torah . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

VI

Table of Contents

Part Two Can the Pentateuch Be Read in Its Present Form? Narrative Continuity in the Pentateuch in Comparative Perspective Jeffrey Stackert Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Jean Louis Ska What Do We Mean by Plot and by Narrative Continuity? . . . . . . . . . 201 Yairah Amit Travel Narratives and the Message of Genesis . . . . . . . . . . . . . . . 223 Joel S. Baden Why Is the Pentateuch Unreadable? – Or, Why Are We Doing This Anyway? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Jeffrey Stackert Pentateuchal Coherence and the Science of Reading . . . . . . . . . . . . 253 Jean-Pierre Sonnet Does the Pentateuch Tell of Its Redactional Genesis? – The Characters of YHWH and Moses as Agents of Fortschreibung in the Pentateuchʼs Narrated World . . . . . . . . . . . . . . . . . . . . . . . . 269 Joel S. Baden Continuity between the Gaps – The Pentateuch and the Kirta Epic . . . . . 283

Part Three The Role of Historical Linguistics in the Dating of Biblical Texts Shimon Gesundheit Introduction – The Strengths and Weaknesses of Linguistic Dating . . . . 295 Erhard Blum The Linguistic Dating of Biblical Texts – An Approach with Methodological Limitations . . . . . . . . . . . . . . . . . . . . . . . . . 303 Jan Joosten Diachronic Linguistics and the Date of the Pentateuch . . . . . . . . . . . 327

Table of Contents

VII

William M. Schniedewind Linguistic Dating, Writing Systems, and the Pentateuchal Sources . . . . . 345 Thomas Römer How to Date Pentateuchal Texts – Some Case Studies . . . . . . . . . . . 357 Noam Mizrahi The Numeral 11 and the Linguistic Dating of P . . . . . . . . . . . . . . . 371 Jakob Wöhrle Thereʼs No Master Key! – The Literary Character of the Priestly Stratum and the Formation of the Pentateuch . . . . . . . . . . . . . . . . 391 Frank H. Polak Oral Platform and Language Usage in the Abraham Narrative . . . . . . . 405 Frank H. Polak Storytelling and Redaction – Varieties of Language Usage in the Exodus Narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

Part Four The Significance of Second Temple Literature and the Dead Sea Scrolls for the Formation of the Pentateuch Bernard M. Levinson Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479 Sidnie White Crawford What Constitutes a Scriptural Text? – The History of Scholarship on Qumran Manuscript 4Q158 . . . . . . . . . . . . . . . . . . . . . . . . . 483 Molly M. Zahn Scribal Revision and the Composition of the Pentateuch – Methodological Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 Reinhard G. Kratz Reworked Pentateuch and Pentateuchal Theory . . . . . . . . . . . . . . . 501 Richard J. Bautch Holy Seed – Ezra 9–10 and the Formation of the Pentateuch . . . . . . . . 525 Sara Japhet What May Be Learned from Ezra–Nehemiah about the Composition of the Pentateuch? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543

VIII

Table of Contents

Part Five Evidence for Redactional Activity in the Pentateuch Konrad Schmid Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563 Jean Louis Ska Some Empirical Evidence in Favor of Redaction Criticism. . . . . . . . . 567 Christoph Levin The Pentateuch – A Compilation by Redactors . . . . . . . . . . . . . . . 579 Konrad Schmid Post-Priestly Additions in the Pentateuch – A Survey of Scholarship . . . . 589

Part Six The Integration of Preexisting Literary Material in the Pentateuch and the Impact upon Its Final Shape Joel S. Baden Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607 Rainer Albertz Noncontinuous Literary Sources Taken Up in the Book of Exodus . . . . . 609 Itamar Kislev The Story of the Gadites and the Reubenites (Numbers 32) – A Case Study for an Approach to a Pentateuchal Text . . . . . . . . . . . . . . . 619 Karin Finsterbusch Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative – Different Solutions in MT Deut 31:1–32:47 and (the Hebrew Vorlage of) LXX Deut 31:1–32:47 . . . . . . . . . . . . . . . . . 631 David P. Wright Source Dependence and the Development of the Pentateuch – The Case of Leviticus 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 651

Table of Contents

IX

Part Seven Historical Geography of the Pentateuch and Archaeological Perspectives Jan Christian Gertz Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685 David Ben-Gad HaCohen Biblical Criticism from a Geographerʼs Perspective – “Transjordan” as a Test Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 687 Israel Finkelstein and Thomas Römer Early North Israelite “Memories” of Moab . . . . . . . . . . . . . . . . . 711 Thomas B. Dozeman The Historical Geography of the Pentateuch and Archaeological Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 729 Jan Christian Gertz Hezekiah, Moses, and the Nehushtan – A Case Study for a Correlation between the History of Religion in the Monarchic Period and the History of the Formation of the Hebrew Bible . . . . . . . . . . . . . . . 745 Angela Roskop Erisman For the Border of the Ammonites Was . . . Where? – Historical Geography and Biblical Interpretation in Numbers 21 . . . . . . . . . . . 761

Part Eight Do the Pentateuchal Sources Extend into the Former Prophets? Konrad Schmid Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 779 Baruch J. Schwartz The Pentateuchal Sources and the Former Prophets – A Neo-Documentarianʼs Perspective . . . . . . . . . . . . . . . . . . . . 783 Cynthia Edenburg Do the Pentateuchal Sources Extend into the Former Prophets? – Joshua 1 and the Relation of the Former Prophets to the Pentateuch . . . . 795 Thomas Römer The Problem of the Hexateuch . . . . . . . . . . . . . . . . . . . . . . . 813

X

Table of Contents

Part Nine Rethinking the Relationship between the Law and the Prophets Dalit Rom-Shiloni Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 831 Konrad Schmid The Prophets after the Law or the Law after the Prophets? – Terminological, Biblical, and Historical Perspectives. . . . . . . . . . . . 841 Marvin A. Sweeney Hoseaʼs Reading of Pentateuchal Narratives – A Window for a Foundational E Stratum . . . . . . . . . . . . . . . . . . . . . . . . . . . 851 Reinhard Achenbach The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah . . . . . 873 Georg Fischer ‫ – ותפשי התורה לא ידעוני‬The Relationship of the Book of Jeremiah to the Torah . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 891 Dalit Rom-Shiloni Compositional Harmonization – Priestly and Deuteronomic References in the Book of Jeremiah – An Earlier Stage of a Recognized Interpretive Technique . . . . . . . . . . . . . . . . . . . . . 913 John Kessler Patterns of Descriptive Curse Formulae in the Hebrew Bible, with Special Attention to Leviticus 26 and Amos 4:6–12 . . . . . . . . . . . . 943 Mark J. Boda Reading Zechariah 9–14 with the Law and the Prophets – Sibling Rivalry and Prophetic Crisis . . . . . . . . . . . . . . . . . . . . . . . . 985 Jakob Wöhrle Jacob, Moses, Levi – Pentateuchal Figures in the Book of the Twelve . . . 997 Christophe L. Nihan Ezekiel and the Holiness Legislation – A Plea for Nonlinear Models . . . 1015 Ariel Kopilovitz What Kind of Priestly Writings Did Ezekiel Know? . . . . . . . . . . . 1041

Table of Contents

XI

Michael A. Lyons How Have We Changed? – Older and Newer Arguments about the Relationship between Ezekiel and the Holiness Code . . . . . . . . . . . 1055 Tova Ganzel and Risa Levitt Kohn Ezekielʼs Prophetic Message in Light of Leviticus 26 . . . . . . . . . . 1075

Part Ten Reading for Unity, Reading for Multiplicity – Theological Implications of the Study of the Pentateuchʼs Composition Benjamin D. Sommer Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1087 Benjamin D. Sommer Book or Anthology? – The Pentateuch as Jewish Scripture . . . . . . . . 1091 Markus Witte Methodological Reflections on a Theology of the Pentateuch . . . . . . 1109 Jean-Pierre Sonnet The Dynamic of Closure in the Pentateuch . . . . . . . . . . . . . . . . 1121 James W. Watts Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture . . . 1135 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1147 Ancient Sources Index  Hebrew Bible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1151  Ancient Near Eastern and Epigraphic Texts and Papyri . . . . . . . . . 1190  Deuterocanonical Works . . . . . . . . . . . . . . . . . . . . . . . . 1192  Second Temple Literature . . . . . . . . . . . . . . . . . . . . . . . . 1192  Texts from the Judean Desert . . . . . . . . . . . . . . . . . . . . . . 1192  New Testament . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1194  Rabbinic Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1194  Early Christian Writings and Greco-Roman Literature . . . . . . . . . 1195  Medieval Writers . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1195 Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1197

Convergence and Divergence in Pentateuchal Theory The Genesis and Goals of This Volume Scholarly advance in the humanities often depends less on sensational new discoveries than upon the questioning and re-evaluation of what had become unquestioned assumptions.1

The Pentateuch lies at the heart of Western humanities. With its notions of divine revelation and social transformation through historical action, it serves as a bedrock document for Judaism, Christianity, and Islam. It contributes powerfully to areas ostensibly far removed from religion, such as the rich literary, intellectual, political, and artistic history of European and later North American civilization, and has also influenced Africa, Asia, and South America. Yet despite nearly two centuries of scholarship, the human origins of this monument of civilization remain shrouded in the past. Indeed, recent developments in scholarship have broken down an earlier consensus, making it even more difficult to date its source documents and gain access to the compositional process by which the Pentateuch first took shape. The traditional conception of a unified, self-consistent foundation narrative that begins with creation and extends to the eve of the Israelitesʼ entry into the promised land of Canaan has long been given up. Critical scholarship has isolated multiple layers of tradition, inconsistent laws, and narratives that could only have originated from separate communities within ancient Israel and were joined together at a relatively late stage by a process of splicing and editing. The so-called New Documentary Hypothesis, often associated with the name of Julius Wellhausen, had dominated academic discourse on the Pentateuch since the end of the nineteenth century.2 It presupposes four originally independent literary sources (the Yahwist, Elohist, Priestly, and Deuteronomic sources, identified by the sigla J, E, P, and D), each with its own set of laws and narratives, which were joined together in stages to produce the composite text of the Pentateuch. Despite challenges and modifications, the explanatory power 1  R. J. Coggins et al., preface to Israelʼs Prophetic Tradition: Essays in Honour of Peter R. Ackroyd (ed. R. J. Coggins et al.; Cambridge: Cambridge University Press, 1982), vvi. 2  J. W ellhausen , Prolegomena zur Geschichte Israels (Berlin: de Gruyter, 2001 [repr. from the 6th ed., 1927]); English translation, idem , Prolegomena to the History of Israel (trans. J. S. Black and A. Menzies; Scholars Press Reprints and Translation Series 17; Atlanta: Scholars Press, 1994 [1st ed., Edinburgh: Adam & Charles Black, 1885]).

2

The Genesis and Goals of This Volume

of the model long permitted it to trump rival hypotheses or to incorporate them as minor modifications of detail (such as adjustments of chronology). Recent developments in academic biblical studies, however, jeopardize the revolutionary progress that has been accomplished over the last two centuries. Over the past forty years, the source-critical method has come under unprecedented attack. In many quarters it has been rejected entirely: many scholars claim it no longer provides a secure starting point for investigating the history of Israelite religion or the literary formation of the Pentateuch. Recent decades have witnessed not simply a proliferation of intellectual models but, in many ways much more seriously, the fragmentation of discourse altogether as scholarly communities in the three main research centers of Israel, Europe, and North America increasingly talk past one another. Even when they employ the same terminology (for example, redactor, author, source, exegesis), scholars often mean quite different things. Concepts taken for granted by one group of scholars (such as the existence of the Elohist or the Yahwist sources) are dismissed out of hand by other scholarly communities. That breakdown in a shared discourse is where this volume seeks to make a contribution, by reflecting on methodological assumptions and the theoretical models that inform the discipline. Admittedly, the evidence for or against the Documentary Hypothesis is at best indirect because only copies of copies, in infinite regress, are preserved: the oldest complete manuscript of the Hebrew Bible, the Leningrad Codex, dates to the year 1008 CE, more than a millennium after the events depicted in the Hebrew Bible.3 As a result, all arguments are based on internal content and literary analysis rather than independent, externally datable evidence. In addition, a series of methodological and demographic revolutions in academic biblical studies has drastically changed the playing field. Although numerous factors have played a role in bringing about these changes, the three most important are that (1) archaeologists have made numerous discoveries that challenge any direct correspondence between the textual presentation and a historical reconstruction of the religion and literature of ancient Israel; (2) the discipline of “Old Testament” studies, long dominated by Protestant scholars in Europe and North America working with implicit Christian theological paradigms, has been irrevocably transformed by the emergence of Israeli biblical scholarship and by societal changes that permitted greater numbers of Jews to gain academic positions at American universities;4 and (3) new methodological insights  The Leningrad Codex is catalogued as Firkowitch B19A in the Russian National Library. See D. N. Freedman et al. (eds.), The Leningrad Codex: A Facsimile Edition (Grand Rapids, MI: Eerdmans; Leiden: Brill, 1998). 4  In continental Europe, Jewish scholars could not hold tenured positions in biblical studies, because all such chairs were housed in faculties of Protestant or Catholic theology, which, according to the Konfessionsvorbehalt, restricted both faculty appointments and the awarding of doctoral degrees along confessional lines. Jews with interests in academic religious studies were forced into other fields, such as rabbinics or Assyriology, or into exclusively Jewish 3

Convergence and Divergence in Pentateuchal Theory

3

have directed increased attention to forms of scribal creativity (such as textual reworking and commentary) and to stages of Judean history (such as the exile and the Second Temple period) that were previously marginalized. As a result, traditional paradigms have been rejected as untenable, and new perspectives are constantly being generated. Yet, the lack of a shared intellectual discourse hampers what might otherwise be a moment of opportunity in the creative development of the discipline. In the three major centers of research on the Pentateuch – North America, Israel, and Europe – scholars tend to operate from such different premises, employ such divergent methods, and reach such inconsistent results that meaningful progress has become impossible. The models continue to proliferate but the communication seems only to diminish. In Israeli scholarship, the Documentary Hypothesis in one or another of its classical forms continues to be highly esteemed. Some scholars working at the Hebrew University in Jerusalem in particular see the future of pentateuchal scholarship in the refinement rather than the abandonment of the sources J, E, P, and D for the reconstruction of the compositional history of the Pentateuch. The Priestly texts of the Pentateuch have garnered special interest, along with the Holiness Legislation. They have been examined more profoundly in the Israeli context than elsewhere in biblical studies, and the results are revolutionary. Examining this literature against the background of cultic and legal material recovered from the ancient Near East has led to a new appreciation of the Priestly sourceʼs historical integrity, antiquity, creativity, and cultural significance. A thorough reassessment of the stages of composition of the Priestly literature has yielded an entirely new approach to the formation of this corpus and the interrelationship of its constituent parts. The European discussion has moved in such a different direction that it has become all but unintelligible in the Israeli academic context. Scholars on the European continent predominantly view the Pentateuch as composed from thematic blocks (primeval history, ancestral history, Moses-exodus story) rather than documentary sources. Indeed, the two sources that under the older model provided the most reliable window into the earliest period of Israelite religion – the Yahwist and the Elohist – are now treated with extreme skepticism by most European scholars, who dispute their antiquity if not their very existence. European scholars focus instead on differentiating between Priestly and non-Priestly text complexes. Finally, a number of European scholars contend that there was no connection between Genesis and Exodus in any pre-Priestly texts and shift the date of much of the Pentateuch to the Persian period (539–331 BCE). In North America, as in Israel, scholarship still largely supports the Docuinstitutions such as the Jewish Theological Seminary at Breslau (1854–1939) or the Lehrhaus in Frankfurt founded by Franz Rosenzweig in 1920 and reopened by Martin Buber in 1933. Such issues continue to affect the discipline.

4

The Genesis and Goals of This Volume 

mentary Hypothesis and places a large number of text complexes in the preexilic period. Because of the very different structure of graduate education, North American scholars tend to draw more intensively on ancient Near Eastern and Second Temple literature (like the Dead Sea Scrolls) in attempting to construct their models. They often contend that the current proliferation of European hypotheses is theory driven and self-generated without adequate consideration of comparative literary evidence. In effect, three independent scholarly discourses have emerged. Each centers on the Pentateuch, each operates with its own set of working assumptions, and each is confident of its own claims. This volume seeks to further the international discussion about the Pentateuch in the hope that the academic cultures in Israel, Europe, and North America can move toward a set of shared assumptions and a common discourse. Like the Pentateuch itself, this volume has a long and multilayered compositional history. The point of departure was an international research group entitled Convergence and Divergence in Pentateuchal Theory: Bridging the Academic Cultures of Israel, North America, and Europe, which was convened at the Israel Institute for Advanced Studies (IIAS) in Jerusalem from September 2012 through June 2013. For the first time in the history of the discipline, an internationally representative, long-term research group was convened at an Institute for Advanced Studies in the attempt to overcome the fragmentation in the field of academic biblical studies. The IIAS is remarkable for its commitment to interdisciplinary research and its focus on creating research teams composed of international scholars. The research group was established to investigate the scholarly debate regarding the formation of the Pentateuch and to trace the genealogy of the three diverging academic cultures involved. By bringing together an international team composed of the leading advocates of the competing positions, and by creating a structured series of intellectual encounters, the research group attempted to break free of the intellectual impasse, foster meaningful communication, and permit new knowledge to develop. The idea and initiative for the research group came from Bernard M. Levinson, who in close collaboration with Konrad Schmid (Zurich) and Baruch J. Schwartz (Jerusalem) prepared the formal research proposal. The research group consisted of eight members: Jan Christian Gertz, Shimon Gesundheit, Sara Japhet, Levinson, Schmid, Schwartz, Dalit Rom-Shiloni, and Benjamin D. Sommer. In addition, Joel S. Baden and Jeffrey Stackert contributed as short-term guests for approximately one month each. Ariel Kopilovitz served as research assistant. The first, extensive stage of the groupʼs intellectual work, which took place from September to December 2012, was directed toward the investigation of the emergence of the distinct academic cultures in pentateuchal research. In this phase, each member of the group selected a publication of his or her own that was deemed characteristic of his or her work; each of these was assigned in turn

Convergence and Divergence in Pentateuchal Theory

5

for reading and critique. These presentations contextualized the selections by providing the scholarʼs own view of his or her basic methodological standpoint and assumptions. In a second phase, the group devoted a significant amount of time to the discussion of specific biblical texts, such as the Joseph story in the book of Genesis or the plague cycle in the book of Exodus. These portions of the Pentateuch provide many peculiarities and difficulties for readers, and there are different ways to evaluate these texts in terms of their historical genesis. Some of the group members defended a source-critical approach; others, while not denying that the Pentateuch is composed of sources, placed more emphasis on redactional expansions of preexisting texts. Each member of the group benefited from the rare opportunity to study these texts intensively in the company of colleagues in the field. In the third phase, the group discussed basic differences regarding historical-exegetical methods and also turned more closely to legal texts of the Pentateuch and their early reception. Seminars were given by members of the group who had published on the dating of pentateuchal texts, on the relationship between the legal collections of the Pentateuch, and on the evidence provided by Ezra–Nehemiah for understanding the formation of the Pentateuch. Further academic guests lecturing to the group included Ed Greenstein (Bar-Ilan University), Steven E. Fassberg (Hebrew University), Itamar Kislev (University of Haifa), Armin Lange (University of Vienna), Naphtali Meshel (Princeton University; now Hebrew University), Frank Polak (Tel Aviv University), Alexander Rofé (Hebrew University), and Emanuel Tov (Hebrew University). Invited colleagues and doctoral students from the Hebrew University helped strengthen the groupʼs ties to the local academic community in the field of Hebrew Bible. As a preliminary capstone to its work, the group organized an international conference (bearing the same title as the research group), which took place at the IIAS on May 12–13, 2013. In addition to the members of the group, the list of chairs and speakers included an additional nineteen scholars from Israel and abroad. A second and much larger international conference, with fifty scholars on the program, entitled The Pentateuch within Biblical Literature: Formation and Interaction, took place at the IIAS on May 25–29, 2014. Grant applications by Dalit Rom-Shiloni and by Jan Christian Gertz were essential in funding these conferences, which could not have taken place without the generous support provided by both Israeli and German foundations committed to furthering international research: the IIAS, the Israel Science Foundation, the Fritz Thyssen Foundation, and the Deutsche Forschungsgemeinschaft. Seeking to bring its goals to the attention of colleagues more broadly, the group also organized panels at the World Congress of Jewish Studies in Jerusalem (July 30–Aug. 1, 2013) and the Annual Meeting of the Society of Biblical Literature in Baltimore (Nov. 17–20, 2013).

6

The Genesis and Goals of This Volume

This volume has been organized into ten parts, each representing a theme that the editors thought important in order to move the discipline forward. Each part has been provided with its own introduction that seeks to highlight the larger intellectual goals and rationales of the papers included. Each of the fifty-six essays, contributed by forty-nine international colleagues, has gone through a process of peer review. In the selection and organization of the ten parts, the authors have sought to reframe conventional approaches to the question of the formation of the Pentateuch, bringing to bear historical linguistics, material culture, geography, and the literature of the Second Temple period:

1. Empirical Perspectives on the Composition of the Pentateuch 2. Can the Pentateuch Be Read in Its Present Form? Narrative Continuity in the Pentateuch in Comparative Perspective 3. The Role of Historical Linguistics in the Dating of Biblical Texts 4. The Significance of Second Temple Literature and the Dead Sea Scrolls for the Formation of the Pentateuch 5. Evidence for Redactional Activity in the Pentateuch 6. The Integration of Preexisting Literary Material in the Pentateuch and the Impact upon Its Final Shape 7. Historical Geography of the Pentateuch and Archaeological Perspectives 8. Do the Pentateuchal Sources Extend into the Former Prophets? 9. Rethinking the Relationship between the Law and the Prophets 10. Reading for Unity, Reading for Multiplicity: Theological Implications of the Study of the Pentateuchʼs Composition

Extensive effort has been placed on bringing to bear the relationship of the prophetic corpus to the Pentateuch, with special attention to matters of innerbiblical exegesis and textual allusion as potentially providing new evidence for standard assumptions about textual dating and literary development. The question of the relation between synchronic and diachronic methodology has also been explored. The volume aims, in these ways, less to provide a set of final answers than to open a dialogue that includes proponents of multiple positions, creating a shared conversation and inviting further participation and response.5 5  The editors wish to acknowledge the international grant support that made the original research year, the two conferences, and this volume possible. Gratitude goes first and foremost to the Israel Institute for Advanced Studies (Jerusalem) for its extraordinary support and remarkable staff. The encouragement of its director, Michal Linial, who fondly called us the Tanakhistim, meant a great deal. Major support was also provided by the European Institutes for Advanced Studies Fellowship Program and by the European Commission under the Marie Curie Scheme. Crucial support for funding the two conferences organized by the research group and for the publication of this volume was provided by the Israel Science Foundation, the Fritz Thyssen Foundation (Cologne), and the Deutsche Forschungsgemeinschaft (Bonn). Important support has been provided by the University of Zurich, Heidelberg University, the University of Tel Aviv, and the University of Minnesota. Without the dedicated, professional academic editing services provided by Sarah Shectman and the remarkable production skills of Samuel Arnet (Zurich), the volume would not have seen the light of day.

Convergence and Divergence in Pentateuchal Theory

7

Only the reader can decide whether the research group has achieved its goals. After having devoted himself to the study of the Pentateuch for many years, Julius Wellhausen finally became weary of the field. In 1889, while teaching at Marburg, he received a Ruf (call) to be appointed to the chair in Old Testament at Eberhard Karls University of Tübingen, one of Europeʼs oldest universities. Declining that offer in a letter to the minister of culture of Baden-Württemberg, Wellhausen wrote: Mich interessieren die Themata der Vorlesungen nicht, mich langeweilt der Pentateuch und die kritische Analyse und das Altersverhältnis der Quellen. [I am not interested in the topics of the lectures. I am bored with the Pentateuch, critical analysis, and the relationship between the sources.]6

With contributions that focus closely on the biblical text while asking new questions from a full range of methodological perspectives, we hope to help the reader avoid Wellhausenʼs ennui. Jan Christian Gertz Bernard M. Levinson Dalit Rom-Shiloni Konrad Schmid

6  J. Wellhausen, Briefe (ed. R. Smend et al.; Tübingen: Mohr Siebeck, 2013), 247 (letter of January 12, 1890).

Part One Empirical Perspectives on the Composition of the Pentateuch

Introduction Jan Christian Gertz The reconstruction of the formation of the Pentateuch is still an equation with several unknowns. Perhaps it will always remain an unsolved riddle. In terms of external evidence, the texts and traditions combined to form the given Pentateuch are lost – most likely forever. Nearly unknown are the social and historical circumstances of the formation of the Pentateuch and its parts. And finally, we know hardly anything about the biblical writers who fashioned the narratives and laws of the Pentateuch by combining separate but complementary written traditions. What we think we know about the formation of the Pentateuch is based on internal critical analysis of the Pentateuch. The evidence for or against the Supplementary Hypothesis, the Documentary Hypothesis, and the Fragmentary Hypothesis or the various combinations thereof is, at best, indirect. In the history of research, phenomena such as anachronisms, inconsistencies, contradictions, and thematic and stylistic variations within the books of the Pentateuch were rightly considered incompatible with the traditional view of uniform date or homogeneous authorship. They were more convincingly explained by the supposition of textual growth. It is a plausible hypothesis that the Pentateuch was formed by the combination of separate written traditions, and it is possible for modern scholarship to retrace these processes to some extent. However, beyond this general agreement on the historical growth of biblical literature, there is little consensus in scholarship on the formation of the Pentateuch. Unfortunately, estimations such as “this text is incoherent” or “this thematic or stylistic variation is in no way comparable with the suggestion of a single author” are highly subjective. An argument that some may consider self-evident may be regarded by others as weak. The evaluation of textual evidence is especially controversial concerning the ability to reconstruct the preliminary stages of the text and its literary growth precisely. Further contentious issues are the proof of textual influence and the determination of textual dependence. No less debated, of course, is the general idea of the literary history of ancient Israel, to which the respective hypotheses on the formation of the Pentateuch belong. Reconstructing the formation of the Pentateuch is a historical issue – even if one disregards the historical question and thinks only in terms of literary analysis. As a result, it is necessary to ask for historical analogies and for the

12

Jan Christian Gertz

correlation of historical phenomena:1 are there analogies for the assumption that the Pentateuch as a whole can be divided into four sources? In the light of historically demonstrable literary processes, is it plausible to assume that the received Pentateuch is the work of a single compiler who combined nearly completely preserved sources and abstained from reworking and reformulating them? Is there an empirical basis for scholarly confidence in reconstructing the growth of the text in every detail? Is there empirical evidence for a series of unlimited smaller adaptations? Concrete analogies would enable those doing source criticism or history of redaction to base their work on something more than subjective self-evidence. Naturally, the search for analogies is not new. Three decades ago, Jeffrey Tigay prepared the introduction to his inspiring edited volume, Empirical Models for Biblical Criticism, at the Institute for Advanced Studies in Jerusalem. It is therefore fitting that the present volume similarly emerges out of a research year and two associated conferences convened at the same institute. In his engaging introduction, Tigay mentioned analogies adduced on the basis of Chronicles, the Qumran Scrolls, the Septuagint, ancient Near Eastern literature, and the nature of the cultural milieu in which biblical literature was produced.2 Every single “empirical model” shows that ancient writers created their literary works out of distinct and overlapping written sources or by supplementing one source with another. Concrete analogies could function as models of literary development, providing the critic with firsthand experience with compilersʼ or redactorsʼ techniques. But where can we find appropriate analogies demonstrating more than the general conditions of literary production in the ancient Near East? Given the relative paucity of evidence from the Bible itself or the time of the formation of the early stages of the Pentateuch, the Qumran scrolls and certain postbiblical texts like the book of Jubilees become attractive as models of literary development. Yet, one cannot rule out the possibility that some of the techniques reflected in the postbiblical material are different from those used earlier by the biblical authors. The analogies are thus not perfect. They are not a “primary source” for the techniques of Israelite writers in the earlier periods. Nevertheless, analogies can show what is possible or realistic by presenting what happened elsewhere. In so doing, they can aid in evaluating the historical realism of an existing theory of the formation of the Pentateuch. Like every historical analogy, they are a proof of plausibility. They serve to indicate which compositional technique might

 See E. T roeltsch , “Über historische und dogmatische Methode in der Theologie,” in Zur religiösen Lage, Religionsphilosophie und Ethik, vol. 2 of Gesammelte Schriften (Tübingen: Mohr, 1913), 729–753. 2  J. H. T igay , introduction to Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 1–20. 1

idem ,

Introduction

13

plausibly have been used by the biblical writers by demonstrating what was done in culturally and historically similar contexts. Tigayʼs question has received new attention in recent research, as can be seen in the present volume. Besides the part entitled “Empirical Perspectives on the Composition of the Pentateuch,” Reinhard G. Kratz and Molly M. Zahn address the question in the part “Second Temple Literature and Its Importance for the Formation of the Pentateuch.” Moreover, Jean Louis Ska and Cynthia Edenburg explicitly deal with the topic in the part “Evidence for Redactional Activity in the Pentateuch.”

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible Intellectual Infrastructure in Tenth- and NinthCentury Israel, Judah, and the Southern Levant Christopher A. Rollston Introduction The dating of biblical materials has long been, and shall always be, the subject of much discussion and dispute.1 As part of some of the discussions and disputes regarding the origins of the earliest biblical materials, some very fine scholars have argued that the capacity for writing texts of substance and sophistication was simply not present in Israel or Judah prior to the eighth century BCE. For this reason, these scholars contend that the origins of the earliest biblical materials cannot antedate the eighth century BCE. For example, Thomas L. Thompson has written that “we cannot seek an origin of literature in Palestine prior to the eighth, or perhaps even better the seventh century.”2 Similarly, regarding the Southern Kingdom of Judah, Israel Finkelstein states that he 1  I am grateful to the National Endowment for the Humanities (Washington, DC), the Albright Institute of Archaeological Research (Jerusalem), and the American Center of Oriental Research (Amman) for grants, fellowships, and libraries that permitted me to conduct the research in this article. In addition, I am indebted to the Department of Antiquities of Jordan, the Amman Citadel Museum, the Israel Antiquities Authority, the Israel Museum (Jerusalem), the Rockefeller Museum (Jerusalem), the Directorate General of Antiquities of Lebanon, the National Museum of Beirut, the Harvard Semitic Museum, the British Museum, and the University of Pennsylvania for permission to collate inscriptions in their collections. As always, I am grateful to Bruce Zuckerman and Marilyn Lundberg of West Semitic Research for photographic and digital expertise and good counsel. I am also grateful to George Washington University for providing research funds for assistance with the completion of this article, to my research assistant Nathaniel E. Greene for his assistance with digital matters, and to Adam Bean for reading a penultimate version of this manuscript and making useful suggestions. Finally, I wish to thank my dear friend Bernard M. Levinson for facilitating my invitation to be part of the Jerusalem Symposium on the Pentateuch in May 2013, a presentation that formed the basis for this article. 2  T. L. Thompson, Early History of the Israelite People from the Written and Archaeological Sources (SHCANE 4; Leiden: Brill, 1992), 391.

16

Christopher A. Rollston

has “argued time and again [that] archaeology shows that meaningful scribal activity appeared in Jerusalem only with the rise of Judah to full statehood in the late eighth century B.C.E.” For this reason, Finkelstein concludes that the “composition of literary works” in Judah could not antedate the eighth century BCE.3 Regarding the Northern Kingdom of Israel, he has stated that “one can expect large-scale building activities such as the ones carried out at Samaria, Jezreel, and the other Omride sites and a prosperous economy to be accompanied by an advanced bureaucratic apparatus, including writing,” but he then goes on to state that “evidence of writing in the entire region in the early ninth century is sparse. In fact, not a single early ninth-century B.C.E. inscription has thus far been found in the heartland of Israel – at Samaria, Jezreel, Megiddo, Yokneam, and Taanach.” He states further that “daily administration related to the recording of agricultural output is evident in Israel only in the first half of the eighth century B.C.E., first and foremost in the Samaria ostraca.” He is, of course, aware that someone might contend that “most scribal activity was carried out on papyrus and parchment” (and thus perished during the course of time), but he argues that there is also an absence of monumental inscriptions in Northwest Semitic until “the second half of the ninth century, for example, the Mesha and Tel Dan inscriptions.” He then concludes that “literacy and scribal activity during the time of the Omrides was weak at most.”4 Regarding the Hebrew Bible, therefore, Finkelstein states that “assembling all available data for scribal activity in Israel and Judah reveals no evidence of writing before approximately 800 B.C.E. In fact, it shows that meaningful writing in Israel began in the first half of the eighth century, while in Judah it commenced only in the late eighth and more so in the seventh century B.C.E.” Then he asserts that “past ideas regarding the date of compilation of biblical texts were based on the testimony of the Bible and hence fell prey to circular reasoning. Recent archaeological and biblical research has made it clear that no biblical text could have been written before circa 800 B.C.E. in Israel and about a century later in Judah.” Along those same lines, he contends that “this means that the earliest northern texts, such as the core of the Jacob cycle in Genesis, were probably put in writing in the first half of the eighth century, during the period of prosperity in Judah, especially under the long reign of Jeroboam II. This indicates, in turn, that ninth-century B.C.E. and earlier memories could have been preserved and transmitted only in oral form.”5  I. Finkelstein, “Digging for the Truth: Archaeology and the Bible,” in The Quest for the Historical Israel (ed. B. B. Schmidt; ABS 17; Atlanta: Society of Biblical Literature, 2007), 9–20, here 12. 4  I. Finkelstein, The Forgotten Kingdom: The Archaeology and History of Northern Israel (SBL Ancient Near Eastern Monographs 5; Atlanta: Society of Biblical Literature, 2013), 113–115. 5  F inkelstein , Forgotten Kingdom (see n. 4), 162–163. For similar statements in earlier publications, see also idem , “Digging for the Truth” (see n. 3), 14, 17. It is worth noting 3

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

17

Ultimately, these sorts of construals of the epigraphic evidence are simply too conservative and fail to factor in the weight of the cumulative evidence of the actual extant inscriptions from the southern Levant. That is, the constellation of the extant Levantine Northwest Semitic epigraphic evidence demonstrates nicely that trained scribal elites, using a linear alphabetic script, were already functioning very capably in Phoenicia (e.g., Byblos) in the late eleventh and early tenth centuries BCE, using the standardized Phoenician script with good letter morphology, consistency regarding the direction of writing (sinistrograde), consistency of stance, and care with regard to letter spacing and letter environment.6 Much the same is true as well (as I shall that those arguing for the eighth and seventh centuries as the earliest possible dates for the production of literary and historical texts in Israel and Judah often rely on the work of D. J amieson -D rake, Scribes and Schools in Monarchic Israel: A Socio-Archaeological Approach (JSOTSup 109; Sheffield: Sheffield Academic, 1991). For example, Thompson , Early History of the Israelite People (see n. 2), 390–391; N. P. L emche, Prelude to Israelʼs Past: Background and Beginnings of Israelite History and Identity (Peabody, MA: Hendrickson Publishers, 1998), 221; Alas, Jamieson-Drakeʼs selection, analysis, and understanding of the archaeological and epigraphic material is plagued with problems. Among the most damning of critiques is that of G. I. D avies , “Were There Schools in Ancient Israel?,” in Wisdom in Ancient Israel: Essays in Honour of J. A. Emerton (ed. J. Day et al.; Cambridge: Cambridge University Press), 199–211, here 207–209. On some of the orthographic and paleographic problems with Jamieson-Drake, see C. A. Rollston, “Scribal Education in Ancient Israel: The Old Hebrew Epigraphic Evidence,” BASOR 344 (2006), 47–74. An additional problem with Jamieson-Drake is that he seems to presuppose that scribal activity would perforce occur in some sort of monumental context. But even in Mesopotamia and Egypt, scribal education often occurred in small-scale contexts, including domestic contexts. For this reason, it seems entirely reasonable that in Israel and Judah, scribal education occurred in small-scale contexts, including domestic ones. For the literature on this, see C. A. R ollston, “An Old Hebrew Stone Inscription from the City of David: A Trained Hand and a Remedial Hand on the Same Inscription,” in Puzzling Out the Past: Studies in Northwest Semitic Languages and Literatures in Honor of Bruce Zuckerman (ed. M. J. Lundberg et al.; Leiden: Brill, 2012), 189–196, here 195–196. Many of the criticisms that can be leveled against Jamieson-Drake can be leveled also against H. M. N iemann , Herrschaft, Königtum, und Staat (FAT 6; Tübingen: Mohr Siebeck, 1993), and G. Garbini , History and Ideology in Ancient Israel (New York: Crossroad, 1988). 6  C. A. Rollston, “The Dating of the Early Royal Byblian Inscriptions: A Response to Benjamin Sass,” Maarav 15 (2008), 57–93, with detailed discussion of the host of paleographic problems that accompany Sassʼs attempt to lower the dates of the monumental royal Byblian inscriptions by some century and a half. Regarding the monumental royal Byblian inscriptions, it should be emphasized that the recent article of I. Finkelstein and B. S ass , “The West Semitic Alphabetic Inscriptions, Late Bronze II to Iron IIA: Archaeological Context, Distribution, and Chronology,” HBAI 2 (2013), 149–220, constitutes a repetition of the same arguments Sass has attempted to use in recent years. Striking (and telling) is the fact that within this article, rather than replying to the paleographic arguments I marshaled, Sass and Finkelstein write the following: “Behind [Rollstonʼs] [. . .] steadfast clinging to the 1000 B.C.E. dating of Ahiram and the tying of two Byblos kings to the two Egyptian Pharaohs

18

Christopher A. Rollston

argue below) in the fledgling states of Israel, Judah, Moab, and Ammon during the ninth century. After all, most of the ninth-century inscriptions from these kingdoms and chronological horizons consist of well-formed letters; good letter morphology, ductus, and stance; sufficient attention to spacing and letter environment; and usage of a standard phonology and orthography. In addition, some of these ninth-century inscriptions from the southern Levant contain some rather impressive content, replete not only with reliable historical content but also with some deft political and religious ideology. Someone might contend that such inscriptions hail primarily from the second half of the ninth century, not the first half. I would counter that the sophisticated usage of script and language is not the sort of thing that “springs forth full-grown from the head of Zeus,” and so the presence of some fine inscriptions from the middle and second half of the ninth century presupposes the presence of very capable scribalism during the preceding chronological horizon(s). And, of course, in addition to this, there is a growing corpus of tenth-century inscriptions from the regions of Israel and Judah (discussed below), thus augmenting the totality of the epigraphic record. Therefore, at this juncture, although the epigraphic evidence cannot be said to support some sort of a “Solomonic enlightenment” in Israel and Judah of the tenth century,7 neither can it convincingly be argued that the eighth century is the earliest possible date for a scribal apparatus was apparently the wish not to let go of Albrightʼs proposed archaeological and textual backdrop to the biblical David, Solomon and Hiram” (181; A. Lemaire and A. Millard are similarly criticized with the same words). At least in my case, I said nothing about David, Solomon, or the biblical Hiram, as they were not part of the evidence I brought to the fore. Moreover, I contended that the Aḥiram Sarcophagus inscription dated not to 1000 BCE but rather to sometime in the tenth century – arguably the early tenth century, but still, the point is that I did not “cling steadfastly” to 1000 BCE for Aḥiram. In any case, after this, Sass and Finkelstein simply repeat their arguments for the late dating of all of these inscriptions. Suffice it to say that the paleographic problems with Sassʼs attempt to lower the chronology that I enumerated in detail in 2008 have not changed. That is, the paleographic data support the standard dating and so attempts to lower the dating one hundred to one hundred and fifty years are not tenable, and I am confident that epigraphers will not embrace the proposed lowering. 7  For the older literature on the so-called Solomonic enlightenment, a term that was used to suggest the presence of scribal elites in the royal court and the capacity for some disinterest in the writing of historical and wisdom texts, see G. von R ad , The Problem of the Hexateuch and Other Essays (trans. E. W. Trueman Dicken; New York: McGraw-Hill, 1966 [1st ed., 1938]), 1–78; A. A lt, “Die Weisheit Salomos,”Theologische Literaturzeitung 76 (1951), 139–144; R. B. Y. S cott , “Solomon and the Beginnings of Wisdom in Israel,” in Wisdom in Israel and in the Ancient Near East: Presented to Professor Harold Henry Rowley (ed. N. Noth and D. Winton Thomas; VTSup 3; Leiden: Brill, 1955), 262–279; G. von R ad , Wisdom in Israel (Nashville: Abingdon, 1972). Significantly, the notion of a Solomonic enlightenment was strongly, cogently, and fatally critiqued long ago by J. L. Crenshaw , “A Review of Wisdom in Israel by Gerhard von Rad,” RelSRev 2 (1976), 6–12; and also in idem , Old Testament Wisdom: An Introduction (2nd ed.; Louisville, Westminster John Knox,

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

19

in Israel, Judah, Moab, and Ammon. Rather, the convergence of epigraphic evidence demonstrates that the intellectual scribal infrastructure was already present in these regions for the production of literature in (at least) the ninth century. And for Phoenicia, the epigraphic evidence demonstrates that there were already very capable trained scribes functioning within a state apparatus in the tenth century, and arguably also in the late eleventh century as well.8 And in previous publications, building on the work of the previous generation of scholars, I have argued for the presence of a very sophisticated scribal apparatus in Israel and Judah during the eighth, seventh, and sixth centuries BCE; thus the scribal productivity of the ninth century persists in the succeeding centuries as well.9 Finally, as a complementary component of this discussion, I shall also contend that there are historical materials in the Hebrew Bible that are most naturally understood as having been based on written royal annals of ninth-century officialdom of Israel and Judah. In essence, therefore, I shall contend that the intellectual infrastructure was certainly present in the southern Levant during the ninth century BCE for writing texts of substance and sophistication, and this is as true of scribes in Israel and Judah as it was of scribes in Moab, Ammon, and Damascus.

1998). Suffice it to say that those who continue to argue against the Solomonic enlightenment are essentially (to use a proverbial phrase from the past) “beating a dead horse.” 8  I have discussed the scripts and inscriptions of all of these regions in various technical articles, a number of which appear in the footnotes in this article, but I have also provided fairly synthetic summaries of the epigraphic and paleographic data in C. A. R ollston, Writing and Literacy in the World of Ancient Israel: Epigraphic Evidence from the Iron Age (ABS 11; Atlanta: Society of Biblical Literature, 2010), 19–82. The finest and most detailed tome regarding the script traditions of the Levant during the ninth century is arguably that of my former student, H. D. Davis Parker, “The Levant Comes of Age: The Ninth Century BCE through Script Traditions” (Ph.D. diss., The Johns Hopkins University, 2013). 9  R ollston, “Scribal Education in Ancient Israel” (see n. 5), 47–74; idem , “The Phoenician Script of the Tel Zayit Abecedary and Putative Evidence for Israelite Literacy,” in Literate Culture and Tenth-Century Canaan: The Tel Zayit Abecedary in Context (ed. R. E. Tappy and P. K. McCarter; Winona Lake, IN: Eisenbrauns, 2008), 61–96; idem , “Scribal Curriculum during the First Temple Period: Epigraphic Hebrew and Biblical Evidence,” in Contextualizing Israelʼs Sacred Writing: Ancient Literacy, Orality, and Literary Production (ed. B. B. Schmidt; AIL; Atlanta: Society of Biblical Literature, 2015), 69–99; idem , “Epigraphy: Writing Culture in the Iron Age Levant,” in The Wiley Blackwell Companion to Ancient Israel (ed. S. Niditch; Hoboken, NJ: Wiley & Sons, 2016), 131–150. See the literature cited in the notes of these articles for reference to earlier secondary literature, especially J. L. C renshaw , Education in Ancient Israel: Across the Deadening Silence (New York: Doubleday, 1998). Among the most important recent works, see in particular D. M. Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (Oxford: Oxford University Press, 2005); K. van der Toorn , Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007).

20

Christopher A. Rollston

I. Select Inscriptions of the Tenth, Ninth, and Early Eighth Centuries from the Southern Levant: Monumental Among the oldest of the monumental inscriptions are the royal inscriptions of Byblos, a famous Phoenician port city located on the Mediterranean Sea. And among the most impressive of the Byblian royal inscriptions is the King Aḥiram Sarcophagus Inscription.10 This inscription dates to the tenth century BCE and was chiseled into a very finely made sarcophagus. It is normally classified as among the oldest of the Byblian Phoenician inscriptions, with only Azarbaʿl (figure 1) being slightly older. The Aḥiram inscription reads as follows: “The sarcophagus that [ʾE]thbaʿl son of Aḥiram king of Byblos made for Aḥiram, his father, when committing him to eternity.” Thus, this is an inscription that a son made for his deceased father. Aḥiram, the father, was the king of Byblos, and it is reasonable to contend that ʾEthbaʿl his son was also his successor. The inscription continues, and it contains some of the traditional curse materials that are part of many royal burial inscriptions, namely, “And if a king among kings or a governor among govnors, or an army commander should come up to Byblos and expose this sarcophagus, may the scepter of his rule be overturned, and may the throne of his kingdom be in tumult, and may rest flee from Byblos. And as for him, may his royal record(s) be erased from By[blos].”11 This inscription is important for a number of additional reasons as well. First of all, although it is not the earliest of the Old Byblian royal inscriptions (again, Azarbaʿl is older), the script mandates a date squarely in the tenth century BCE, so it is quite early indeed. Furthermore, the script is beautifully executed, and it reflects the fact that there were some very gifted Phoenician scribes and stonemasons during this early period of the Phoenician script. The letter morphology reflects some variation but always within a strict standard of deviation. The spacing between letters is done with precision and consistency. The stonemason did begin to reduce the size of the letters as he neared the end of the sarcophagus, but even with these sorts of constraints the letter morphology and stance are quite neat and clean (with the exception of the rewriting in the chipped-out area at the end of the inscription). This inscription is also interesting (but not unique) in that a letter was accidentally omitted from a word, namely, in the word “governor.” No one, not even a gifted scribe, is perfect. The entire inscription is the work of a trained professional. It is among the most famous of the Old Byblian Phoenician inscriptions, and it is a sterling demonstration of the sophistication of the Phoenician scribal apparatus in the Iron Age Levant of the tenth century.

10  R. Dussaud, “Les inscriptions phéniciennes du tombeau dʼAhiram, roi de Byblos,” Syria 5 (1924), 135–157. 11  Translation from C. A. R ollston , Northwest Semitic Royal Inscriptions (SBLWAW; Atlanta: Society of Biblical Literature, forthcoming).

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

21

The King Yeḥimilk of Byblos Inscription (figure 2) is also representative of the gifted royal scribes and stonemasons in Byblos of Phoenicia. This inscription also dates to the tenth century BCE and is a temple dedication. In it, King Yeḥimilk says: “May Baʿl-Shamēm and Baʿlat of Byblos and the Council of the Holy Gods of Byblos lengthen the days of Yeḥimilk and (lengthen) his years over Byblos.”12 This inscription is particularly nice because it refers to two heads of

Figure 1. Azarbaʿl Inscription (Phoenician, drawing by Christopher A. Rollston).

Figure 2. Yeḥimilk Inscription (Phoenician, drawing by Christopher A. Rollston).

the pantheon, a god and a goddess, and then refers to the divine council as well. The script of this inscription is beautifully done, with spacing between letters consistent and methodical. Moreover, word dividers are used. This inscription is carefully lined as well (double lines separate each line of text), a further demonstration of the care and capability of the scribal professional(s) who were 12

 Translation from Rollston, Northwest Semitic Royal Inscriptions (see n. 11).

22

Christopher A. Rollston

responsible for it. Similarly, the King Abibaʿl of Byblos Inscription, on a statue of the Egyptian pharaoh Sheshonq I (r. ca. 945–924 BCE) is Phoenician and dates to the tenth century. It reads as follows: “[The statue which] Abibaʿl King of [Byblos, son of Yeḥimilk king of] Byblos, brought from Egypt for Baʿlat [of Byblos, his sovereign. May Baʿlat of Byblos lengthen the days of Abibaʿl and (lengthen) his years] over Byblos.”13 Likewise, the Byblian inscription of King Elibaʿl was inscribed on a bust of the Egyptian pharaoh Osorkon I (r. ca. 924–889 BCE) and is also carefully executed. Of some significance, of course, is the fact that two of these great royal Byblian inscriptions were made on imports from Egypt, and not just any imports, but rather royal statuary from Egypt. Moreover, the tradition of professional scribalism at Byblos continued during the late tenth and early ninth century as well with the King Shipiṭbaʿl Inscription (figure 3), a fine royal inscription from that chronological horizon. In sum, there can be no doubt that some very gifted scribes served the royals of the Phoenician city-states, with Byblos a prime demonstration of the presence of such capabilities. The Tel Dan Stela (figure 4) is written in the Old Aramaic language and in the Phoenician script, that is, the prestige script used to write both Phoenician and Aramaic at that time. Although it is not a particularly long inscription, and although it is not entirely preserved, it is impressive. Based on the historical content and the paleography, this inscription can be dated confidently to the ninth century BCE. Tel Dan was a northern Israelite site in this period. The person who commissioned this stela was an Aramean king. Avraham Biran and Joseph Naveh have argued convincingly that it was King Hazael of Damascus (r. ca. 842–806 BCE) who commissioned this inscription.14 George Athas has contended that it was commissioned not by Hazael of Damascus but by Ben-Hadad III of Damascus, the son of Hazael. Most scholars concur with the proposal of Biran and Naveh, as do I.15 This Old Aramaic inscription, commissioned by an Aramean king, was placed on Israelite soil as a reminder to the Israelites of a major defeat. The preserved portion of the stela begins with reference to “cutting” (perhaps about the making of a treaty or about the breach or a treaty) and then mentions that “my father went up against him in war [. . .] and my father lay down and he went to his fathers.” That is, the stela contains reference (but without preserved details, per se) to the history that antedates Hazaelʼs own kingship and conquest. Among the things mentioned is the death of his father (“he lay down and went to his fathers” is a standard idiom for this). The inscription then moves into the  Translation from Rollston, Northwest Semitic Royal Inscriptions (see n. 11).  A. B iran and J. Naveh , “The Tel Dan Inscription: A New Fragment,” IEJ 45 (1995), 1–18. 15  G. Athas, The Tel Dan Inscription: A Reappraisal and a New Interpretation (London: T&T Clark, 2003). 13

14

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

23

casus belli, the reason for Hazaelʼs war against Israel and Judah, namely, “the king of Israel had gone formerly into the land of my father” (i.e., territorial encroachment by Israel). The term formerly is used here to indicate that this happened prior to Hazaelʼs coronation. The inscription then moves into a discussion of King Hazaelʼs rise to kingship: “As for me, Hadad made me king and Hadad went before me (in battle).” This Old Aramaic inscription also recounts some of Hazaelʼs victories, but because this portion of the inscription is not well

Figure 3. Shipiṭbaʿl Inscription (Phoenician, drawing by Christopher A. Rollston).

Figure 4. Tel Dan Stela Inscription (Old Aramaic language, in Phoenician script; drawing by Christopher A. Rollston).

preserved the precise contents cannot be discerned. Some details, nevertheless, are preserved and are particularly important. Namely, Hazael states that he killed King Jehoram (r. ca. 849–842 BCE), son of Ahab, the king of Israel, and that he killed King Ahaziah (r. ca. 842 BCE), son of Jehoram, king of the house of David (i.e., Judah). Only a few words of the remainder of the stela are preserved, but it seems reasonable to argue that King Hazael commissioned this inscription

24

Christopher A. Rollston

in the wake of his rise to power (and in the wake of King Jehu of Israelʼs rise to power, as will be discussed below). In any case, the Tel Dan Stela is a very nice, paradigmatic victory stela, in which a king claims victory over an enemy, on the enemyʼs territory, and then leaves a stela to remind the conquered of their military losses. The script of this inscription is very carefully executed, with standard morphology and ductus and with spacing between letters done with deftness and consistency. The use of word dividers in this inscription is quite consistent, at least as consistent as the great Tell Fakhariyeh Inscription from northern Syria. The Old Aramaic of the Tel Dan Stela is very good, equal to some of the best of the Old Aramaic inscriptions from the eighth century BCE. Of course, there are also some precise historical details in this inscription, replete with personal names and patronymics. There can be no doubt that there were some very finely trained military scribes accompanying King Hazael of Damascus, a conqueror of some significance in the southern Levant of the ninth century BCE. Phoenicia was no backwater in the tenth century BCE and Damascus was no backwater in the ninth century BCE. Scribes with formal standardized training in script and language were there. The Mesha Stela is among the longest and most impressive of all Iron Age Northwest Semitic inscriptions. It is written in the Moabite language, a language that is distinct from Phoenician, Hebrew, Aramaic, and Ammonite. The script of this inscription is the Old Hebrew script, arguably a reflex of northern Israelite (Omride) political and cultural hegemony during much of the first half of the ninth century BCE.16 Based on the historical content and the paleography, this inscription can confidently be assigned to the ninth century BCE, probably during the early years of the second half of the century. The inscription is fairly long, with some thirty-five lines preserved, with around twenty-five to thirty letters per line. The script of this inscription is beautifully executed, with letter spacing consistent, letter morphology consistent (i.e., falling within a certain standard of deviation), and the use of word dividers. The lines of this inscription are straight, a reflection of the careful layout that preceded the actual chiseling of the stone. The orthography used within this inscription is consistent, as is also the phonology. Within this inscription, King Mesha of Moab provides his patronymic (“son of Kemoshyat”) and refers to himself as a “Dibonite.” Then he states that Kemosh (the national God of Moab) had been “angry” with Moab and had allowed Moab to be subjected to the hegemony of King Omri of Israel. During the reign of Omriʼs “son,” Mesha (with the full support of Kemosh) rebelled against his Omride suzerain and was able to regain much of the historic Moabite territory. It is important to emphasize in this context that Meshaʼs public works are also a strong component of this inscription. For  So, J. Naveh, Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography (2nd ed.; Jerusalem: Magnes, 1987), 65; Rollston, Writing and Literacy (see n. 8), 42, 54. 16

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

25

example, Mesha states, “I built Baʿl-Maon and I made a water reservoir in it, and I built Qiryatēn.” Later, Mesha states that he went up against Yaḥaṣ and seized it to augment the territory of Dibon, and he built “Qarḥoh, the walls of the lower acropolis and the walls of the upper acropolis.” He goes on to state, “I myself built its gates and I myself built its towers and I myself built its royal palaces and I myself made the retaining walls of the water reservoir for the

Figure 5. El-Kerak Inscription (Moabite language, in Old Hebrew script; drawing by Christopher A. Rollston).

water spring within the city [. . .] and I built ʿAroʿer and the surfaced road at the ʿArnon.”17 In the ancient Near Eastern world, including the southern Levant, kings were supposed to excel in public works, and Mesha contends that he did just that.18 The main point is that the Mesha Stela Inscription is the product of a formal, sophisticated scribal apparatus in Moab during the early years of the second half of the ninth century BCE. Moreover, the script that it uses is the Old Hebrew script, a reflection of Omride hegemony and the sophisticated nature of the Omride scribal apparatus. The el-Kerak Inscription (figure 5) is also a ninth-century Moabite inscription. It is fragmentary but certainly royal, as the first preserved line indicates: “[. . . K]emoshyat, King of Moab.” Two additional lines of text follow but without enough preserved letters to yield all that much in terms of readings.  Translations of the Mesha Stela are from Rollston, Northwest Semitic Royal Inscriptions (see n. 11). 18  For a very good, detailed discussion of statements about the accomplishment of royal works for oneʼs own kingdom, see D. J. G reen , I Undertook Great Works: The Ideology of Domestic Achievements in West Semitic Royal Inscriptions (FAT 2/41; Tübingen: Mohr Siebeck, 2010). 17

26

Christopher A. Rollston

The script of this inscription is the same script as that of the Mesha Stela, with all of the letters carefully chiseled into the stone, straight lines of text, general consistency of the morphology and stance of the letters, consistent letter spacing, and the presence of word dividers. Of particular importance for reading this text is the recent work of Heather Dana Davis Parker and Ashley Fiutko Arico, who argue convincingly that this ninth-century Moabite inscription was actually inscribed on a statue, with the pleats of a skirt, a belt, and a navel all still present (with the latter two now added into the most recent version of my drawing of this inscription). Furthermore, they also argue that the closest stylistic parallels for this statue are Egyptian royal statuary.19 This evidence demonstrates again that the southern Levant was not some sort of backwater during the ninth century BCE but rather a region with a sophisticated scribal apparatus and the capability of producing statuary similar to contemporary Egyptian statuary or the capability of importing royal statuary from Egypt. Reference should also be made to the Amman Citadel Stela (figure 6), a fine monumental inscription found on the citadel of Amman, ancient Rabbat Amman, the capital of the Ammonite kingdom. It is normally dated to the ninth century BCE on the basis of its script (a lapidary series of the ninth century BCE). Word dividers are used within this inscription. Although this inscription is not complete and there is some damage to the surface, sufficient text (ca. eight lines) is preserved for it to be affirmed with some confidence that this inscription is a monumental dedicatory inscription. For example, at the beginning of the first preserved line, the inscription arguably refers to the Ammonite national deity Milkom, a deity also mentioned in the Bible in connection with Ammon (1 Kgs 11:5; Jer 49:1–3). Moreover, the script of this inscription is very good, with a consistent morphology for the letters, substantial attention to spacing between letters, and frequent usage of word dividers.20 The original quality of this monumental inscription from Rabbat Ammon demonstrates that during the ninth century BCE, the Ammonite scribal apparatus was very capable of producing a first-class royal inscription. Again, the ninth century in the southern Levant was home to some very talented royal scribes from multiple countries. Finally, although only a small fragment of this stela has been preserved, it is important to note that the northern Israelite capital of Samaria has also yielded the remains of a monumental inscription in Old Hebrew. Just four letters of it are preserved, but the limestone is of high quality and the stone has been nicely dressed. The letters were chiseled with precision, the work of a master. This

19  H. D. Davis Parker and A. F. Arico, “A Moabite-Inscribed Statue Fragment from Kerak: Egyptian Parallels,” BASOR 373 (2015), 105–120. 20  On this inscription, see F. M. Cross, “Epigraphic Notes on the Amman Citadel Inscription,” BASOR 193 (1969), 13–19.

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

27

Figure 6. Amman Citadel Stela (Ammonite language, in Phoenician script; photo courtesy of West Semitic Research and the Department of Antiquities of Jordan).

inscription has been dated to the (early) eighth century BCE, but a date in the late ninth century is also tenable.21 Ultimately, the most rational conclusion on the basis of the cumulative linear Northwest Semitic epigraphic data is that trained royal scribes were functioning in important capacities during the eleventh, tenth, ninth, and early eighth centuries BCE in much of the southern Levant (and later, of course).22 These scribes were producing monumental inscriptions chiseled carefully into stone. The inscriptions are characterized by the deft writing of the letters (in terms of morphology and stance), careful attention to spacing between letters, frequent use of word dividers, and general straightness of lines (on a horizontal plane). Moreover, the content of these inscriptions routinely refers to kings and kingdoms, often also using personal names and patronymics. These are royal inscriptions of a very high order, not the work of some fledgling beginners 21  J. W. Crowfoot et al., The Objects from Samaria (London: Palestine Exploration Fund, 1957), plate 4; Rollston, Writing and Literacy (see n. 8), 55. 22  For some of the discussion of script traditions and scribalism during various periods, including those that are later than those discussed in this article, see the following and the literature cited there: C. A. R ollston , “Northwest Semitic Cursive Scripts of Iron II,” in An Eye for Form: Epigraphic Essays in Honor of Frank Moore Cross (ed. J. A. Hackett and W. E. Aufrecht; Winona Lake, IN: Eisenbrauns, 2014), 202–234; idem , “The Iron Age Phoenician Script,” in An Eye for Form: Epigraphic Essays in Honor of Frank Moore Cross (ed. J. A. Hackett and W. E. Aufrecht; Winona Lake, IN: Eisenbrauns, 2014), 72–99; and idem, “Scribal Education in Ancient Israel” (see n. 5), 47–74.

28

Christopher A. Rollston

but rather the work of consummate professionals who were trained well in the technology of writing.

II. Select Inscriptions of the Tenth, Ninth, and Early Eighth Centuries: Nonmonumental From the site of Khirbet Qeiyafa hail some very important inscriptions that date to the late eleventh or early tenth centuries BCE. Among the most interesting is the Ishbaʿl Jar Inscription (figure 7; discovered during the 2012 excavation season). This inscription is dated by the excavator to ca. 1020–980 BCE, based on radiometric dating of the relevant layer in which it was found. According to the editio princeps, the readings of the Incised Ishbaʿl inscription are: [ ] ʾšbʿl bdʿ. I believe that these readings are correct. The authors of the editio princeps render the extant portion of this inscription (the very beginning of the inscription is not preserved, though some traces are present) as “Ishbaʿl son of Bedaʿ.” The authors of the editio princeps correctly note that this inscription was incised before firing and that it is written sinistrograde (i.e., right to left). Because of the stance and direction of the letters, it is readily apparent that this inscription is written in the script that is classified as Early Alphabetic, though a late stage of it.23 There is no doubt that the script of this inscription is that of a trained scribal professional. The letters were executed with precision and deftness. The spacing between words was careful and precise. The word dividers were nicely done and consistent. This inscription constitutes further evidence for the presence of trained scribal professionals in the southern Levant during the late eleventh and early tenth centuries BCE. Of course, the famed Qeiyafa Ostracon was discovered prior to this inscription, in 2008.24 The script of the Qeiyafa Ostracon is also Early Alphabetic (though without the degree of deftness that is present in the Ishbaʿl Inscription), in writing best understood as dextrograde. Dividing lines are used between each line of this ostracon, and some word dividers are present. As is often the case in Early Alphabetic inscriptions, there is substantial variation in the stance and direction of the letters of the Qeiyafa Ostracon. Since this inscription is faded and abraded, the readings have been much discussed and debated. Nevertheless, this ostracon does constitute evidence for scribal activity in the late eleventh and early tenth centuries in the southern Levant, something that is not surprising in light of the impressive monumental remains at the site  Y. Garfinkel et al., “The ʾIšbaʿal Inscription from Khirbet Qeiyafa,” BASOR 373 (2015), 217–233. For further discussion of the script, see my forthcoming article in the Festschrift for Dennis Pardee. 24  H. Misgav et al., “The Ostracon,” in Qeiyafa, Volume 1: Excavation Report, 2007–2008 (ed. Y. Garfinkel and S. Ganor; Jerusalem: Israel Exploration Society/Institute of Archaeology, Hebrew University of Jerusalem, 2009), 243–257. 23

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

29

of Khirbet Qeiyafa.25 It is worth mentioning that recent years have witnessed the discovery of multiple inscriptions in Early Alphabetic script in addition to those of Khirbet Qeiyafa. For example, at Tell es-Safi, an incised Early Alphabetic sherd was found in an archaeological context that the excavators consider late

Figure 7. Ishbaʿl Jar Inscription from Qeiyafa (Canaanite linguistic dialect, in Early Alphabetic script; drawing by Christopher A. Rollston).

Iron Age I (ca. 1050–1000 BCE) or early Iron Age II (1000–900 BCE).26 Of these dates, I prefer one in the late eleventh or early tenth century.27 Also of import are the brief but significant Early Alphabetic inscriptions from Tel BethShemesh28 and the Ophel in Jerusalem.29 In any case, the inscriptions from such sites are not monumental in nature, but some (such as the Ishbaʿl Inscription 25  Finally, it should be emphasized that although it is tempting to argue for a precise linguistic classification of these inscriptions from Qeiyafa (as was done in the editio princeps of the ostracon discovered in 2008), the fact of the matter is that there are no diagnostic elements that constitute evidence for such a classification, neither for the ostracon discovered in 2008 (C. A. R ollston , “The Khirbet Qeiyafa Ostracon: Methodological Musings and Caveats,” Tel Aviv [2011], 67–82) nor for this incised jar inscription discovered in 2012. Since the script of both can be classified as Early Alphabetic, the safest conclusion is that the language is Canaanite. 26  A. M. Maeir et al., “A Late Iron Age I/Early Iron Age II Old Canaanite Inscription from Tell eṣ-Ṣafi/Gat, Israel: Palaeography, Dating, and Historical-Cultural Significance,” BASOR 351 (2008), 39–71. 27  See now also some additional inscriptions from Safi, published in A. M. M aeir and E. Eshel, “Four Short Alphabetic Inscriptions from Late Iron Age IIa Tell eṣ-Ṣafi/Gath and Their Implications for the Development of Literacy in Iron Age Philistia and Environs,” in See, I Will Bring a Scroll Recounting What Befell Me (Ps 40:8): Epigraphy and Daily Life from the Bible to the Talmud Dedicated to the Memory of Professor Hanan Eshel (ed. E. Eshel and Y. Levin; Göttingen: Vandenhoeck & Ruprecht, 2014), 69–88. 28  P. K. McCarter et al., “An Archaic Baal Inscription from Tel Beth-Shemesh,” Tel Aviv 38 (2011), 179–193. 29  E. Mazar et al., “An Inscribed Pithos from the Ophel, Jerusalem,” IEJ 63 (2013), 39–49.

30

Christopher A. Rollston

from Qeiyafa) are quite impressive in terms of script quality. The main point is that during the late eleventh century and the tenth century, writing was part and parcel of scribal life in important polities such as Qeiyafa, Safi, Jerusalem, and Beth-Shemesh. Also worth emphasizing is the fact that although the Phoenician

Figure 8. Kefar Veradim Inscribed Bowl (written in the Phoenician script; drawing by Christopher A. Rollston).

script had already displaced the Early Alphabetic script in Phoenicia during the eleventh century, the usage of the Early Alphabetic script persisted longer in regions of the southern Levant that were not as central as Phoenicia. In many respects, this is predictable, but in any case, at this juncture the evidence is sufficient to mandate that conclusion based on the epigraphic data. The Kefar Veradim Inscribed Bowl (figure 8) is a stunning artifact, made of bronze and fluted.30 It is obviously a prestige object. This inscription, excavated at a site in the north, consists of just four words, all preserved quite well, with two word dividers present. It reads as follows: “The Bowl of Pesaḥ ben Shemaʿ.” This inscribed bowl was found in a burial cave at Kefar Veradim (Israel). Moreover, the script is definitively Phoenician. The script of this inscription reflects the work of a trained, consummate scribal hand. I consider this inscribed bowl to hail from the same basic chronological horizon as the Azarbaʿl Inscription from Byblos, that is, the late eleventh century or the very early tenth century BCE. I believe that it is quite probable that this was an heirloom piece, rather than something that was made at the time of the deceasedʼs death. It is also important to mention in this connection that the Phoenician script was the prestige script during the tenth century BCE, and as such its usage is attested in multiple 30  Y. Alexandre, “A Canaanite Early Phoenician Inscribed Bronze Bowl in an Iron Age IIA-B Burial Cave at Kefar Veradim, Northern Israel,” Maarav 13 (2006), 7–41.

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

31

countries (as seen also above, with various inscriptions), including early Israel. However, during the early ninth century, the Old Hebrew national script arose and rapidly displaced the Phoenician script in the land of Israel.31 Based on the evidence at hand (some of it discussed below), it seems most natural to connect the rise of the Old Hebrew script with the Omrides of the Northern Kingdom of Israel. In any case, this piece is another nice example of the varied usages of writing during the Iron Age. During the 2005 excavation season at Tel Zayit, a large stone was found with an incised, two-line inscription.32 Although abraded in a few places, the inscription is an abecedary, that is, the letters of the alphabet in order. The excavator has dated the archaeological context to the mid-tenth century BCE. The morphology and stance of the letters is impressive, though the writing surface is certainly not optimal. There has been substantial discussion about the script, with the editio princeps arguing that it anticipates the features of the Old Hebrew script. Based on the general absence of features long considered markers of the early Old Hebrew script (such as the curvature of the terminal portions of letters such as kaph, mem, and nun, among other things), I have argued that the script is just the standard Phoenician script, that the inscription dates to the late tenth or very early ninth century BCE, and that it is roughly contemporary with the Gezer Calendar (which is also written in the Phoenician script).33 Among the most important things to emphasize here is the fact that this inscription is evidence of the scribal apparatus in the sphere of Judah in the tenth or early ninth century BCE. Note that Tel Zayit is located just a short distance from Lachish (a Judean site). Someone might suggest that this stone was brought to Tel Zayit from someplace else, but this is a fairly crude stone with an abecedary and it seems acceptable to posit that it was inscribed at Tel Zayit. Thus, from my perspective, the most convincing conclusion is to contend that the Tel Zayit Abecedary constitutes evidence for a scribal apparatus at an important southern site, during the late tenth century BCE or the very early ninth century BCE. At the northern Israelite site of Tel Reḥov, almost a dozen inscriptions have been found, brief and fragmentary, but critically important. Inscriptions 1–4 were found in Stratum VI, inscription 5 in Stratum V, and inscriptions 6–11 in Stratum IV. According to the excavator, Stratum VI is to be dated to the mid-tenth century BCE, Stratum V to the last decades of the tenth century, and Stratum IV to the ninth century BCE, with the terminus for Stratum IV being no

 I discuss the rise of the Old Hebrew script briefly in this article (below) and in some detail, with reference to much of the primary and secondary literature, in Rollston, “Phoenician Script of the Tel Zayit Abecedary” (see n. 9), 61–96. 32  R. E. Tappy et al., “An Abecedary from the Mid-Tenth Century B.C.E. from the Judaean Shephelah,” BASOR 344 (2006), 5–46. 33  Rollston, “Phoenician Script of the Tel Zayit Abecedary” (see n. 9), 61–96. 31

32

Christopher A. Rollston

later than 834 BCE.34 Significantly, the script of the inscriptions from Stratum VI has hallmark features of the Phoenician script and none of the hallmarks of the Old Hebrew script. Similarly, the inscription from Stratum V has hallmark

Figure 9. Reḥov Inscription 5, Stratum V (written in the Phoenician script; photo and drawing by Christopher A. Rollston).

features of Phoenician and none of Hebrew. Notice in particular that the terminal portion of the fifth stroke of the mem and the third stroke of the nun do not have the curvature that is a marker of the Old Hebrew script. Rather, they are straight lines, as is customarily the case in the Phoenican script of this period (figure 9). The same is the case for Reḥov Inscription 2 (from Stratum VI), which also has 34  See S. Aḥituv and A. Mazar, “The Inscriptions from Tel Reḥov and Their Contribution to the Study of Script and Writing during Iron Age IIA,” in See, I Will Bring a Scroll Recounting What Befell Me (Ps 40:8): Epigraphy and Daily Life from the Bible to the Talmud Dedicated to the Memory of Professor Hanan Eshel (ed. E. Eshel and Y. Levin; Göttingen: Vandenhoeck & Ruprecht, 2014), 39–68, and figures 1–12. For an earlier publication of some of these inscriptions, see A. Mazar, “Three 10th–9th Century B.C.E. Inscriptions from Tel Reḥov,” in Saxa Loquentur: Studien zur Archäologie Palästinas/Israels: Festschrift für Volkmar Fritz zum 65. Geburtstag (ed. C. G. D. Hertog et al.; Münster: Ugarit-Verlag, 2003), 171–184. Of course, there have been attempts to lower Mazarʼs dating of these strata and thus also of the inscriptions found there. For this, see F inkelstein and Sass, “West Semitic Alphabetic Inscriptions” (see n. 6). Suffice it to say that Mazarʼs archaeological evidence for tenth- and ninth-century dating are strong and the paleographic evidence dovetails perfectly with the archaeological evidence. Indeed, in my case, I first looked at the published photos and then personally collated all of these inscriptions in Jerusalem with my microscope and assigned paleographic dates to them; after this, I carefully read Mazarʼs dating of the strata and found that my paleographic dating and Mazarʼs archaeological dating both lined up perfectly. I am confident that the dating of inscriptions 1–5 to the tenth century and 6–11 to the ninth century will stand the test of time. I am grateful to Amihai Mazar for allowing me to photograph and collate these inscriptions. Both Heather Dana Davis Parker and I will soon be publishing separate articles on the Reḥov inscriptions in a forthcoming issue of Maarav, dealing in more depth with the paleography.

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

33

a fully preserved mem. This makes sense in light of the fact that evidence for a distinctive Old Hebrew script is first attested in the ninth century BCE. Of great significance, the best preserved inscriptions from Stratum IV do

Figure 10. Reḥov Inscription 8, Stratum IV (written in the Old Hebrew script, photo and drawing by Christopher A. Rollston).

have the hallmark features of the Old Hebrew script, with curvature of the terminal portion of the fifth stroke nicely present in Reḥov Inscription 8 (figure 10). Similarly, there is a fully preserved mem and a fully preserved nun in Reḥov Inscription 6 (also from Stratum IV), and both have curvature in the terminal portion of the final strokes of these two letters.35 That is, the inscriptions from Stratum IV are written in the fledgling, but distinctive, Old Hebrew script. Therefore, this small group of inscriptions from Tel Reḥov Stratum IV constitutes one of the finest, and earliest, groups of Old Hebrew inscriptions, something that Heather Dana Davis Parker has cogently argued.36 Furthermore, as has been noted in the past, some of the Arad Ostraca (from a Judean site) are also from the ninth century BCE and written in the Old Hebrew script.37  As I have mentioned in the past (e.g., Rollston, “Northwest Semitic Cursive Scripts of Iron II” [see n. 22], 220), there are rare occasions, in an inscription written in the Phoenician or Aramaic script, in which there is some curvature present with these letters, but it constitutes the exception, not the norm. 36  Davis Parker, “Levant Comes of Age” (see n. 8), 258–263, 278–304. 37  I discuss the archaeological context of Arad 76 in some detail in R ollston , “Scribal Education in Ancient Israel” (see n. 5), 52, n. 18, along with references to inscriptions from Hazor and Tel Batash. In this connection, see also C. A. R ollston , “What Is the Oldest Hebrew Inscription?,” BAR 38 (May/June 2012), 32–40, 66, 68. Particularly surprising is William Schniedewindʼs recent statement (in a paragraph critiquing my statements that the Old Hebrew script is first attested in the ninth century BCE) that “by the late eighth century, 35

34

Christopher A. Rollston

Although fragmentary and faded, Arad Ostracon 76 constitutes a prime example of this. Of course, the data from Arad regarding the presence of the Old Hebrew script are now confirmed by the inscriptions from Reḥov Stratum IV. Finally, it should also be emphasized that from the site of Kuntillet ʿAjrud hail a number of Phoenician and Old Hebrew inscriptions, with most (Hebrew and Phoenician both) dating to the late ninth and early eighth centuries BCE (some of the Phoenician may be earlier, heirloom pieces, such as the inscribed stone bowl).38 The Old Hebrew inscriptions from Kuntillet ʿAjrud are very finely done (as are the Phoenician), often with a beautiful scribal hand and careful attention to letter environment and morphology. These are the product of a well-educated scribe or of a well-educated official with training in the scribal curriculum. The fact that this site is deep on the southern border of Judah (though perhaps frequented also by northern Israelites) and the fact that the Old Hebrew inscriptions from this site are from the late ninth or early eighth century are important data. the Hebrew script was differentiated from Phoenician” (W. M. S chniedewind , A Social History of Hebrew: Its Origins through the Rabbinic Period [ABRL; New Haven, CT: Yale University Press, 2013], 82). Alas, I am confident that Schniedewind will find few followers among trained epigraphers. After all, the epigraphic evidence demonstrates that a distinctive Old Hebrew script is being used in ninth-century inscriptions from Arad (e.g., Arad 76), as I (among others) have noted in the past, and now in the Reḥov inscriptions from Stratum IV and also, of course, in the Mesha Stela and el-Kerak Inscription, arguably as a result of the well-documented Omride hegemony. It is unfortunate that, in a book that will be as widely read as Schniedewindʼs, he misunderstands the epigraphic Old Hebrew data to this degree. Along those same lines, it is, alas, perhaps not surprising that Schniedewind entirely misconstrues and misunderstands the evidence that I (following J. Naveh) have marshaled regarding the chronological horizon for the origins of the Old Hebrew script in the ninth century BCE. Namely, Schniedewind comments on “the elongation of tails on letters that Rollston identifies as the beginnings of a distinctive Hebrew script” (82). In reality, I have stated precisely the opposite of the view that Schniedewind attributes to me! Here are my actual words: “I do not consider elongation to be a distinctive marker of a particular script series. My reason for this view is as follows: the Phoenician, Aramaic, and Old Hebrew script series all reflect elongation” (Rollston, “Phoenician Script of the Tel Zayit Abecedary” [see n. 9], 83–89). I am afraid that I do not know how I could have been clearer. Also interesting is the fact that Schniedewind refers to my discussion in Rollston , Writing and Literacy (see n. 8) as his putative source for my discussion of elongation (42–46). Actually, though, I do not discuss elongation there, but rather I discuss subjects such as letter stance and letter curvature (in a comparative discussion of the Phoenician, Hebrew, and Aramaic scripts). In any case, for a nice example of the differences between the Old Hebrew script and the Phoenician script in the late ninth or early eighth century from a single archaeological site, I would suggest a simple glance at the Phoenician and Old Hebrew scripts attested side-by-side at the site of Kuntillet ʿAjrud, as I think that the differences between these two national scripts are discernible even to nonspecialists in epigraphy. For photos, drawings, and discussions of the Kuntillet ʿAjrud inscriptions, see S. A ḥituv et al., “The Inscriptions,” in Kuntillet ʿAjrud (Ḥorvat Teman): An Iron Age II Religious Site on the Judah-Sinai Border (ed. Z. Meshel; Jerusalem: Israel Exploration Society, 2012), 73–142. 38  S. Aḥituv et al., “Inscriptions” (see n. 37), 73–142.

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

35

It is perhaps useful for me to emphasize that script typology is the same sort of science as is pottery typology. Indeed, they are twins, both empirical in nature, both important. And just as someone trained well in pottery typology should have the ability to discuss in an empirical manner the dating and variants

Figure 11. Samaria Ostracon 17a (Old Hebrew language and script; drawing by Christopher A. Rollston).

of pottery vessels, so also someone trained well in script typology should be able to discuss in an empirical manner the dating and variants of major and minor script series (e.g., Phoenician, Old Hebrew, and Aramaic). Of course, Frank Moore Cross was famous for suggesting that someone attempting to do typology (script, pottery) must also have an “eye for form,” hence, the title of the Gedenkschrift in his honor, a volume that begins with a statement about epigraphic methodology.39 The Old Hebrew ostraca from Samaria derive from the first three decades of the eighth century BCE and are administrative in nature (figure 11). These ostraca, that is, the Reisner Samaria Ostraca, number in excess of one hundred. Some sixty were published long ago.40 The remainder of the ostraca were not considered legible at the time, but Ivan Kaufman was able to read and decipher many of these.41 Most contain reference to a year (arguably a regnal year of King Jeroboam II, r. 786–746 BCE), with some place names, personal names, and commodities referred to. There has been much ink spilled in attempting to 39  C. A. Rollston, “Prolegomenon to the Study of Northwest Semitic Palaeography,” in An Eye for Form: Epigraphic Essays in Honor of Frank Moore Cross (ed. J. A. Hackett and W. E. Aufrecht; Winona Lake, IN: Eisenbrauns, 2014), 1–4. 40  G. A. R eisner et al., Harvard Excavations at Samaria, 1908–1910 (Cambridge, MA: Harvard University Press, 1924). 41  I. T. Kaufman, “The Samaria Ostraca: A Study in Ancient Hebrew Palaeography, Texts and Plates” (Ph.D. diss., Harvard University, 1966).

36

Christopher A. Rollston

debate the precise function of these ostraca. Some have considered them to be records of commodities coming to crown coffers. After all, Samaria was the capital of the Northern Kingdom. In such a case, the commodities might have been produced on royal farmland and production centers (e.g., olive presses), or the commodities might have been payment of taxes (in the form of commodities) coming to the crown. Conversely, some have argued that the commodities listed on the ostraca are the crownʼs expenditures, given as payment to those serving the crown. One of the ostraca reads: “In the tenth year, to Gadyaw, from ʾAza, Abibaʿl, Ahaz, Sheba, Meribaʿl.” Another reads: “In the tenth year, wine of the vineyard of the Tel (and) a jar of refined olive oil.” Regardless of whether these are viewed as receipts of income or expenditures, these sixty or so legible ostraca, plus the additional ones that Kaufman was able to decipher on the basis of Reisnerʼs glass negatives, do provide some reliable evidence for the bureaucracy of the Northern Kingdom of Israel in the early eighth century BCE. In addition, and at least as important, these ostraca reveal that there were two distinct dialects of Old Hebrew, namely, the southern dialect of Old Hebrew known from Judean sites such as Lachish and Arad and the northern dialect that is attested on the Reisner Samaria Ostraca (a site in the Northern Kingdom of Israel). Although many of the differences between these two dialects will never be known, some are apparent. For example, in the northern dialect of Old Hebrew (“Israelite Hebrew”), the word for “year” was šat, to be contrasted with the southern dialect of Old Hebrew (“Judahite Hebrew”), which used šānâ. In addition, within the southern dialect of Old Hebrew, the diphthong ayi was preserved; hence the word for “wine” is written yyn (for yayin) in Judahite, but in the northern dialect of Old Hebrew, this diphthong was contracted and so this same word is written yn (for yēn) in Israelite Hebrew. Of course, in this connection it is also worth mentioning that a seal was found at Megiddo (in 1904) that reads “Belonging to Shemaʿ, servant of Jeroboam.”42 The Jeroboam referenced here is King Jeroboam II of the Northern Kingdom of Israel, and Shemaʿ was one of his high royal official, as can be deduced from his title. Finally, it can also be mentioned that from the Moabite site of Tell Ataruz in Jordan hails a very difficult and brief but very important inscription from the ninth century.43 This inscription employs some very impressive hieratic numer42  See N. Avigad and B. S ass , Corpus of West Semitic Stamp Seals (Jerusalem: Israel Exploration Society, 1997), 49, no. 2. 43  The site is being excavated by C.-H. Ji. I am in the process of preparing this inscription for publication collaboratively with P. K. McCarter, S. Wimmer, and A. Bean. For my preliminary discussion of the inscription, see C. A. Rollston, “The Ninth Century ‘Moabite Pedestal Inscription’ from King Meshaʼs Ataruz: Preliminary Synopsis of an Excavated Epigraphic Text and its Biblical Connections,” Rollston Epigraphy: Ancient Inscriptions from the Levantine World, Dec. 17, 2013, www.rollstonepigraphy.com/?p=631. For discussion of the archaeological context, see especially C.-H. Ji, “The Early Iron Age II Temple at Khirbet Atarus and Its Architecture and Selected Cult Objects,” in Temple Building and Temple Cult: Architecture

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

37

als, accompanied by writing in the same script as the Mesha Stela and el-Kerak Inscription. The inscription arguably dates to a decade or two after the Mesha Stela; the presence of hieratic numerals at such an early date is impressive. The cumulative evidence is weighty, demonstrating that professional scribes were functioning nicely in the southern Levant during the tenth, ninth, and eighth centuries BCE, mostly within administrative contexts associated with governments. But even in the case of nonadministrative materials (e.g., presumably those of Kuntillet ʿAjrud), the script and orthography is good enough to argue that these inscriptions were from the hands of scribes trained in a professional manner. Ultimately, the epigraphic evidence is clear: there were scribes writing texts during the tenth, ninth, and early eighth centuries in the southern Levant. There is really no convincing way to avoid this conclusion. The epigraphic data are simply overwhelming.

III. The Epigraphic Record and the Bible: Evidence for Royal Annals in Ninth and Eighth Century Israel and Judah The books of Kings are literary compositions with discernible political and religious vantage points and a long history of transmission and redaction,44 but there is some core historical material in some of the pericopes of Kings, a fact that can be demonstrated in a cogent manner on the basis of textual evidence from the ancient Near Eastern and southern Levantine worlds. Because it is the contention of this article that there were already scribal elites in Israel and Judah during the ninth century BCE, the focus here shall be on correspondence between biblical and epigraphic materials about historical figures of the ninth century and the beginning of the eighth century BCE.45 and Cultic Paraphernalia of Temples in the Levant (ed. J. Kamlah; Wiesbaden: Harrassowitz, 2012), 203–221; I. Finkelstein and O. Lipschits, “Omride Architecture in Moab: Jahaz and Ataroth,” ZDPV 126 (2010), 29–42. 44  With regard to the vantage points of ancient writers of historical and literary texts, I find myself very much in agreement with the brilliant analyses of scholars such as Baruch Halpern and Marc Brettler. See especially B. Halpern, The First Historians: The Hebrew Bible and History (Cambridge: Harper & Row, 1988); M. Z. B rettler , The Creation of History in Ancient Israel (London: Routledge, 1995). One of the finest histories of Israel during the recent past is that of M. Liverani, Israelʼs History and the History of Israel (trans. C. Peri and P. R. Davies; London: Equinox, 2005), with nuanced and detailed discussion of the histories of both Israel and Judah (104–199). 45  I am indebted to L. J. M ykytiuk , Identifying Biblical Persons in Northwest Semitic Inscriptions of 1200–539 BCE (AcBib 12; Atlanta: Society of Biblical Literature, 2004), idem , “Archaeology Confirms Fifty Real People in the Bible,” BAR 40 (Mar./Apr. 2014), 42–50, 68; and A. R. Millard, “Israelite and Aramean History in the Light of Inscriptions,” in Israelʼs Past in Present Research: Essays on Ancient Israelite Historiography (ed. V. P. Long;

38

Christopher A. Rollston

The biblical book of Kings refers to the fact that Omri was the king of the Northern Kingdom of Israel (1 Kgs 16:16–28), reigning for some “twelve years” (ca. 882–871 BCE). There is also reference to Ahab as Omriʼs son and successor (1 Kgs 16:28–30), credited with a reign of some “twenty-two years” (ca. 873–852 BCE). Kings also refers to King Mesha of Moab as a vassal of the Northern Kingdom of Israel and to Meshaʼs rebellion against the Omrides after the death of Ahab (2 Kgs 1:1; 3:4–5). The Hebrew Bible refers to Kemosh as the national God of the Moabites (Num 21:29; 1 Kgs 11:7, 33; 2 Kgs 23:13). Among the cities that the Hebrew Bible places in the geographic region of Moab are Dibon, Ataruz, and Nebo (Num 32:3,38; 33:47, etc.). The Mesha Stela (discussed briefly above) hails from the ninth century BCE. It is written in the Moabite language and in the Old Hebrew script. In this inscription, written in the first person, Mesha refers to himself as the king of Moab and states that he is from the city of Dibon (ll. 1–3). He also states that “King Omri of Israel had oppressed Moab for many days” (ll. 4–5), a situation that continued during much of his “sonʼs” reign, totaling some forty years (a stock, round number). But after a time, Mesha decided to rebel, with the full support of the national God, Kemosh. Ultimately, according to his stela, Mesha was able to regain much of the historic Moabite territory from the Omrides, even “dragging the vessels of Yahweh from Nebo into the presence of Kemosh” after he retook Nebo (ll. 7–18). Mesha also states that he wrested control of Ataruz from Israel.46 As is often the case with texts from antiquity (literary or epigraphic), there are some tensions between the Hebrew narratives in Kings and the Moabite narrative in the Mesha Stela.47 Nevertheless, the fact that there are broad correspondences SBTS 7; Winona Lake, IN: Eisenbrauns, 1999), 129–140; and M. Cogan, The Raging Torrent: Historical Inscriptions from Assyria and Babylonia Relating to Ancient Israel (Jerusalem: Carta, 2008). 46  The translations of the Mesha Stela are from R ollston , Northwest Semitic Royal Inscriptions (see n. 11). 47  For example, the biblical narrative suggests that the rebellion of Mesha occurred shortly after the death of Ahab, at the beginning of the reign of King Jehoram of Israel (r. ca. 849–842; 2 Kgs 3:5–6). But the Moabite account seems to suggest that the rebellion occurred shortly after the death of Omri and thus during the beginning of the reign of Omriʼs “son.” Part of the difficulty here is the fact that the term son (bn) can function in a broader sense and so could readily refer to a grandson of Omri, just as F. M. Cross and D. N. Freedman, Early Hebrew Orthography [New Haven: American Oriental Society, 1952], 39–40, have argued. Furthermore, in terms of tensions between the two narratives, from the Hebrew narrative in Kings it would seem that Meshaʼs rebellion was quelled (2 Kgs 3:6–27), but based on the Moabite narrative in the Mesha Stela it would seem that Mesha enjoyed victory after victory (Mesha Stela, ll. 10–21). Of course, it should be noted that it is plausible to argue that the Mesha Stela and the material about Meshaʼs rebellion in Kings are about two different incidents in the history of Israelite hegemony and Moabite rebellion, during the course of Meshaʼs reign. Naturally, it is also possible that either the Moabite text or the Hebrew text just got the facts about the rebellion or the succession wrong (and do note that the Mesha Stela does not provide

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

39

between the epigraphic text of the ninth century and the Deuteronomistic History is readily apparent. The book of Kings notes that Omri was a king of Israel during the first half of the ninth century BCE, as does the Mesha Stela. The book of Kings mentions Mesha as a king of Moab who was under Omride hegemony, as does the Mesha Stela. The national God of Moab is named as Kemosh in both the Mesha Stela and the Hebrew Bible. The Hebrew Bible and the Mesha Stela refer to the national God of Israel as Yahweh. The Mesha Stela and the book of Kings both declare that during the reign of King Mesha of Moab, Mesha rebelled against his Omride hegemon. The Mesha Stela mentions cities such as Dibon, Ataruz, and Nebo as Moabite territories, details affirmed in the Hebrew Bible. The totality of such correspondences suggests that the Deuteronomistic Historian(s) had access to written records with accurate historical details about international relations, including major regal figures, in the ninth century BCE. The book of Kings refers to Omriʼs son and successor, Ahab, having married a Phoenician princess, namely, Jezebel the daughter of Ethbaʿl of Sidon (1 Kgs 16:31). Naturally, this functioned as a marital alliance between Israel and Phoenicia, a practice very well attested in the ancient Near Eastern world.48 There are numerous references to military tensions with Syrian kings (1 Kgs 20), alongside the statement that a treaty was established between King Ahab of Israel and King Ben-Hadad (= Hadad-ʿezer) of Damascus (1 Kgs 20:34). The narrative in Kings suggests that the treaty held for some three years before tensions erupted again between Israel and Syria, culminating in the battle of Ramoth-Gilead, around the year 850 BCE, in which King Ahab of Israel was slain (1 Kgs 22). The Kurukh Monolith of the Neo-Assyrian King Shalmaneser III (r. 858– 824 BCE) recounts Shalmaneserʼs victory at the battle of Qarqar in 853 BCE (ii 78–102). A. Kirk Grayson has argued that the Kurukh Monolith of Shalmaneser III was engraved in late 853 or early 852 BCE.49 In his account of the battle, Shalmaneser III states that a coalition of twelve Levantine kingdoms had formed the personal name of the “son” of Omri, perhaps suggesting a dearth of precise knowledge about succession there). Finally, it is also worth noting that this material in Kings is in the midst of the Elijah and Elisha Cycle, that is, in the midst of narratives about two prophetic figures of the Northern Kingdom of Israel. It seems reasonable to contend that this Cycle hailed from circles in the Northern Kingdom of Israel, just as M. Cogan and H. Tadmor, II Kings: A New Translation with Introduction and Commentary (AB 11; New York: Doubleday, 1988), 48, among others, have done. 48  For a very useful discussion of diplomatic marriages attested in the ancient Near Eastern world, see especially S. A. M eier , “Diplomacy and International Marriages,” in Amarna Diplomacy: The Beginnings of International Relations (ed. R. Cohen and R. Westbrook; Baltimore: Johns Hopkins University Press, 2000), 165–173; For a further sense of the prominence and intricacies of marital alliances in ancient international relations, see some of the following Akkadian, Hurrian, and Hittite letters in W. L. Moran, The Amarna Letters (Baltimore: Johns Hopkins University Press, 1992), EA 1–4; 11–13; 19–21; 23–24; 29; 31–32. 49  A. K. G rayson , Assyrian Rulers of the Early First Millennium BC: Volume II, 858– 745 BC (RIMA 3; Toronto: University of Toronto Press, 1996), 11.

40

Christopher A. Rollston

against him, and it was these with which he did battle. The first three kings listed are “Hadad-ʿezer (Adad-idri) of Damascus,” who contributed “1200 chariots, 1200 cavalry, and 20,000 troops”; “Irhuleni of Hamat,” who contributed “700 chariots, 700 cavalry, and 10,000 troops”; and “Ahab the Israelite,” who contributed “2000 chariots and 10,000 troops.”50 Among the remaining members of the coalition, according to Shalmaneser III, the Phoenicians were particularly prominent (e.g., Byblos, Irqanatu, Arvad, Usanatu, Shianu). The narratives of the biblical book of Kings, therefore, and the narrative in the Kurukh Monolith contain a number of correspondences. First and foremost is the fact that Kings refers to Ahab as king of Israel, and the Kurukh Monolith refers to King Ahab of Israel as well (note also that the determinative for a country is used in conjunction with the term Israel in the latter). Moreover, both place Ahab squarely in the first half of the ninth century BCE. In addition, the narrative of Kings refers to a marital alliance between Ahab of Israel and Ethbaʿl of Sidon. Similarly, the Kurukh Monolith refers to a number of Phoenician citystates that had allied with Ahab of Israel. Furthermore, Kings states that although battles were fought between Ahab of Israel and Ben-Hadad (a name that is used to refer to multiple Damascene kings, including Hadad-ʿezer of Damascus), at one point there was a treaty established between them, a treaty that lasted some three years (1 Kgs 20:34; 22:1). It is entirely reasonable to contend that the treaty between Damascus and Israel mentioned in Kings is the same treaty that Shalmaneser III is referencing in the Kurukh Monolith as having taken place in 853 BCE (that is, the treaty that was part of the formation of the anti-Assyrian alliance of Levantine states). To be sure, Shalmaneser III provides additional details about the breadth of membership in this coalition of Levantine states, but it is difficult to avoid the conclusion that the same alliance is referenced. Note that there has been a long discussion in the secondary literature regarding the association of some of this material (1 Kgs 20) with Ahab, with some scholars arguing that connecting this material with Ahab is not convincing. Among the things that are often emphasized in this vein is that Ahab of Israel seems to be portrayed as very strong in the Kurukh Monolith, supplying more forces than either Hadad-ʿezer of Damascus or Irhuleni of Hamat, but then Ahab seems to be portrayed as substantially weaker than Damascus in the narrative in Kings (1 Kgs 20:1–6).51 But note that the Kurukh Monolith numbers the foot soldiers 50  These translations are those of G rayson , Assyrian Rulers (see n. 49), 23 (ii 89–92). N. Na‍ʾaman has argued that Ahab contributed two hundred, not two thousand, chariots to the coalition (N. Na ‍ʾaman , “Two Notes on the Monolith Inscription of Shalmaneser III from Kurukh,” Tel Aviv 3 [1976], 89–106, here 97–102). Personally, I do not find the number two thousand to be too high, especially in light of the fact that Ahab is not said to have contributed cavalry to the coalition (in contrast to Irhuleni and Hadad-ʿezer, both of whom did contribute substantial cavalry). 51  For discussion of the primary and secondary literature, see especially W. T. P itard , Ancient Damascus: A Historical Study of the Syrian City-State from Earliest Times until Its

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

41

of Hadad-ʿezer as twenty thousand, whereas those of Ahab are numbered ten thousand. Furthermore, Ahab is not said to have provided any cavalry, but Hadad-ʿezer is said to have supplied 1,200. Thus, Ahab may have supplied some eight hundred more chariots than Hadad-ʿezer (i.e., 1,200 to Ahabʼs two thousand), but it does not seem reasonable to suggest that the numbers of the Kurukh Monolith demonstrate that Ahab was a very powerful regional king with military resources that greatly surpassed those of Hadad-ʿezer. Finally, although I suspect there is substantial hyperbole in the numbers, the narrative in Kings suggests that the Damascene king had formed a substantial coalition to assist him in his punitive raid against Israel. Suffice it to say that there are some broad correspondences between the Mesopotamian cuneiform records regarding Ahab of Israel and the account in the Deuteronomistic History. This fortifies further the contention that the Deuteronomistic Historian had access to some sort of written historical records. The book of Kings calls Jehu (r. ca. 842–815 BCE) the king of Israel and Hazael (r. ca. 842–806 BCE) as the king of Damascus (Aram). According to the tradition enshrined in Kings, Elijah the prophet is commissioned by Yahweh to anoint Jehu as king of the Northern Kingdom of Israel and to anoint Hazael as the king of Damascus (1 Kgs 19:15–16). Moreover, this pericope contains a striking statement, “And it shall be that whoever might escape from the sword of Hazael, Jehu will kill, and whoever may escape from the sword of Jehu, Elisha shall kill” (1 Kgs 19:17), suggesting an alliance of sorts between Jehu and Hazael at the beginning of their reigns. Ultimately, it is Elijahʼs successor, Elisha, who orchestrates the anointing of Jehu as king of Israel (2 Kgs 9:1–6), and it is Elisha who predicts to Hazael that he will be the king of Aram (2 Kgs 8:7–13). The book of Kings indicates that both Jehu of Israel and Hazael of Damascus are usurpers. Hazael is reported to have smothered Ben-Hadad to death in his own bed (2 Kgs 8:15). And according to Kings, Jehoram of Israel (r. 849–842 BCE), a son and successor of Ahab of Israel (2 Kgs 3:1), along with his ally, King Ahaziah of Judah (r. 842 BCE), a son and successor of King Jehoram of Judah (2 Kgs 8:25), had been engaged in battle against Hazael of Damascus (2 Kgs 8:26–28). During a pitched battle, Jehoram of Israel was wounded by Hazael and was recovering in Jezreel (2 Kgs 9:14–15). After his anointing, the book of Kings mentions that Jehu traveled by chariot toward Jezreel. According to Kings, it was from Jezreel that King Jehoram of Israel and King Ahaziah of Judah set out in their chariots to meet Jehu, hoping that Jehu was coming in peace. Upon realizing that it was treachery, Jehoram of Israel attempted to flee, but Jehu pulled his bow and fired an arrow into Jehoramʼs back, piercing his heart and killing him immediately. Seeing the fate of Jehoram of Israel, Ahaziah of Judah attempted to flee in his chariot, but those loyal to Jehu shot Ahaziah of Judah, mortally wounding him. Ahaziah fled to the northern Israelite fortress Fall to the Assyrians in 732 B.C.E. (Winona Lake, IN: Eisenbrauns, 1987), 114–125.

42

Christopher A. Rollston

of Megiddo and died there. Later, his body was transferred to Jerusalem, where he was buried in his ancestral tomb (2 Kgs 9:21–29). Therefore, the material in Kings can be understood to state that both Hazael of Damascus and Jehu of Israel came to power at the same time, both through usurpations. Moreover, both Hazael and Jehu are said to have engaged in violence against Jehoram of Israel and Ahaziah of Judah, with Hazael seriously wounding Jehoram of Judah but with Jehu responsible for the mortal blows. Tensions later developed between Hazael and Jehu and between Hazael and Jehoahaz (r. 815–801 BCE), the son and successor of Jehu, but in ca. 842 BCE (2 Kgs 10:32–34; 2 Kgs 13:1–9), at the time that both Jehu and Hazael came to the throne, the narrative states that they were allies, as does also a brief but important text that is buried in the midst of the Elijah and Elisha cycles (1 Kings 19:17). Finally, it should be noted that Kings mentions that Hazael also decided to make a punitive raid against Jerusalem and that Jehoash of Judah (r. 837–800 BCE) plundered his own treasuries to give to Hazael, who then withdrew (2 Kgs 12:17–18). According to the various inscriptions in Mesopotamian cuneiform, Shalmaneser III made numerous punitive campaigns to the Levant. After the battle of Qarqar, Shalmaneser III consistently refers to his battles with Hadad-ʿezer of Damascus and Irhuleni of Hamat but not to Ahab of Israel (who had died in ca. 850 BCE). However, in his eighteenth year (ca. 841 BCE), Hadad-ʿezer of Damascus is no longer mentioned; rather, Hazael is now the king of Damascus. The Kurbail Statue Inscription of Shalmaneser III (which Grayson dates to 839 or 838 BCE) contains the following words (ll. 21–28): In my eighteenth regnal year, I crossed the Euphrates for the sixteenth time. Hazael of Damascus, trusting in the might of his soldiers, carried out an extensive muster of his troops. He fortified Mount Saniru, the mountain peak which is before Mount Lebanon. I fought with him (and) defeated him. I put to the sword 16,000 of his fighting men (and) took away from him 1,121 of his chariots (and) 470 of his cavalry with his military camp. To save his life, he ran away (but) I pursued (him). I imprisoned him in Damascus, his royal city, (and) cut down his gardens. I marched to Mount Hauranu (and) razed, destroyed, burned, (and) plundered cities without number. I marched to Mount Ba‍ʾalira‍ʾasi, which is a cape (jutting out into) the sea, (and) erected my royal statue there.

Immediately after these words, Shalmaneser III declares that “at that time, I received tribute from the people of Tyre, Sidon, (and) from Jehu (Iaua) of the house of Omri (Humri).”52 From the Black Obelisk of Shalmaneser III, which Grayson dates to late 828 BCE or 827 BCE (based on the fact that the last reference is to a campaign in Shalmaneser IIIʼs thirty-first year, that is, 828 BCE) come these details: “I received tribute from Jehu (Iaua) of the house of Omri  Grayson, Assyrian Rulers (see n. 49), 60; for additional references to both Hazael and Jehu in the inscriptions of Shalmaneser, see pp. 48 (monumental bulls from Calah) and 54 (Assur stone tablet). 52

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

43

(Humri): silver, gold, a gold bowl, a gold tureen, gold vessels, gold pails, tin, the staffs of the kingʼs hand, (and) spears.”53 And, of course, the Black Obelisk of Shalmaneser III also contains a relief that depicts Jehu of Israel prostrating himself before Shalmaneser III. Moreover, on a separate inscription known as the Broken Statue of Shalmaneser III from Assur, Shalmaneser III writes: Hadad-ʿezer (Hadad-Idri) passed away (and) Haza‍ʾel, son of nobody, took the throne. He mustered his numerous troops (and) moved against me to wage war and battle. I fought with him (and) defeated him. I took away from him his walled camp. He fled to save his life (and) I pursued (him) as far as Damascus, his royal city. [I cut down his gardens . . .].54

Note, of course, that the term son of nobody is used to indicate that Haza‍ʾel was not the legitimate heir to the throne but rather a usurper (perhaps from a collateral royal line, but a usurper in the eyes of Shalmaneser III). In any case, according to the Black Obelisk of Shalmaneser III, Hazael continued his bellicose methodology; thus, Shalmaneser III also refers in the Black Obelisk to defeating Hazael in his (Shalmaneser IIIʼs) twenty-first regnal year: “In my twenty-first regnal year, I crossed the Euphrates for the twenty-first time (and) marched to the cities of Hazael of Damascus. I captured four cities (and) received tribute from the people of the lands Tyre, Sidon, (and) Byblos.”55 A small cylinder (bored through the center) of black stone, discovered at Assur, reads as follows: “Booty from the temple of the deity Sheru of the city Malaha, a royal city of Hazael of Damascus, which Shalmaneser III, son of Ashurnasirpal II, king of Assyria, brought back inside the wall of Inner City (Assur).”56 From Nimrud hails a piece of ivory with the words “[Belonging to Lo]rd Haza‍ʾel.”57 Within the corpus of Northwest Semitic inscriptions, a pair of horse forehead ornaments from the Heraion of Samos are inscribed in Old Aramaic with the words, “That which Hadad gave our lord Hazael from ʿUmqi, in the year that our lord crossed the river.”58 A pair of horse blinkers from the Apollo Daphnephoros temple at Eretria in Euboia bear the same Old Aramaic inscription.59 Similarly, an ivory inscription from Arslan-Tash reads: “That [. . .] the son of Amma for our lord Hazael in the year [. . .].”60 The Zakur Stela refers to Zakur as the king of Hamat and Luʿat and as a successor of Iruleni of Hamat. Zakur refers to his own rise to power with the full support of the God Baʿlshamayin. In addition,  Grayson, Assyrian Rulers (see n. 49), 149.  Grayson, Assyrian Rulers (see n. 49), 118 (i 25–ii 6). 55  Grayson, Assyrian Rulers (see n. 49), 67 (ll. 102–104). 56  Grayson, Assyrian Rulers (see n. 49), 151 (ll. 1–8). 57  Cogan, Raging Torrent (see n. 45), 35. 58  The readings and translations are those of I. E ph ʿal and J. N aveh , “Hazaelʼs Booty Inscriptions,” IEJ 39 (1989), 192–200, here 193. 59  Ephʿal and Naveh, “Hazaelʼs Booty Inscriptions” (see n. 58), 193. 60  Ephʿal and Naveh, “Hazaelʼs Booty Inscriptions” (see n. 58), 197. 53 54

44

Christopher A. Rollston

he states that a coalition of kings, including “Bar-Hadad son of Hazael, king of Aram, established an alliance (against me).”61 Thus, Hazael figures prominently in the epigraphic record. The Tel Dan Stela Inscription (written in Old Aramaic, as noted earlier) is of course relevant as well. Among other things, King Hazael of Damascus asserts in the Tel Dan Stela that he killed “[Jeho]ram son of [Ahab], king of Israel” and “[Ahaz]iah son of [Jehoram kin]g of the House of David.” At first blush, someone might suggest that Hazaelʼs declaration is in serious tension with the biblical account, which attributes the killing of Jehoram of Israel and Ahaziah of Judah to Jehu (2 Kgs 9), but further probing of the textual data is useful. As noted previously, Kings says that King Jehoram of Israel and King Ahaziah of Judah had been warring against King Hazael of Damascus. During one of the battles against Hazael, Jehoram of Israel was seriously wounded, causing him to convalesce in the northern Israelite city of Jezreel (2 Kgs 8:26–29). Subsequently, according to the narrative in Kings, Jehu traveled to the region of Jezreel and killed Jehoram (who had first been seriously wounded by Hazael), and he also killed Ahaziah of Judah. Furthermore, of course, Mesopotamian texts (e.g., the Black Obelisk of Shalmaneser III) demonstrate that Hazael of Damascus and Jehu of Israel had been part of an anti-Assyrian alliance. In addition, a text from the Elijah and Elisha cycles strongly suggests that Jehu and Hazael had formed an alliance of some sort (1 Kgs 19:17). The biblical book of Kings states that Hazael of Damascus was a usurper who assassinated Ben-Hadad (Hadad-ʿezer). The cuneiform inscriptions of Shalmaneser III refer to Hadad-ʿezer of Damascus as the king of Damascus often, but in the records of Shalmaneser IIIʼs campaign in his eighteenth year, 841 BCE, the king of Damascus is Hazael. In addition, Shalmaneser IIIʼs inscriptions refer to Jehu as the king of Israel, dovetailing with the material in Kings. Furthermore, Shalmaneser III refers to Hazael as a “son of nobody,” a term indicating that he was as usurper, not the heir apparent of the Damascene throne. Again, this dovetails with the account in Kings of Hazaelʼs rise to power. The biblical text states that Jehoram of Israel and Ahaziah of Judah had combined forces to war against Hazael of Damascus in ca. 842 BCE (probably with Ahaziah the junior member of the alliance or perhaps entirely a vassal). The biblical text states that in a battle against Hazael Jehoram of Israel was seriously wounded and that Jehu of Israel killed both Jehoram of Israel and Ahaziah of Judah. That is, the biblical book of Kings indicates that Hazael and Jehu were engaged in battles with Jehoram of Israel and Ahaziah of Judah (battles that occurred in 842 BCE), and in those battles the sitting monarchs of Israel (Jehoram) and Judah (Ahaziah) were wounded and died. This dovetails reasonably well with the Tel Dan Stela, with the predictable difference being that in his inscription Hazael claims to have killed both, whereas in the Hebrew version in Kings the Israelite king is credited 61

 Rollston, Northwest Semitic Royal Inscriptions (see n. 11).

Inscriptional Evidence for the Writing of the Earliest Texts of the Bible

45

with (or blamed for) this. The end result is that Jehu is the king of Israel and Hazael (after his assassination of Ben-Hadad, that is, Hadad-ʿezer) is the king of Damascus. Of course, as noted, this is precisely the description Shalmaneser III gives as well – that is, Jehu is king of the Northern Kingdom of Israel and Hazael is the king of Damascus. It is also worth mentioning that at the end of the ninth century and beginning of the eighth, Jehoash (r. ca. 801–786 BCE) is reported in the book of Kings to be the king of Israel (2 Kgs 13:10), and the inscriptions of the Neo-Assyrian king Adad-narari III (r. ca. 810–783) note that he “received the tribute of Joash (Iuʾasu), the Samarian, (and) of the people of Tyre (and) Sidon.”62

Conclusion Based on the total weight of the epigraphic evidence (including the correspondences between the epigraphic and biblical data), it is readily apparent that royal scribes, writing royal inscriptions in the Phoenician script, were functioning nicely in Byblos during the late eleventh, tenth, and early ninth centuries (and later). Moreover, it is also readily apparent that in the ninth century, the southern Levantine states of Damascus, Moab, and Ammon were producing monumental royal inscriptions. Moreover, a monumental inscription has been found in Samaria that hails from the late ninth or early eighth century. There is a growing number of nonmonumental inscriptions (which I believe hail mostly from officialdom, reflective especially of administrative activities) from the regions of Israel, Judah, and Philistia, demonstrating in a decisive manner that trained scribes were functioning in this region during the tenth, ninth, and early eighth centuries BCE (and, of course, later as well). Finally, it is most reasonable to argue that the writers of the biblical book of Kings who were responsible for narratives about the ninth century also had access to royal records of some detail, something that is evidenced by some strong correspondences (in some very precise details at times) between the epigraphic material and Kings (cf. also the references in the book of Kings to the “Annals of the Kings of Israel,” the “Annals of the Kings of Judah,” the “Annals of Solomon,” etc.). Certainly the debates about the dating of the biblical documents and putative sources will continue. But it is absolutely certain that anyone who argues that there was no capacity for the production of historical or literary texts in ancient Israel or Judah prior to the eighth century is not au courant with the epigraphic data. That argument is moribund.

62

 Grayson, Assyrian Rulers (see n. 49), 211 (ll. 7–8).

The Covenant Code Appendix (Exodus 23:20–33), Neo-Assyrian Sources, and Implications for Pentateuchal Study David P. Wright I have argued in detail that the Covenant Code (CC; Exod 20:23–23:19) used the Laws of Hammurabi (LH) as a primary generative source, both for the casuistic laws at the center of the composition and for the surrounding apodictic laws.1 This chief conclusion was accompanied by an embryonic argument that CC appears to have been composed in connection with a narrative that contextualized the law collection.2 My study, however, avoided discussing Exod 23:20–33, the passage that concludes the revelation of CC. Though attached to CC, this Covenant Code appendix (to which I will refer in what follows as CCA) reflects narrative more than legal themes. It exhorts the Israelites to faithfulness and promises that a divine messenger will bring them to their land and that divine terror will help them conquer their enemies. It now appears to me that CCA was in fact created as part of the process out of which the Covenant Code grew. It continues the use of the sequential template of LH and, through this, accesses motifs from other Mesopotamian sources – i.e., Assyrian royal inscriptions – to shape native traditions about Israelite origins and to chart a vision of pending military campaign in CCʼs narrative. It is part of CCʼs supporting narrative and points to the breadth of that narrative. All this has implications for pentateuchal study.

Background The argument that CC used LH as a primary source for its casuistic and apodictic laws underlies the argument of this essay. This need not be repeated in any great detail here, though some review is necessary as a basis. To do this, let me

 D. P. W right , Inventing Godʼs Law: How the Covenant Code of the Bible Used and Revised the Laws of Hammurabi (New York: Oxford University Press, 2009). 2  Wright, Inventing (see n. 1), 332–345. 1

48

David P. Wright

direct attention to the recent review by William Morrow.3 Using strict criteria for determining textual relationships, he concludes that the laws in Exod 21:18–32 (casuistic laws on injury, homicide of a slave, miscarriage, talion, slave injury, and the goring ox) do appear to be dependent on LH. Though he avoids extending the genetic argument to other parts of CC, seeing this core set of laws as dependent, in my view, provides an alternate entry point to the argument that much or all of CC is indeed dependent on LH.4 That is to say, if this core passage is dependent on LH, then it is reasonable to conclude that other passages that have similarities with LH are also dependent on LH or related sources, even though these similarities are less dense. This reasoning applies to various passages in the casuistic laws, including debt slavery and the related law on seduction (21:2–11; 22:15–16); homicide and some capital crimes (21:12–17); an ox goring an ox (21:35–36); animal theft (21:37–22:3); deposit (22:6–7); animal loss (22:9–12); and animal rental (22:13–14). These laws all have correlations with material in LH. Yet these correlations are not simple citations from the source. CC appears to have creatively transformed its source material, using a particular law from LH as a foundation while blending in motifs from other laws in LH and sometimes even from auxiliary sources. The Covenant Code appears to be responding to legal problems in the sources and reworking them to resolve these. For example, CCʼs debt-slave laws (21:2–11) correlate in their basic topic with LH 117. The latter describes a case where a debtor surrenders members of his household – wife, son, or daughter – to a creditor to pay off a debt. The law may also refer to the creditor surrendering himself for the debt, depending on how the verb ittandin (“he gave” versus “he surrendered himself”) is understood. The Covenant Codeʼs particular concern, it appears, was to distinguish how these various family members are to be treated. It uses laws from other places in LH or from an auxiliary source to supply details and to operate as a lens for reinterpretation. The Covenant Codeʼs primary response was splitting off the  W. Morrow, “Legal Interactions: The Mišpāṭîm and the Laws of Hammurabi,” BO 70 (2013), 309–331. For recognition of the similarity of Exod 21:18–32 to LH, see also C. Nihan, “Révisions scribales et transformations du droit dans lʼIsraël ancien: Le cas du talion (jus talionis),” in Loi et Justice dans la Littérature du Proche-Orient ancien (ed. O. Artus; BZABR 20; Wiesbaden: Harrassowitz, 2013), 123–158, here 127–128, n. 19. 4  For an entry to the argument through these laws, see D. P. Wright, “The Origin, Development, and Context of the Covenant Code (Exodus 20:23–23:19),” in The Book of Exodus: Composition, Reception, and Interpretation (ed. T. B. Dozeman et al.; FIOTL; VTSup 164; Leiden: Brill, 2014), 220–244; idem, “Miscarriage, Talion, and Slave Injury in the Covenant Code and Hammurabiʼs Laws,” in Gazing on the Deep: Ancient Near Eastern and Other Studies in Honor of Tzvi Abusch (ed. J. Stackert et al.; Bethesda, MD: CDL Press, 2010), 539–564. See my fuller examination of the implications of Morrowʼs limited conclusions, “Method in the Study of Textual Source Dependence: The Covenant Code,” in Subtle Citation, Allusion, and Translation in the Hebrew Bible (ed. Z. Zevit; Sheffield: Equinox, forthcoming). 3

The Covenant Code Appendix and Neo-Assyrian Sources

49

case of a daughter (vv. 7–11). In LH the daughter does not marry the creditor, but in CC she does. This resolved an expected eventuality: the creditor would probably have sex with the unattached woman. This required marriage. The logic and support for this was provided by a law similar to MAL A 55–56, which requires a seducer or rapist to marry his victim. The Covenant Code included a version of this law in 22:15–16 as a footnote at the end of its casuistic laws. To provide other details for the case of the daughter, CC drew on LH 148–149, which prescribes that a person must support a first wife if he takes a second wife when the first becomes displeasing. The Covenant Code wrote its law on a male debt slave (vv. 2–6) so that it could apply to a debtorʼs son or the debtor himself. It dealt with the wife of the creditor only insofar as this woman might enter servitude with her husband, the implication being that a wife cannot be given over to the creditor independently as in LH 117. This probably had to do with guarding against sexual exploitation, the fundamental concern in the law about a daughter. In a case where an unmarried male enters debt slavery, is given a wife, has children, and does not wish to leave his family, CC provided a legal rite that integrated the debt slave into the creditorʼs household as, essentially, a chattel slave. The Covenant Code derived this procedure by conceptually inverting the details of LH 282, a case where a chattel slave says “You are not my master!” in response to which his master cuts off his ear. Thus CCʼs debt-slavery law builds on LH 117, enfolding materials and motifs from LH 148–149, 282 and a law similar to MAL A 55–56 in a way that solves problems and otherwise transforms the basic case. It is difficult to claim that these correlations with LH and other Near Eastern legislation are merely fortuitous whereas those in 21:18–32 are due to dependence on LH, especially when both 21:18–32 and 21:2–11 remodel material from LH in a similar way. This logic extends to CCʼs apodictic laws (20:23–26; 22:20–23:19). Though these contain fewer specific phraseological contacts with LH, in combination with the casuistic laws they match the A-B-A pattern of Hammurabiʼs prologue-law-epilogue. Moreover, like Hammurabiʼs outer sections, the apodictic laws are the place where the personalities of the deities are on full view and where the lawgiver speaks in first person – in contrast to the theologically dry bodies of casuistic laws in both works. It is in the apodictic laws that CCʼs chief hermeneutical transformation, replacing Hammurabi with YHWH as lawgiver, appears.5 This contextually motivates and governs the transformations imple This point is particularly important in view of some recent analyses that see dependence on LH only in the casuistic laws and not the apodictic laws. See B. M. L evinson , “Is the Covenant Code an Exilic Composition? A Response to John Van Seters,” in In Search of Pre-Exilic Israel (ed. J. Day; JSOTSup 406; London: T&T Clark, 2004), 272–325 (republished in idem , “The Right Chorale”: Studies in Biblical Law and Interpretation [FAT 54; Tübingen: Mohr Siebeck, 2008], 276–330); D. M. Carr, Formation of the Hebrew Bible: A 5

50

David P. Wright

mented in the casuistic laws, even though the replacement of a human lawgiver with the deity does not explicitly appear there. The Covenant Codeʼs apodictic laws also reflect themes from the outer sections of LH, the cult in the initial apodictic laws, and a broader set of themes – the poor, justice, and the cult – in the final apodictic laws. Table 1 summarizes these correlations in themes and emphases between the two texts. Table 1: General Correlations in the Covenant Codeʼs Apodictic Laws and Prologue/Epilogue of the Laws of Hammurabi A

B

A

LHʼs Prologue/CCʼs Initial Apodictic Laws (20:23–26)

Casuistic Laws of LH and CC

LHʼs Epilogue/CCʼs Final Apodictic Laws (22:20–23:19)

The godsʼ personalities on full display

Deities appear limitedly, in juridical contexts

The godsʼ personalities on full display

First person of lawgiver

No reference to lawgiver (exception in 21:13–14 tied to 20:24)

First person of lawgiver

Common theme: cult

Common themes: the poor, justice, and cult

An ostensible problem with seeing a correspondence between the outer sections of the texts is the genre difference between them. The Covenant Code has apodictic law whereas LH has praise of the king. But this is resolved when we realize that, despite the broad correlation in the A-B-A structure, CCʼs apodictic laws follow more particularly the themes of a series of exhortations and admonitions in the epilogue of LH, whose genre is akin to apodictic law (LH cols. 47:59–49:17). This exhortatory block, as we may call it, outlines the model of justice and specific behaviors expected of a future king. It is accompanied by a short promise of blessing for the obedient king (49:2–17) and a long list of curses for a disobedient king (49:18–51:91), with which the epilogue ends.6 The future-king passage, cited in abbreviated form in appendix 1 to this paper, will be of importance later in this study. New Reconstruction (New York: Oxford University Press, 2011), 470–472; Morrow, “Legal Interactions” (see n. 3), 314–316. If the apodictic laws are omitted (along with Exod 21:1, 13–14), there is no explicit portrayal of the deity as author of the laws. The apodictic laws provide this hermeneutical innovation. This then raises the question of what motivated the later editor, who supposedly added the apodictic laws, to attribute those and the casuistic laws to YHWH. It would seem that the particular motivation for such was a response to the portrayal of Hammurabi as lawgiver. 6  Some of the language of the initial exhortations and admonitions (48:59–94) is repeated in the following blessing and curse sections (49:2–17 and 49:18–52).

The Covenant Code Appendix and Neo-Assyrian Sources

51

How the themes of the exhortatory block are reflected in CCʼs apodictic laws is clarified by first recognizing the organizational plan of the final apodictic laws. These laws have two parallel passages or “strings” (22:20–30 and 23:9–19) that surround a chiastic center (23:1–8).7 The two parallel strings each contain laws about the poor (22:20–26 // 23:9–12), speaking about sovereigns (22:27 // 23:13), and the cult (22:28–30 // 23:14–19).8 The chiastic core contains laws mainly about the pursuit of justice. The themes of the parallel strings in the final apodictic laws follow the sequence of similar themes in Hammurabiʼs exhortatory block, as summarized in table 2. The initial apodictic laws (20:23–26), for their part, reflect themes from Hammurabiʼs exhortatory block, but in a different way (again, see table 2). The focus here is more on cultic themes, including the cultic symbol for the sovereign (an altar instead of a statue), and on the memorialization of the name of the sovereign in the cult. Where all three passages of apodictic law intersect is in the theme of speaking about sovereigns, human or divine (20:24; 22:27; 23:13b).9 In sum, it seems arbitrary to claim that Exod 21:18–32 and other sections of casuistic law such as 21:2–11 are textually dependent on LH but that the apodictic laws, despite their similarities with the thematic structure of LH, are unrelated to this generative process. The complexity of CC as a whole, in terms of themes and genre, though usually seen as deriving from a complex process of textual growth, really appears to stem from its being a wide-ranging composition formulated on the plan and in imitation of LH. Table 2: Correlations between the Laws of Hammurabiʼs Epilogue and the Covenant Codeʼs Apodictic Laws and the Covenant Code Appendix LH Epilogue (cols. 47:59–51:91) = “Exhortatory Block,” ends in future-king passage (48:59–49:44) with blessings and curses (49:45–51:91)

Initial Apodictic Laws (Exod 20:23–26)

String I of Final Apodictic Laws plus Chiastic Core (Exod 22:20–23:8)

String II of Final Apodictic Laws plus Appendix (Exod 23:9–19+20– 33)

7  For the recognition of the parallelism between these strings in D and H, see my other paper in this volume, “Source Dependence and the Development of the Pentateuch: The Case of Leviticus 24.” 8  The basic structure was identified by C. C armichael , “A Singular Method of Codification of Law in the Mishpatim,” ZAW 84 (1972), 19–25; the observation was developed by B. S. Jackson, Wisdom-Laws: A Study of the Mishpatim of Exodus 21:1–22:16 (Oxford: Oxford University Press, 2006), 457–458. It is more fully articulated in Wright, Inventing (see n. 1), 10–16, 51–90, 286–321, 322–332, and connected to stimuli from the epilogue of Hammurabiʼs Laws. 9  For a diagram mapping these themes, see Wright, Inventing (see n. 1), 320.

52

David P. Wright

Three persons (immigrant, widow, orphan) not to be oppressed (22:20–23) Two laws on the poor: (A) no interest (v. 24) (B) pledge (vv. 25–26)

Three persons (the “weak,” orphan girl, widow) not to be oppressed (47:59–73)

Hammurabiʼs image and stela set up in the Esagil temple (47:75–78)

Immigrant not to be oppressed (23:9)

Two laws on the poor: (A) seventh-year produce (vv. 10–11) (B) rest seventh day (v. 12)

[beginning of CC] Altar instead of images to be made in cult (20:23–24a) Call to obedience. (23:13a) [related to future-king passage, below]

Hammurabiʼs name (“my name” šumī) recalled (zakāru) in the Esagil temple (47:93–48:2)

God and chieftain Recall (‫ )זכר‬of YHWHʼs name (“my (= king) not to be cursed (22:27) name”; ‫ )שמי‬in cult place (20:24bα) Sacrificial and cultic prescriptions (22:28–30)

Wronged man to visit the temple, appears before Hammurabiʼs statue and stela, and offers praise of Hammurabi to Marduk and Zarpanitu (called “lords,” bēlum/ bēltum) (48:3–58) End of prayer of praise says: well-being for the people, good omens, gods that “enter” (erēbum) the Esagil temple (48:34–58)

Names (‫ )שם‬of other gods not to be recalled (‫)זכר‬ (23:13b) Every male to appear before (emended: “see”) YHWH (the “Lord” ‫ )אדן‬at the sanctuary for festivals; offerings (23:14–19) [end of CC proper]

YHWH comes (‫)בוא‬ to the cult place and blesses (‫ )ברך‬the people (20:24bβ) Two laws on the altar: (A) no hewn stone (v. 25) (B) no stairs (v. 26) [casuistic laws follow, 21:1–22:19]

The Covenant Code Appendix and Neo-Assyrian Sources

Counsel to future king to obey and ensure justice Laws not to be altered Chiastic structure (48:59–49:16) Short blessing and many curses Includes name motif Predicated on obedience or disobedience (49:16–51:91)

Laws ensuring justice Justice not to be perverted Chiastic structure (23:1–8)

53

CC appendix, with blessings (‫ )ברך‬of food, health, fertility, long life for Israelites and implicit curses of political and military destruction of Canaanites Predicated on obedience Includes name motif (23:20–33)

Hammurabiʼs Epilogue and the CC Appendix The foregoing sets the stage for understanding the generative relationship of CCA to CC proper. The first point of evidence is that CCA, which immediately follows the final apodictic laws, reflects motifs of the curse section in the last half of Hammurabiʼs epilogue and thus continues the sequential use of LH following from the final apodictic laws (see table 2, lower right corner). This passage completes the sequence of themes in the second string of the final apodictic laws. It matches the last section of the exhortatory block that enjoins a future king to follow and preserve Hammurabiʼs model of justice with attendant blessings or curses. If we allow for the conceptual inversion of ideas, a hermeneutic observable in other places where CC appears to have used LH (the preceding section of this essay drew attention to some examples of this), the blessings promised the Israelites in CCA (vv. 25–26) have a number of correlations with the curses of the epilogue. Citations from the texts are provided in table 3 and are referenced by paragraph numbers, showing their relative position in the list of curses.10 The blessing of food and water in v. 25a relates to the curses from the gods Ea (§ 4) and Adad (§ 7); the removal of sickness in v. 25b to the curse by Ninkarrak (§ 12); the prevention of infertility in v. 26a to the curse by Nintu (§ 11); and long life in v. 26b to the curse by Enlil (§ 2).

10  § 1 Anu (col. 49:45–52), § 2 Enlil (49:53–80), § 3 Ninlil (49:81–97), § 4 Ea (49:98–50:13), § 5 Shamash (50:14–40), § 6 Sin (50:41–63), § 7 Adad (50:64–80), § 8 Zababa (50:81–91), § 9 Ishtar (50:92–51:23), § 10 Nergal (51:24–39), § 11 Nintu (51:40–49), § 12 Ninkarrak (51:50–69), § 13 great gods of heaven (51:70–83), § 14 Enlil (51:84–91).

54

David P. Wright

Table 3: Blessings for the Israelites and Epilogue Curses (Rothʼs translation of the Laws of Hammurabi11) Exod 23:25a: He will bless (‫)וּב ַרְך‬ ֵ your food and your water (‫ימיָך‬ ֶ ‫ת־מ‬ ֵ ‫ת־ל ְח ְמָך וְ ֶא‬ ַ ‫)א‬. ֶ Ea § 4: May he (Ea) dam up his rivers at the source; may he not allow any life-sustaining grain (ašnan) in his land. Adad § 7: May the god Adad [. . .] deprive him of the benefits of rain from heaven and flood from the springs, and may he obliterate his land through destitution and famine (ina ḫušaḫḫim u bubūtim) Exod 23:25b: I will remove sickness (‫)מ ֲח ָלה‬ ַ from among you. Ninkarrak § 12: May the goddess Ninkarrak [. . .] cause a grievous malady (murṣam kabtam) to break out upon his limbs, and evil demonic disease (assakam lemnam), a serious carbuncle (simmam marṣam) which cannot be soothed, which a physician cannot diagnose, which he cannot ease with bandages, which, like the bite of death, cannot be expunged (la innassaḫu) [. . .]. Exod 23:26a: There will be no (female) who miscarries or is barren (‫)מ ַשׁ ֵכּ ָלה וַ ֲע ָק ָרה‬ ְ in your land. Nintu § 11: May the goddess Nintu [. . .] deprive him of an heir and give him no offspring; may she not allow a human child to be born among his people. Exod 23:26b: I will let you live the full number of your days (‫ת־מ ְס ַפּר יָ ֶמיָך ֲא ַמ ֵלּא‬ ִ ‫)א‬. ֶ Enlil § 2: May the god Enlil [. . .] incite against him [. . .] disorder that cannot be quelled and a rebellion that will result in his obliteration (gabaraḫ ḫalāqišu); may he cast as his fate a reign of groaning, of few days (ūmī īṣūtim), of years of famine, of darkness without illumination, and of sudden death (mūt niṭil īnim); may he declare with his venerable speech the obliteration of his city, the dispersion of his people, the supplanting of his dynasty, and the blotting out of his name and his memory from the land (šumšu u zikiršu ina mātim la šubšâm). cf. Ea § 4: May the god Ea [. . .] who lengthens the days of my life [. . .].

A more direct correspondence in tenor with Hammurabiʼs curses is found in the prediction of political collapse and military defeat for the Canaanites, as outlined in the texts in table 4. Several of the curses announce the political disempowerment and defeat of the disobedient king and his people. Table 4: Correlations between Canaanite Destruction and Epilogue Curses Exodus 23:20–33 [. . .] I (YHWH) will fight your foes and hound your enemies. 23 [. . .] He (the messenger) will bring you to the Amorite, the Hittite, Perizzite, the Canaanite, the Hivite, the Jebusite, and I will wipe them out. [. . .] 27 I will send my terror before you and put in panic all the people among whom you go and make all your foes flee. 28 I will send 22

11  M. T. Roth, Law Collections from Mesopotamia and Asia Minor (2nd ed.; SBLWAW 6; Atlanta: Scholars Press, 1997), 136–140.

The Covenant Code Appendix and Neo-Assyrian Sources

55

hornets/plague/dread and they/it will drive out the Hivite, Canaanite, and Hittite before you. [. . .] 30 I will expel them from before you gradually, until you have flourished and take possession of the land. 31 I will set your border from the Reed Sea to the Philistine Sea, and from the Desert to the River, for I am delivering the inhabitants of the land into your hands, and you shall expel them from before you. [. . .] 33 They shall not remain in your land so they do not cause you to sin against me – because you will (end up) serving their gods (and) it will be a trap for you.

Curses of the Laws of Hammurabi Epilogue (Rothʼs Translation) Anu § 1: May the great god Anu [. . .] deprive him of the sheen of royalty, smash his scepter, and curse his destiny. Enlil § 2: see table 3 for this (including dispersion of the people) Ninlil § 3: May the goddess Ninlil [. . .] induce the divine king Enlil to pronounce the destruction of his land, the obliteration of his people [. . .]. Shamash § 5: May the god Shamash [. . .] confuse his path and undermine the morale of his army; [. . .] may he (Shamash) provide an inauspicious omen portending the uprooting of the foundations of his kingship and the obliteration of his land; [. . .] may he uproot him from the land of the living [. . .]. Sin § 6: May the god Sîn [. . .] deprive him of the crown and throne of kingship [. . .]. Zababa § 8: May the god Zababa [. . .] smash his weapon upon the field of battle; may he turn day into night for him, and make his enemy triumph over him. Ishtar § 9: May the goddess Ishtar, mistress of battle and warfare [. . .] curse his kingship with her angry heart and great fury; [. . .] may she smash his weapon on the field of war and battle, plunge him into confusion and rebellion, strike down his warriors, drench the earth with their blood, make a heap of the corpses of his soldiers upon the plain, and may she show his soldiers no mercy; as for him, may she deliver him into the hand of his enemies, and may she lead him bound captive to the land of his enemy. Nergal § 10: May the god Nergal [. . .] burn his people with his great overpowering weapon like a raging fire in a reed thicket; may he have him beaten with his mighty weapon, and shatter his limbs like (those of) a clay figure. Nintu § 11: See table 3 for this. The great gods § 13: May the great gods of heaven and earth [. . .] curse that one, his seed, his land, his troops, his people, and his army with a terrible curse.

Significantly, the blessings for the Israelites and threats held out for the Canaanites, taken together, have conceptual correlations with every curse paragraph from the epilogue (i.e., §§ 1–13) save the short final Enlil imprecation (§ 14).12 Another correlation between CCA and the future-king passage is in the call to obedience and the conditional predication of blessing on obedience. The Covenant Code appendix reads: “Take care before him (the messenger) and heed his voice. Do not disregard him, because he will not pardon your disobedience, because my name is in him. If you do heed his voice and do everything that I 12

 This forms a literary envelope with or complement to the initial Enlil curse (§ 2).

56

David P. Wright

say [. . .]” (Exod 23:21–22). After this comes the forecast of destruction for the Canaanites and blessing for the Israelites. A similar construction appears in the future-king passage. It begins with a call to obedience: “In the future at any time, may any king who appears in the land keep the just commands (words) that I have written on my stela [. . .]” (col. 48:59–74). Then the two options are laid out as conditions, introducing blessing or curse: “If that man gives heed to my words that I have written on my stela [. . .]” (49:2–5). A short blessing follows. This is counterbalanced with the provision “if that man does not give heed to the words that I have written on my stela [. . .]” (49:18–22). The curses follow. The larger evidential point is this: CCA fits like a puzzle piece into the larger context of the final apodictic laws in their mapping of the sequence of themes found in the epilogue (again, see the lower right-hand corner of table 2). In its position, CCA carries on the sequential use of the epilogue material from the final apodictic laws (right column of table 2). In fact, CCA at the end of the second string of apodictic laws is precisely parallel, structurally speaking, to the section of laws on justice in 23:1–8 at the end of the first string of final apodictic laws. They articulate materials from the future-king passage of the epilogue in different ways: the laws on justice take up the theme of executing justice from the future-king passage, and CCA takes up the matter of blessings, curses, and obedience from the same passage.

Assyrian Royal Inscriptions and the Covenant Code Appendix Another foreign textual catalyst – Assyrian royal inscriptions – appears to have been influential in the creation of CCA. To serve as a basis for the discussion that follows, I have assembled in appendix 2 to this paper a collection of relevant passages from eight illustrative texts (A–H) from the time of Adad-Nirari II (d. 891 BCE) to Assurbanipal (d. 631 BCE). The beginning of appendix 2 includes an index of motifs (i–vi) that will be discussed in the body of the essay here. Citations that exemplify particular motifs are then given in appendix 2 using the index numbers as they occur in the sample texts. The discussion here will make selective or abbreviated reference to examples. Readers will want to go to appendix 2 to survey the evidence in more detail. Appendix 2 contains full references to the text editions and translations. The relevance of these texts becomes clear when it is realized, as Victor Hurowitz has demonstrated and described in detail, that the general form and genre of LH as a whole is actually that of a royal inscription.13 If the long legal code is bracketed off, the prologue and epilogue constitute a royal inscription  V. A. Hurowitz, Inu Anum ṣīrum: Literary Structures in the Non-Juridical Sections of Codex Hammurabi (Occasional Publications of the Samuel Noah Kramer Fund 15; Philadelphia: University Museum, 1994). 13

The Covenant Code Appendix and Neo-Assyrian Sources

57

that touts the election of the king and his achievements and exhorts readers to protect those achievements. Recognition of this genre correlation, it appears, provided the avenue for CCA to bring in motifs from other royal inscriptions. The particular latch opening the door to the broader genre of royal inscriptions was presumably Hammurabiʼs future-king passage with its blessings and curses, discussed above and cited in appendix 1 to this essay.14 Many royal inscriptions, not just those sampled in appendix 2, have similar passages at their ends. Examples can be found under motif vi in the texts of appendix 2 in texts A, B, E, F, G, and H. These sections counsel a future prince or king to maintain the inscriptional kingʼs prestige and accomplishments, often with blessings for obedience and curses for disobedience. They can be rather extensive, like that in text B from Aššur-Naṣirpal II, and may have a long list of gods, as in text H from Assurbanipal. Because these passages are a rather standard feature of royal inscriptions, it would be a matter of reflex for one familiar with such texts to think of them when working with the similar passage in Hammurabiʼs epilogue. Several thematic correlations are visible between CCA and the royal inscriptions. The primary overall correlation is in the motif of conquest. This is the backbone that runs through CCA, and the theme is similarly pervasive in royal inscriptions. Because of this, and also because most scholars of the Bible and ancient Near East are broadly familiar with royal inscriptions, I have not cited examples of this particular motif in appendix 2. Nevertheless, the theme is visible there incidentally in passages cited to illustrate other motifs. If the future-king passage in LH was the particular gateway to other royal inscriptions, then the topic of conquest seems to have been the particular attraction for going to these other texts. This largely accounts for the shift in genre and theme in Exod 23:20–33. The Covenant Code appendix maintained the general topic of future fortunes, including motifs related to blessing and curse, raised by the future-king passage of Hammurabiʼs epilogue. But instead of focusing on a future individual leader, CCA used the wider genre of royal inscriptions to engage the topic of national conquest. A more specific and conspicuous correlation between CCA and royal inscriptions appears in the motif of the gods or the godsʼ symbols going before the king or the army in the campaign, the vanguard motif.15 This is motif ii in the sample corpus and appears in texts A, B, C, D, and H. The Akkadian texts use the verb 14  For the pattern, see C. M eltzer , “Concluding Formulae in Ancient Mesopotamian Royal Inscriptions: The Assyrian Sources” (PhD diss., Toronto University, 1983), 141–145, as discussed in I. Zsolnay, “The Function of Ištar in the Assyrian Royal Inscriptions” (PhD diss., Brandeis University, 2009), 88–106. Zsolnayʼs discussion of this motif and the vanguard motif (discussed below) were influential in my seeing the relevance of royal inscriptions for CCA. 15  For this divine escort or vanguard motif, see V. A. H urowitz and J. G. W estenholz , “LKA 63: A Heroic Poem in Celebration of Tiglath-Pileser Iʼs Murṣu-Qumanu Campaign,” JAOS 42 (1990), 1–49, here 30–34; T. M ann , Divine Presence and Guidance in Israelite Tradition: The Typology of Exaltation (Eugene, OR: Wipf & Stock, 1977). See the idioms ālik

58

David P. Wright

alāku and preposition (ina) panāt-/maḫr-/IGI. For example, an inscription of Shalmaneser III (text C) claims: “With the exalted might of the divine standard which goes before me (a-lik IGI-ia) (and) with the fierce weapons which Aššur my lord gave to me, I fought (and) defeated them.” The Letter to Aššur for the Eighth Campaign of Sargon II (text D) describes the advance with the words “Against Zikirtu and Andia I guided the yoke of the standard wagon of Nergal and Adad, who go before me” (a-li-kut maḫ-ri-a). Assurbanipalʼs Rassam Decagon (text H) describes the attack: “Assur, Sin, Shamash, Adad, Bêl, Nabû, Ishtar of Nineveh, the queen of Kidmuri, Ishtar of Arbela, Urta, Nergal (and) Nusku, who march before me (ša ina maḫ-ri-ia il-li-ku), slaying my foes, cast Shamash-shum-ukîn, my hostile brother, [. . .] into the burning flames [. . .] and destroyed him.” Variants of the motif include the description of the god going or standing at the side as in texts F and G. For example, an inscription from Esarhaddon (text F) says, “The goddess Ištar, the lady of war and battle, who loves my priestly duties, stood at my side (i-da-a-a ta-zi-iz-ma), broke their bows, (and) she split open their tight battle ranks.” The vanguard motif occurs emphatically four times in CCA. Verse 23 says, “Indeed, my messenger will go before you (‫אָכי ְל ָפנֶ יָך‬ ִ ‫)כּי־יֵ ֵלְך ַמ ְל‬, ִ and bring you to the Amorite, the Hittite, the Perizzite, the Canaanite, the Hivite, the Jebusite, and I will wipe them out.” The verse uses the verb go (*‫ )הלך‬and the preposition before (‫)ל ְפנֵ י‬, ִ similar to the Akkadian expression noted above. The other three cases of the vanguard motif use the verb send (*‫ )שלח‬with the preposition before (‫)ל ְפנֵ י‬ ִ plus suffix pronoun. This occurs in the first verse of CCA: “I am herewith sending a messenger before you” (‫הנֵּ ה אָנ ִֹכי שׁ ֵֹל ַח ַמ ְלאְָך ְל ָפנֶ יָך‬, ִ v. 20). In the other two cases, further into the passage (vv. 27–28), the messenger is bifurcated into a pair of negative entities: “I will send my terror before you (‫ימ ִתי ֲא ַשׁ ַלּח‬ ָ ‫ת־א‬ ֵ ‫ֶא‬ ‫)ל ָפנֶ יָך‬ ְ and put in panic all the people among whom you go” and “I will send hornets/plague/dread (‫ת־ה ִצּ ְר ָעה ְל ָפנֶ יָך‬ ַ ‫ )וְ ָשׁ ַל ְח ִתּי ֶא‬before you and they/it will expel the Hivite, Canaanite, and Hittite before you.” The negative avatars16 and the verbal effects they have on the enemy, especially ‫ימה‬ ָ ‫א‬, ֵ whose meaning “terror”17 is clear, along with its effect of putting the enemies into panic (‫וְ ַהמּ ִֹתי‬, root *‫ )המם‬and making them flee (‫ת־כּל־‬ ָ ‫וְ נָ ַת ִתּי ֶא‬ maḫri “herald, forerunner” CAD A/1 344 (of gods and divine emblems) and ālik pani “leader, superior” CAD A/1 344–345 (sometimes of gods). See also CAD P 80 (s.v. panāt ummāmi). 16  I draw this term from the insightful discussion of conceptualization of the deity in biblical texts by B. D. S ommer , The Bodies of God and the World of Ancient Israel (Cambridge: Cambridge University Press, 2009), 78 and 232, n. 113. 17  ‫ימה‬ ָ ‫ ֵא‬is paired with ‫ ַפ ַחד‬in Exod 15:16. The meaning “fear, terror” is clear from Josh 2:9. The terror “falls” also in Gen 15:12; Ps 55:5 (in this last case it is the “terrors of death”). It verbally terrifies in Job 9:34; cf. 13:21; 33:7. It is paired with sword in Deut 32:35 and royal terror is compared to a lionʼs roar in Prov 20:2. See also Isa 33:18; Jer 50:38; Ps 88:16; Job 20:25; 39:20; 41:6; Ezra 3:3.

The Covenant Code Appendix and Neo-Assyrian Sources

59

‫ֹיְביָך ֵא ֶליָך ע ֶֹרף‬ ֶ ‫א‬, v. 27), correlate with motifs in the royal inscriptions. Several inscriptions describe enemies being overwhelmed by the puluḫtu “terror” or melammu “splendor” of the gods or kings (motif iiia in appendix 2; for puluḫtu, see texts D, F; for melammu, see texts B, H).18 Synonyms or parallel terms may be used (ḫātu/ḫattu “terror” in text D and ḫattu puluḫtu “terror and fear” in text E; namrīru “splendor” of the deity in text H; rašubbatu “frightening appearance” of a divine weapon in text H; cf. the namurratu “splendor” of the king in text G or of weapons in texts C, H). Sometimes the terms appear in combination, such as pulḫe melamme or puluḫti melamme “fearful splendor” (texts C, D, F).19 In addition, this fear or splendor may be attributed to the gods or to the king (gods: B, C, F, H; king: B, D, H, cf. G), and when attributed to the king, it may be contextualized with first-person pronouns, e.g., “the fear of my splendor” or “the splendor of my kingship/lordship” (texts B, D, H). The main verbs that describe being overwhelmed by this fear or splendor are saḫāpu “overwhelm,” katāmu “cover,” and tabāku “pour out” (texts B, C, D, E, F, H).20 The term ‫ימ ִתי‬ ָ ‫א‬, ֵ “my terror,” in the CCA matches closely puluḫtu “fear, terror” in the inscriptions, especially when the latter is attributed to the deity or king.21 A correlation with royal puluḫtu or melammu is consistent with the hermeneutic of CC proper, particularly as manifested in the apodictic laws, where various motifs relating to the human royal lawgiver (as well as the Mesopotamian gods) were transferred to YHWH. The “messenger” or “my messenger” as the divine avatar is novel in the CCA in contrast to the inscriptions, but it is functionally equivalent broadly to the melammu “splendor” of the deity or king. Concomitant native tradition and idiom may partly account for the appearance of the messenger motif in particular. For this, see the end of the conclusion of this paper. The second negative avatar in CCA, the ‫צ ְר ָעה‬, ִ does not have a clear correlate in the royal inscriptions.22 It is often rendered “hornet” based on the versions (and 18  For a recent extensive examination of melammu and puluḫtu and biblical counterparts, see S. Z. Aster, The Unbeatable Light: Melammu and Its Biblical Parallels (AOAT 384; Münster: Ugarit-Verlag, 2012). His main discussion of the Akkadian terminology is on pp. 22–121. 19  Aster, Unbeatable Light (see n. 18), 84; idem, “The Image of Assyria in Isaiah 2:5–22: The Campaign Motif Reversed,” JAOS 127 (2007), 249–278, here 274. He observes that the individual occurrence of puluḫtu is just shorthand for puluḫti melamme or pulḫe melamme. 20  See Aster, Unbeatable Light (see n. 18), 45–52; idem, “Image of Assyria” (see n. 19), 274–277, on verbs for “overwhelming” associated with puluḫti melamme. 21  Greek has τὸν φόβον for MTʼs ‫ימ ִתי‬ ָ ‫ ֵא‬in Exod 23:27, without the first-person possessive pronoun. A reverse variation occurs in 23:20, where the Greek has τὸν ἄγγελόν μου (similarly Samaritan and Vulgate) against MTʼs unmodified ‫מ ְלאְָך‬. ַ In v. 23 both MT and Greek have the pronoun ‫אָכי‬ ִ ‫מ ְל‬/ὁ ַ ἄγγελός μου. It is possible that the pronoun in v. 27 is secondary, by attraction to a more original formulation with the pronoun in v. 23. Nevertheless, that ‫ַה ִצּ ְר ָעה‬ in v. 28, parallel to v. 27, lacks the pronoun is not necessarily proof that the pronoun in v. 27 is secondary. 22  It is tempting to connect ‫ ִצ ְר ָעה‬with the Akkadian noun ṣiriḫtu “anger” or the verb ṣarāḫu A (N-stem) “be angry,” which are used in royal inscriptions to describe the kingʼs animosity

60

David P. Wright

cf. Deut 1:44), though etymologically the term may refer to plague (cf. ‫)צ ַר ַעת‬, ָ and in the literary context it is parallel to ‫ימה‬ ָ ‫א‬. ֵ 23 The use of the term ‫צ ְר ָעה‬, ִ may come not from source stimuli but from elaborate paronomasia. Both ‫ימה‬ ָ ‫ ֵא‬and ‫צ ְר ָעה‬, ִ echo the messenger motif and descriptions of the deityʼs acts of conquest earlier in the passage. In the following citation (as opposed to some of my more straightforward translations of the same verses elsewhere in this paper), I have creatively worded the English translation to draw attention to the wordplay in Hebrew:24 I am herewith sending a messenger before you. [. . .] 21 Take care before him and heed his voice. Do not disregard him. [. . .] 22 If you do heed his voice and do everything that I say, I will fight your foes (‫אָיַב ִתּי ֶאת־א ֶֹיְביָך‬ ְ ְ‫)ו‬ and I will distress your discomforters (‫)וְ ַצ ְר ִתּי ֶאת־צ ְֹר ֶריָך‬ 23 Indeed, my messenger will go before you. [. . .] 27 I will send my fear (‫ימ ִתי‬ ָ ‫)א‬ ֵ before you and cause frenzy among the folk you find (‫ל־ה ָעם ֲא ֶשׁר ָתּבֹא ָבּ ֶהם‬ ָ ‫ת־כּ‬ ָ ‫)וְ ַהמּ ִֹתי ֶא‬, and make all your foes flee (‫)א ֶֹיְביָך ֵא ֶליָך ע ֶֹרף‬. 28 I will send dread (‫)ה ִצּ ְר ָעה‬ ַ before you and it/he will drive out (‫ )וְ גֵ ְר ָשׁה‬the Hivite, Canaanite, and Hittite before you. 20

According to v. 22, the deity will do two things: fight the peopleʼs enemies (‫אָיַב ִתּי ֶאת־א ֶֹיְביָך‬ ְ ְ‫ו‬, v. 22bα) and harass their adversaries (‫וְ ַצ ְר ִתּי ֶאת־צ ְֹר ֶריָך‬, v. 22bβ). He performs these acts in the context and by virtue of sending his messenger (vv. 20, 23). The two negative avatars are described later, and they correlate alliteratively with the respective descriptions of divine aggression in v. 22. The term ‫ימה‬ ָ ‫ ֵא‬along with other language in v. 27 (‫] ָתּב ֹא ָבּ ֶהם‬. . .[ ‫] ָה ָעם‬. . .[ ‫וְ ַהמּ ִֹתי‬ ְ ְ‫ו‬ ‫] ע ֶֹרף‬. . .[ ‫] א ֶֹיְביָך‬. . .[) echo the description of divine aggression in v. 22bα (‫אָיַב ִתּי‬ toward his enemies (motif iiib in appendix 2 in texts F and H). But this is not extensively attested and does not appear in the immediate contexts of melammu or puluḫtu. A connection to Akkadian ṣirḫu “flare, flash” and ṣarāḫu C “light up” is attractive because the terms look like they relate conceptually to melammu. But the terms do not occur in royal inscriptions and refer to astronomical phenomena and often occur in omens. 23  See, for example, W. H. C. Propp, Exodus 19–40 (AB 2A; New York: Doubleday, 2006), 290; T. B. Dozeman, Exodus (ECC; Grand Rapids, MI: Eerdmans, 2009), 557. Both take the term as “hornets.” 24  The citation here draws attention to a quasi-poetic form that appears throughout the passage. There are a number of bi- or tricola (depending how one analyzes them) that are primarily hung on the vanguard or avatar lines. The art of this construction can be compared to that of the string and chiastic core structure of the final apodictic laws of CC. But CCA does not have the same sort of structuring as the final apodictic laws.

The Covenant Code Appendix and Neo-Assyrian Sources

61

‫ֹיְביָך‬ ֶ ‫את־א‬, ֶ glottals, labials, semivowel /y/). The term ‫צ ְר ָעה‬, ִ and possibly also the verb ‫וְ גֵ ְר ָשׁה‬,25 in v. 28 echoes the description of divine aggression in v. 22bβ (‫וְ ַצ ְר ִתּי ֶאת־צ ְֹר ֶריָך‬, alveolar emphatic fricatives/affricates, approximant /r/). Thus ‫ימה‬ ָ ‫ ֵא‬and ‫צ ְר ָעה‬, ִ from this literary-structural point of view, are specifications or attributes of the ‫מ ְלאְָך‬, ַ all of whom are sent by the deity – avatars of an avatar.26 The Covenant Code appendix shares other less frequently occurring but still notable motifs with the royal inscriptions. The biblical passage commands the Israelites to destroy the Canaanitesʼ gods and smash their pillars (v. 24). Texts E and H describe the smashing of enemy gods and the tearing down of temples. These texts use the verb šubburu, cognate with the Hebrew ‫שׁ ֵּבר‬. ִ For example, Sennacherib (text E) notes, “My men took the (images of the) gods who dwell there and smashed them (ú-šab-bi-ru-ma) [. . .]. I destroyed and tore down and burned with fire the city (and) [. . .] I tore out the inner and outer walls, temples, the ziggurat of brick and earth [. . .].” Assurbanipal (text H) says, “I struck down the people living (in Bashimu). I smashed (ú-šab-bir) their gods and pacified the divine heart of the lord of lords” and “The zikkurat of Susa, which was built of enameled bricks, I destroyed [. . .]. The sanctuaries of Elam I destroyed totally (lit., to nonexistence). Its gods (and) goddesses I turned into ghosts.”27 The Covenant Code appendixʼs list of enemies in v. 23 (abbreviated in v. 28) has parallels in lists of enemies in some of the inscriptions. Text C from Shalmaneser III (motif v) has a list of kings and/or cities identified, many with gentilic adjectives, including Hama, Israel (i.e., ma-ḫa-ab-bu KUR sir-ʼa-la-a-a), Byblos, Egypt, Irqanatu, Arvad, Usanātu, Šianu, Arabia, and Ammon. Lists of conquered cities appear in various royal inscriptions, along with more-extended accounts of conquests of successive cities or locales. Their goal is to describe the extensiveness of acquisition and conquest, and that is a function of the list in 25  For divine expulsion with *‫ גרש‬in a Northwest Semitic text, see Mesha (KAI 191) line 19. This is a small piece of evidence pointing to the type of traditional local idiom that may be used in the passage. See the conclusion toward the end. 26  One may wonder if this paronomasia marks the original boundaries of CCA, consisting of only vv. 20–28 and focusing on the avatars of the deity and vanguard motif. The following verses, about delayed expulsion (vv. 29–30), borders (v. 31), and the covenant and foreigners as a snare (vv. 32–33), have fewer correlations with Hammurabiʼs epilogue or royal inscriptions. But note that the rationale for the nations staying in the land is different and apparently predates that which is presented in Judg 2:1–5 (see n. 48, below). For issues of redactional development, see n. 34. 27  For the related issue of “godknapping,” see A. Berlejung, “The Assyrians in the West: Assyrianization, Colonialism, Indifference, or Development Policy?” in Congress Volume Helsinki, 2010 (ed. M. Nissinen; VTSup 148; Leiden: Brill, 2012), 21–60, here 33–34. On p. 32 she notes, “There is no proof of any Assyrian interest in uprooting non-Assyrian gods.” This is part of her main argument, as she summarizes on p. 51: “The only goal (and success parameter) of Assyrian imperial policy was an Assyro-centric economic policy, not assyrianization. [. . .] Assyria did not want your belief or identity. The slogan was: Believe whatever you want, but you will never escape death and taxes!”

62

David P. Wright

CCA. For the list in CCA, as in the case of the term messenger, native tradition or wording appears to have been brought in to articulate a theme suggested by the external source (see the end of the conclusion).28 The Covenant Code appendix has some correlations with the future-prince or future-king passages of the inscriptions (see motifs under vi).29 Like Hammu­ rabiʼs epilogue, these sections contain short blessings and longer curses, though the curses in our sample texts are not as extensive as in LH. Text B provides a rather full exposition of the themes. The blessings are as general as the godsʼ hearing the prayers of the future king (texts A, F) but also contain specific blessings of political security (text B), military success (B), and abundance (B). Curses may be generally outlined (H) or deal with specifics such as political failure (A, B, E, F), military conflict and defeat (B, G), extinction of posterity (A), and famine (B). These blessings and curses broadly relate to the blessings promised the Israelites and the defeat held out for the Canaanites in CCA. The kingʼs name (A, B, F, G) and to a lesser extent the godsʼ names (E, F) are prominent in the future-leader passages. The inscriptional kingʼs name is to be maintained on his monument or, if missing or effaced, restored. Text B provides an extensive example that shows the connection of the themes of maintaining the inscription with its name and the necessity of obedience. It prefaces the section with a call to restore the inscriptional kingʼs name: “May a later prince restore its weakened (portions and) restore my inscribed name to its place. (Then) Aššur will listen to his prayers” (v. 24–26). It then talks about the faithful king who “does not alter the ordinances of my text” (v 47), followed by a blessing of abundance. Next it describes and condemns in some detail an unfaithful future leader who destroys the monument or shows it disrespect (see v 55–90 in text B of appendix 2), followed by curses of political destruction, distress, and famine. This section features the motif of obedience, which is connected to or manifested in the preservation of the inscription. This is metonymically represented or en28  For some discussion of the lists of nations in biblical texts, see K. Schmid, Genesis and the Moses Story (Siphrut 3; Winona Lake, IN: Eisenbrauns, 2010), 226–227; J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven, CT: Yale University Press, 2012), 139–140. If a documentary analysis is to be preferred, one of the questions to ask is if E is earlier than J and if its lists influenced Jʼs. See the conclusion. 29  A possible correlation between CCA and the future-leader passages, in addition to those in the main discussion, is the motif of treaty-like obligations. Sennacheribʼs text E warns, as Luckenbill translates (see appendix 2), “If ever there is a future prince among the kings, my sons, who destroys the work which I have done, (and) breaks the covenant I have (hereby) made with him (rik-sa-te ar-ku-su i-paṭ-ṭa-ru) [. . .].” CAD P 292 § 4 renders this similarly but more in line with the broader theme of obedience: “whoever breaks the rulings I have made.” This could correlate inversely and broadly with CCAʼs prohibition of making a treaty with the Canaanites (‫יהם ְבּ ִרית‬ ֶ ‫אֹלה‬ ֵ ‫א־ת ְכר ֹת ָל ֶהם וְ ֵל‬ ִ ֹ ‫ ;ל‬v. 32). The problem is that the interpretation of the phrase in Sennacheribʼs text is not without dispute. CAD R 248 § 2a renders it “whoever dismantles the (canal) system I have constructed.”

The Covenant Code Appendix and Neo-Assyrian Sources

63

tailed by the preservation of the royal name. To preserve the inscriptional kingʼs name is to preserve his inscription, which provides a model and regulations for behavior. The nexus of name and obedience is found in CCA. After the initial announcement of the attending messenger, the passage warns: “Take care before him and heed his voice [. . .], because my name is in him (‫”)כּי ְשׁ ִמי ְבּ ִק ְרבֹּו‬ ִ (Exod 23:21). When we remember that CC hermeneutically replaces the human Mesopotamian king with YHWH, this rationalization for obedience in CCA correlates with the interest in the royal name in royal inscriptions. This verse, in fact, continues the articulation of the theme of the divine name that is prominent in CC proper (Exod 20:24; 22:27; 23:13) and which, as noted earlier, is dependent on Hammurabiʼs epilogue. In the first of these instances YHWH announces: “In every place where I (or emended: you) recall my name” (‫ל־ה ָמּקֹום ֲא ֶשׁר אַזְ ִכּיר‬ ַ ‫ְבּ ָכ‬ ‫וּב ַר ְכ ִתּיָך‬ ֵ ‫ת־שׁ ִמי אָבֹוא ֵא ֶליָך‬ ְ ‫א‬, ֶ 20:24). This contains the terminology and conception in the call to memorialize Hammurabiʼs name: “in the Esagil, which I love, may my name be recalled kindly forever” (ina Esagil ša arammu šumī ina damiqtim ana dār lizzakir). The correlating features include the verb zkr, cognate words for “name” with the first-person suffix attached, and a cult place as the locus of memorialization. Later, as a type of measure-for-measure punishment (a motif explicit in text A of appendix 2), Hammurabiʼs epilogue threatens a disobedient king with name destruction in one of its first curses: “Let him (Enlil) pronounce with his revered mouth the disappearance of his (the disobedient kingʼs) name and memory from the land” (šumšu u zikiršu ina mātim lā šubšâm ina pīšu kabtim liqbi, 47:76–80). Hammurabiʼs text portrays the display of obedience through the maintenance of the inscription and memorialization of Hammurabiʼs name. It may not be a coincidence that CCʼs internal injunction to obedience in 23:13a, which correlates with the call to obedience for a future king (col. 48:59–74; 49:2–5, 18–22; see the discussion of Hammurabiʼs epilogue and CCA, above), accompanies the last of the three parallel passages dealing with speaking about deities and their names in CC (23:13b). Thus, name and obedience are interrelated in both CC proper and CCA, and these resonate with the connection of the royal name and obligations to maintain the inscriptional kingʼs stela and injunctions.30

30  L. S chwienhorst-S chönberger , Das Bundesbuch (BZAW 188; Berlin: de Gruyter, 1990), 410, notes correlations between CCA and CCʼs initial altar law: name, blessing, and “place” (23:20, 21, 25; see 20:24). The laws on images thematically correlate (23:24; cf. 20:23) as well.

64

David P. Wright

Conclusions and Ramifications Here I will draw some conclusions based on the previous analysis and then explore implications for pentateuchal research. In my original study of CC I was perplexed about why CC abandoned its use of the epilogue halfway through, given its apparent preoccupation with this part of Hammurabiʼs text. The furthest correlation that I saw was between the chiastic passage on justice in 23:1–8 and exhortations that the future king follow Hammurabiʼs example of justice (see table 2, bottom of the column, second from right). Though CC alluded to blessing (e.g., 20:24) and curse (e.g., 22:22–23), themes that follow in Hammurabiʼs future-king passage, it did not develop these. It was particularly odd that CC ignored the curses, which are so prominent and ripe for ideological recasting. The correlations between CCA and the epilogue curses, along with themes in the partner genre of royal inscriptions, now resolve this conundrum. The rest of the epilogue was in fact used, to help craft CCA. The forward-looking orientation of the future-king passage facilitated the shift in perspective to future historical events. Hammurabiʼs theme of blessing in the context of obedience, along with motifs from Hammurabiʼs curses by means of conceptual inversion, influenced the description of blessings for the Israelites. Aspects of the curses, unconverted, also helped paint the picture of Canaanite disempowerment. In addition, that LH as a whole was cast as a royal inscription opened the door for using motifs from other, later royal inscriptions in building CCA, including the conquest, the deity or his avatar at the head of attack, and overwhelming divine terror. What is particularly compelling about this evidence is that CCA fits neatly into the mosaic of correlations (see table 2, lower right corner). Not only does it vertically finish the sequence of themes from the epilogue following from the end of the final apodictic laws, but structurally and (in terms of source) thematically it horizontally complements the passage on justice (23:1–8) as an alternative reflection and extension of the future-king passage. That is to say, Hammurabiʼs future-king passage entailed two possible roads for thematic development: the formulation of injunctions based on admonitions and exhortations given to the future king and the description of national future prospects based on descriptions of how a royal successor might react and what consequences both he and his people might face. The chiastic core of laws on justice in 23:1–8 provided the injunctive realization of the future-king passage, and the appendix in 23:20–33 provided the prospective realization. It seems difficult to attribute this vertical and horizontal patterning, which correlates with the thematic structure of an epilogue, along with themes from kindred royal inscriptions, to independent development, especially when the body of CCʼs casuistic laws clearly appear to rely on LH for content and struc-

The Covenant Code Appendix and Neo-Assyrian Sources

65

ture. The whole web of correlations between CC and CCA work together to make this hypothesis about textual dependence reasonable. This analysis also explains compellingly why CCA is where it is, after the laws but before the end of the revelation.31 It accounts for the genre change found in CCA. It also helps explain how CCA relates thematically and conceptually to CC. Even though CCA does not make any explicit reference to CC, the creators of CC and CCA may have seen the promises held out in CCA as dependent upon the maintenance of the preceding law, just as the blessings for an obedient future king are dependent on his maintenance of Hammurabiʼs model of justice. Support for this generative understanding of CCA is found in recent studies that have recognized the influence of Assyrian royal inscriptions on other biblical texts, such as preexilic passages of Isaiah.32 In several cases, motifs from the inscriptions were rearticulated in terms of YHWHʼs power in ideological counterstatement to Assyrian hegemony. For example, Aster has argued that Isa 2:5–22 reflects a description of an annual military campaign as found in royal inscriptions. The two elements in the phrase ‫וּמ ֲה ַדר גְּ אֹנֹו‬ ֵ ‫מ ְּפנֵ י ַפּ ַחד יְ הוָ ה‬, ִ “from the fear of YHWH and his powerful majesty,” in 2:10, 19, 21 are calques on the inscriptional terms puluḫtu “fear” and melammu “splendor.” The reflection of these motifs and their attribution to YHWH is precisely what I have proposed 31  Various studies have sought to explain the textual location of CCA. Some have seen CCA as a type of epilogue to the laws, with some observing specifically that it is similar to blessing and curse sections of law collections, reflecting on the question of the future generated by observing the laws or anticipating the covenant ceremony to come in 24:3–8. See Wright, Inventing (see n. 1), 503, n. 93; Propp, Exodus 19–40 (see n. 23), 287; C. Houtman, Exodus, Volume 3: Exodus 20–40 (HCOT; Leuven: Peeters, 2000), 270; W. Johnstone, “Reactivating the Chronicles Analogy in Pentateuchal Studies, with Special Reference to the Sinai Pericope in Exodus,” ZAW 99 (1987), 16–37, here 25–26; H. A usloos , “The ‘Angel of YHWH’ in Exod. xxiii 20–23 and Judg. ii 1–5: A Clue to the ‘Deuteronom(ist)ic’ Puzzle?” VT 58 (2008), 1–12, here 2–3. Several have noted, however, that it is out of place and really does not act like an epilogue (E. Blum, Studien zur Komposition des Pentateuch [BZAW 189; Berlin: de Gruyter, 1990], 375–376; B. S. Childs, The Book of Exodus [OTL; Philadelphia: Westminster, 1974], 461, 486). See n. 50. 32  See P. M achinist , “Assyria and Its Image in the First Isaiah,” JAOS 103 (1983), 719–737; A ster , “Image of Assyria” (see n. 19); idem , The Unbeatable Light (see n. 18), 204–256; B erlejung , “Assyrians in the West” (see n. 27), 31–33; H. L iss , “Undisclosed Speech: Patterns of Communication in the Book of Isaiah,” JHS 4 (2002), http://www. jhsonline.org/cocoon/JHS/a026.html (accessed 2/24/15); W. R. G allagher , Sennacheribʼs Campaign to Judah (SHCANE 18; Leiden: Brill, 1999), 22–90. See the other literature in Berlejung, “Assyrians in the West,” 32–33, n. 35. She notes that “[b]iblical literature reflects the fact that the elites in Jerusalem were well informed about Assyrian royal propaganda and their worldview” (31). Her discussion of the function of the qīpu (responsible for monitoring local treaties; 28–29, 32) points to one possible institutional route for learning about some Assyrian and Mesopotamian ideas.

66

David P. Wright

for CCA.33 The various examples of influence from Assyrian royal inscriptions in these other biblical texts demonstrate familiarity with the foreign powerʼs ideological descriptions. They seek to counter it by conceptual inversion and appropriation. More will be said about implications that the royal inscriptions have for dating CC and its goals as a piece of legal and narrative literature momentarily, but first it should be observed that the analysis of how CCA relates to the larger structure inspired by LH points to the compositional integrity of CCA in the production of CC. I would not assert that one hand is responsible for all of CC and CCA. The complex use of sources may point to a multiplicity of contributors. But it does seem that CC and its appendix are part of a single compositional plan and process. It does not make sense to argue that the compositors of CCʼs final apodictic laws stopped halfway through using the epilogue and that a generation or more later a different school added CCA, still using LH as a template. The elaborateness of the final apodictic laws, with their two strings placed around a chiastic core and their thirty verses versus the four of the initial apodictic laws, belies this sort of redactional assessment. The length and complexity of the final apodictic laws seem to have arisen in part from a plan to manifest the injunctive and future prospective themes in Hammurabiʼs future-king passage in separate passages, 23:1–8 and 23:20–33.34 33  A ster , Unbeatable Light (see n. 18), 204–224; idem , “Image of Assyria” (see n. 19), 272–278; he posits that in Isa 2:10, 19, 21, ‫ ַפּ ַחד יְ הוָ ה‬is a calque on Akkadian puluḫtu and that ‫ ֲה ַדר גְּ אֹנֹו‬is a calque on melam belūti where the first noun parallels melammu. 34  The source analysis outlined in this essay also challenges some standard redactional analyses of CCA. Analyses generally separate out words and phrases with second-person plural formulation from those with the second-person singular. Moreover, the two negative avatars of vv. 27–28 and vv. 29–30 about gradual expulsion are seen as secondary. For ex­ ample, Y. Osumi, Die Kompositionsgeschichte des Bundesbuches Exodus 20,22b–23,33 (OBO 105; Fribourg: Editions Universitaires; Göttingen: Vandenhoeck & Ruprecht, 1991), 212–215, posits an original text of 23:20–21a, 21bβ–22, 23a, 24, 32, 33bβ. The problem for such an analysis is that it omits material and motifs that appear to be generated by the sources (the LH epilogue or royal inscriptions), including the blessings of vv. 25–26 and the vanguard of terror in v. 27. Most of vv. 20–29 appear tied to sources: the theme of conquest; the vanguard motif, in vv. 20, 23, 27, 28; obedience in vv. 21–22; the name in v. 21; the “terror” avatar of v. 27 and with this the paronomasia between vv. 22 and 27–28; the destruction of foreign gods in v. 24; and the blessings in vv. 25–26. This would indicate that such criteria as the second-person plural phrases (e.g., v. 25) or duplication of motifs (e.g., multiple avatars) are not decisive for redactional analysis. (For problems with the second-person plural forms as a criterion for redactional analysis in CC proper in connection with sources used, see Wright, Inventing [see n. 1], 322–332.) Shifts in style, repetition, and other inconsistencies may well be ascribed to working with different sources, or parts of sources, and blending them. For example, that the blessings in vv. 25–26 mainly correlate with the curses of Hammurabiʼs epilogue as opposed to material in surrounding verses that have correlations more with royal inscriptions may help explain the change to second-person plural in v. 25. For compositional complexity arising from the use of multiple sources, see Wright, “Source Dependence” (see n. 7). See also n. 26,

The Covenant Code Appendix and Neo-Assyrian Sources

67

Several of the foregoing considerations bear on the dating of CC along with CCA. The Covenant Code proper still fits best in the window that I originally proposed, i.e., between 740 and 640 BCE. The recently discovered law fragments from Hazor, though tantalizing for the reconstruction of the history of biblical law and law texts, do not change the picture, because they do not provide parallels to CC or LH. The terminus post quem of 740 BCE is based on the fact that this was about the time that the Neo-Assyrian Empire began to exert strong and mostly continuous influence over Israel and Judah. It is also in the Neo-Assyrian period that we have the most copies of LH, next to the OB period. The terminus ante quem of 640 BCE is based mainly on the fact that Dʼs basic laws use CC as a source. These D laws date, most reasonably, to the mid- to later seventh century BCE. If CC and CCA are part of a larger narrative about the enslavement and liberation of the Israelites from Egypt (discussed below), and if the story of Mosesʼs birth, which seems to be based on the legend of the birth of Sargon of Agade that was composed or resuscitated in the time of Sargon II, then the terminus post quem may be around 700 BCE.35 A major consideration in dating CC to 740–640 and perhaps more narrowly to 700–640 are the affinities between CC and D. These indicate that CCʼs composition may not have been more than a generation or so before Deuteronomy, and in certain respects CC is proto-Deuteronomic. Deuteronomy picks up several motifs from CC and develops these to greater prominence, such that some have argued some of these motifs are Deuteronomistic additions to CC.36 Continuity in scribal culture is also seen in Deuteronomyʼs recasting of parts of a Mesopotamian royal administrative and ideological text, Vassal Treaties of Esarhaddon (672 BCE), in Deut 13 and 28 as law revealed by YHWH and with YHWH taking the place of the Assyrian king.37 This makes sense in a context above. For other observations on redactional development, see Schwienhorst-Schönberger, Bundesbuch (see n. 30), 407–410 (he says that the second-person singular materials look like they belong to his “gottesrechtliche Redaktion” of CC); cf. W right , Inventing (see n. 1), 18–19; Blum, Studien (see n. 31), 375–376 n. 61; Houtman, Exodus (see n. 31), 273. 35  See Wright, Inventing (see n. 1), 342–345, on the Sargon tale. 36  These motifs include those such as the trio of the immigrant, orphan, and widow; “release” with the verb *‫ ;שמט‬the connection of the deityʼs name with the cult “place” as the location where the deityʼs name is manifested; and the holiness of people. I have noted elsewhere that a majority of attestation does not necessarily demonstrate chronological priority, especially when one text uses another as a source. See D. P. Wright, “Profane Versus Sacrificial Slaughter: The Priestly Recasting of the Yahwist Flood Story,” in Current Issues in Priestly and Related Literature: The Legacy of Jacob Milgrom and Beyond (ed. R. Gane and A. Taggar-Cohen; SBLRBS; Atlanta: Society of Biblical Literature, 2015), 125–154, here 146, n. 45. 37  A recent copy of VTE from ancient Syria was found at Tell-Tayinat (now SE Turkey), which shows that the text was known in the west. This copy was put on display at the temple in Tell-Tayinat, and it has been argued that a copy was similarly put on display in the Jerusalem temple. See J. L auinger , “Esarhaddonʼs Succession Treaty at Tell Tayinat: Text and Com-

68

David P. Wright

where Dʼs scribes knew and continued the strategy employed by CCʼs scribes in recasting other Mesopotamian texts that are expressive of and paradigmatic for royal power and administration. Both D and CC are preoccupied with crafting new legal expressions that assert Israelite or, as specifically seems to be the case, Judean political identity and therefore presumably both arise in the same general sociopolitical and historical context.38 The evidence of the use of Assyrian royal inscriptions in CCA as a component in the CC complex now definitively dates the whole to the Neo-Assyrian period. As noted above, other biblical texts from this period, which is also around the time of Hezekiah, reflect knowledge of and recontextualize motifs from Assyrian royal inscriptions in order to articulate Judean distinctiveness and power in their own way. The Covenant Code and CCA are part of – and one of the early parts of – this general late-eighth- and seventh-century response to Neo-Assyrian power. This analysis indicates that CCA is not a late addition to CCʼs context. Though it may contain some redactional supplementation,39 as part of the CC complex it has to be viewed as pre-Deuteronomic. The expressions that it contains that are similar to Deuteronom(ist)ic expressions should be considered part of the material that characterizes the complex as proto-Deuteronomic, a detail that ties CC to the period not long before the creation of Deuteronomy, as noted above.40 mentary,” JCS 64 (2012), 87–123. For Dʼs dependence, see B. M. Levinson and J. Stackert, “The Limits of ‘Resonance’: A Response to Joshua Berman on Historical and Comparative Method,” JAJ 4 (2013), 310–333; idem , “Between the Covenant Code and Esarhaddonʼs Succession Treaty: Deuteronomy 13 and the Composition of Deuteronomy,” JAJ 3 (2012), 123–140. 38  For interesting observations about how such literature functions in terms of analogical application of perspectives from postcolonial theory, see W. M orrow , “Resistance and Hybridity in Late Bronze Age Canaan,” RB 115 (2008), 321–339; idem, “ ‘To Set the Name’ in the Deuteronomic Centralization Formula: A Case of Cultural Hybridity,” JSS 55 (2010), 365–383; W right , Inventing (see n. 1), 350–351. If the story of Mosesʼs birth in Exod 2 models the birth legend of Sargon of Akkad (as many argue) and if this belongs to a CC narrative, then this narrative reflects material and influence from three different paradigmatic representations of Mesopotamian royal ideology: Hammurabi and his law code, Sargon of Akkad and the story of his divine election, and royal inscriptions and their model of conquest. The CC narrative used this material to create a picture of Israelite origins that subversively mimics or flatteringly emulates Mesopotamian ideals. 39  See n. 34, above. 40  Hence CCA is not a Dtr (or post-Dtr) addition (see the summary of views in Ausloos, “Angel” [see n. 31], 2–6 and nn. 4, 22, 28 there; also see R. G. Kratz, The Composition of the Narrative Books of the Old Testament [London: T&T Clark, 2005], 142–143 [he says the passage presupposes the events of 587 BCE]; Childs, Book of Exodus [see n. 31], 261, 468; H outman , Exodus [see n. 31], 273; P ropp , Exodus 19–40 [see n. 23], 287). For the pre-D character of at least a core passage, see Schwienhorst-Schönberger, Bundesbuch (see n. 30), 406–414 (his basic stratum is pre-D, with vv. 29–30 as a Dtr redaction); H.-D. Neef, “ ‘Ich selber bin in ihm’ (Ex 23,21): Exegetische Beobachtungen zur Rede vom ‘Engel des

The Covenant Code Appendix and Neo-Assyrian Sources

69

This justifies and facilitates viewing the passage as the progenitor and model for a group of other passages that have multiple thematic correlations, including Exod 34:10–16; Deut 7; Josh 23; 24; and Judg 2:1–7, 20–23.41 Comparison of these passages to CCA confirms that they are subsequent to CCA. In briefest terms, Deut 7 appears secondary because it incorporates other motifs and thus looks expansive;42 is conceptually more advanced than CCA;43 and, most tellingly, conflates motifs from CC, the Decalogue, CCA, and the narrative of these texts.44 The laws of Exod 34:17–26 are clearly based on CC and so, presumably, the preface to these laws in vv. 10–16, which is similar to CCA, is based on CCA. Exodus 34:10–16 also features motifs in common with Deut 7.45 Joshua 23 proves derivative in combining CCʼs prohibition against speaking the name of other gods with CCAʼs prohibition against serving and bowing to other gods, Herrn’ in Ex 23,20–22; 32,34; 33,2; Jdc 2,1–5; 5,23,” BZ 39 (1995), 54–75, here 60–74; S ommer , Bodies (see n. 16), 210–212; Baden , Composition (see n. 28), 28, 110, 118, 119; Ausloos, “Angel” (see n. 31). 41  For Blumʼs ‫ ַמ ְלאְָך‬stratum, to which some of these passages belong, see below. 42  The broader or expanded range of motifs in Deut 7 includes the ‫ ֵח ֶרם‬motif (v. 2); a command not to show mercy, attached to the covenant prohibition (v. 2); prohibition of intermarriage (v. 3); the punishment of children of Israelites for apostasy (v. 4); a longer list of cult objects (v. 5); a longer description of blessings of fertility (vv. 13–14); an additional call to show no mercy (v. 16); a more extensive description of Canaanite destruction, including of their kings (vv. 23–24); and a repeated description of destruction, including of cultic objects (vv. 1–5, 22–26). From a literary point of view, the discursiveness of Deut 7 seems secondary to the succinct and quasi-poetic literary formulation of CCA (see n. 24). Though Deut 7 has only the “hornet” motif, without the companion “terror,” it puts this in a context of exodus miracles (vv. 17–21 [v. 20]). 43  It can be viewed as rounding up the nations from six to seven (v. 1). The object of obedience is all of the divine commands, not just the voice of the messenger (vv. 11–12). The deity is in the midst of the people (rather than the messenger before; v. 21). The second command to destroy cult objects is accompanied by a prohibition against keeping the objects because of their value (vv. 25–26). 44  Being a holy and treasured people (Deut 7:6; cf. Exod 19:5–6; 22:30), reference to those that love and hate the deity (Deut 7:9–10, 13, 15, from Exod 20:5–6 but complexly interwoven with blessings from CCA), and reference and analogy to the exodus (v. 8) and its miracles (v. 18–19). It is easier to argue that Deut 7 has brought these motifs together than to argue that they have been broken up and distributed throughout the context of Exod 19–24. This is consistent with the use by Dʼs laws of both CCʼs laws and its associated narrative. (Dʼs laws regularly assume the narrative in references to land acquisition and slavery in and the exodus from Egypt.) On CCA as the base text for Deut 7, see Neef, “Ich selber bin in ihm” (see n. 40), 60–61 (referring to G. Schmitt, Du sollst keinen Frieden Schließen mit den Bewohnern des Landes [BWANT 91; Stuttgart: Kohlhammer, 1970, 13–24]); M. Weinfeld, The Promise of the Land (Berkeley: University of California Press, 1993), 84–98. 45  Another sign that Exod 34:11–16 is secondary to CCA is the theme of intermarriage in 34:16, which arguably comes from Deut 7:3–4. Exod 34 develops the logic of how the lack of cultic destruction leads to intermarriage. That Exod 34:10–26 depends on CC, see Wright, Inventing (see n. 1), 481–482, n. 82, with literature.

70

David P. Wright

having a more elaborate snare motif, and depending on chapters surrounding Deut 7.46 Joshua 24 presents motifs common with CCA in an elaborate recounting of history from the patriarchs to the conquest and distributes them in four different speeches, two by Joshua and two by the people.47 Judges 2 appears secondary by virtue of its reference back to previous divine speech (presumably CCA) and the abbreviation of motifs from CCA.48  For correlations with CCA, see Josh 23:4, 5, 7, 9–13. Josh 23 appears to expand the snare motif to include several features: ‫יכם‬ ֶ ‫וּלשׁ ֵֹטט ְבּ ִצ ֵדּ‬ ְ ‫מֹוקשׁ‬ ֵ ‫וּל‬ ְ ‫( ְל ַפח‬v. 13), which echoes and may be connected with the textual development of Judg 2:3, which has two of these: ‫וְ ָהיוּ ָל ֶכם‬ ‫מֹוקשׁ‬ ֵ ‫יהם יִ ְהיוּ ָל ֶכם ְל‬ ֶ ‫אֹלה‬ ֵ ֵ‫ל ִצ ִדּים ו‬.ְ Josh 23 makes reference back to a previous promise (vv. 5, 10), which is presumably to CCA if not Deut 7. The passage is otherwise more elaborate than CCA in dealing with apportioning the land. Perhaps the clearest delimited proof is the phrase ‫וּב ֵשׁם‬ ְ ‫א־תזְ ִכּירוּ‬ ַ ֹ ‫יהם ל‬ ֶ ‫ֹלה‬ ֵ ‫( ֱא‬v. 7), which echoes CCʼs prohibition against speaking the names of other gods (and which is influential in Lev 24; see Wright, “Source Dependence” [see n. 7]). This verse combines language about serving and bowing down to other gods that correlates with CCA (see also v. 16; cf. Exod 23:24). At the same time, though there are no striking direct correlations with the wording of Deut 7, the passage shows a tie to the introductory chapters of Deut in the theme of loving the deity (v. 11; cf. Deut 6:5, 10:12; 11:1, 13, 22; and perhaps the motif of the whole heart and soul in Josh 23:14; cf. Deut 4:29; 6:5; 10:12; 11:13). 47  For correlations with CCA, see Josh 24:8, 11, 12, 17, 18, 19, 24. Josh 24 looks secondary because of its greater elaborateness; broad summary of patriarchal, exodus, and conquest history; contextualizing the hornet and expulsion motifs in the conquest history (v. 12); distribution of motifs related to CCA in four different speeches (two by Joshua, vv. 8, 11, 12, 19; two by the people, vv. 17, 18, 24); focus on foreign gods and images (vv. 14–15, 16, 19–24) reminiscent of the developed concerns of Deut 7 and Exod 34 (see also Gen 35:2, 4; Blum, Studien [see n. 31], 363); having a slightly larger list of peoples expelled and sins forgiven (vv. 11, 19; cf. Exod 23:21, 23); contextualizing forgiveness in a broad context of worshiping YHWH alone as opposed to the specific context of obeying the messenger (v. 19; cf. Exod 23:21); associating forgiveness with the motif of divine jealousy reminiscent of Exod 34, which otherwise shares motifs from CCA (v. 19; cf. Exod 34:14); tying the obedience motif to YHWH rather than a messenger (v. 24; cf. Exod 23:22); and describing a covenant made with YHWH and not just prohibiting one with the Canaanites (v. 25; cf. Exod 23:32). 48  See Judg 2:3 for the citation of previous divine speech. The two parts of that verse (2a, 2b) have language virtually identical to that in Exod 23:29 and 33. Judg 2:2a (the nonexpulsion clause) is only found in the CCA among the various passages being considered here. Judg 2:3 appears to bring these two elements together from CCA as an abbreviated citation. Judg 2:2 also appears to unite the separated motifs of cult destruction and covenant making from Exod 23:24 and 32. These themes are already juxtaposed in the secondary Exod 34:12–15, and the making of the covenant with the “inhabitant of the land” in Judg 2:2 is similar to Exod 34:12, 15 against Exod 23:32. Furthermore, Judg 2 emphasizes sin as opposed to CCA and has a rationale for the remaining nations that contradicts the beneficent reason in CCA and Deut 7:22. See Ausloos, “Angel” (see n. 31), broadly for the priority of CCA to Judg 2:1–5. For discussion of Judg 2:1–5 in the Joshua–Judges context, see B lum , Studien (see n. 31), 366–369. For correlations between CCA and Judg 2:1–5, see E. Blum, “Der kompositionelle Knoten am Übergang von Josua zu Richter: Ein Entflechtungsvorschlag,” in Deuteronomy and Deuteronomic Literature: Festschrift C. H. W. Brekelmans (BETL 133; Leuven: Leuven Uni46

The Covenant Code Appendix and Neo-Assyrian Sources

71

Another set of passages, in the Pentateuch, feature the vanguard motif (Gen 24:7, 40; Exod 13:21–22; 14:19, 24 [cf. v. 14]; 32:1, 23, 34; 33:2; Num 20:16; Deut 31:3, 6, 8).49 Several of these have the specific motif of a divine messenger in a context of the exodus and march to conquest. Their relationship to the vanguard motif in CCA warrants short discussion, especially as they lead to the question about the narrative context of CC and implications for the study of the Pentateuch. My original study argued that CC must have been created in connection with some larger narrative primarily because this legal text lacks contextualizing information. It is only implicitly (though rather clearly) presented as a revelation to Moses and the people. The Covenant Code appendix now provides more definitive evidence that CC was composed to fit a narrative context. Unless CCA can stand with CC as merely an ideal or self-contained statement about the future fortunes of the people, it indicates that CC and CCA would have been part of a story of national origins that led to or included territorial conquest. Erhard Blum has argued quite compellingly within a framework of a supplemental theory of Pentateuchal development that most of the passages featuring the messenger in a conquest context belong to a post-D and post-P 50 Fortschreibung of the text, a ‫מ ְלאְָך‬-Bearbeitung. ַ Nevertheless, if these passages are to be viewed as redactional additions, the argument of this paper requires them to be viewed as pre-Deuteronomic supplements. Though some have argued versity Press/Peeters, 1997), 181–212, here 190; Johnstone, “Reactivating” (see n. 31), 25; Neef, “Ich selber bin in ihm” (see n. 40), 69; Weinfeld, Promise (see n. 44), 157–159. 49  Some passages with the vanguard motif are reasonably post-CCA. Those related to the wilderness trek or conquest include Deut 1:30–32 (cf. Deut 20:1; 23:15); Josh 3:6; 6:7; 10:10 (implicit); Judg 4:14. Passages not directly related to the exodus or conquest include Isa 52:12; 58:8; Mal 3:1; cf. Isa 40:3. More broadly related to the vanguard motif in CCA are passages about a messenger helping in trouble, aiding in flight, attacking, or blocking travel plans: Gen 16:7–11; 21:17; 31:11; 32:2; 48:16; Num 22:22–35; Josh 5:13–15; 10:10; Judg 5:23; 13:6, 9; 1 Sam 29:9; 2 Sam 14:17, 20; 19:28; 24:16–17 (1 Chr 21:15–16); 1 Kings 19:5–8, 35 (1 Chr 32:21); 2 Kings 19:35 (Isa 37:36); Isa 63:9. 50  Blum, Studien (see n. 31), 361–382; idem, “Der kompositionelle Knoten” (see n. 48), 181–212. His main group of texts is Exod 14:19a; 23:20ff.*; (32:34aβ;) 33:2, 3b*, 4; 34:11–27; Judg 2:1–5. Deut 7 relates to this group. Other texts may belong, such as Gen 24:7, 40 (for Num 20:16, see n. 54). He warns against attributing all these passages to the same hand. His argument mainly lies in the perceived coherence formed by this group of texts as well as the relative lateness several of them betray in their contexts (most demonstrably Judg 2:1–5). He notes that there may be some older traditions reflected in the texts, but the wording and redactional place of the texts is primary. See also R. A lbertz , “The Late Exilic Book of Exodus (Exodus 1–34*): A Contribution to the Pentateuchal Discussion,” in The Pentateuch (ed. Thomas Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 243–256, here 246 and n. 12, and see the chart on p. 255 for the position of his late-fifth-century ‫מ ְלאְָך‬-Redaktion ַ (MalʼakR). Neef, “Ich selber bin in ihm” (see n. 40), 71–75, also sees a messenger redactional layer but sees texts like CCA as pre-Deuteronomistic (p. 71). That CCA does not fit the CC context is an important datum in recognizing this text as part of a secondary redactional layer. For a discussion of this redactional hypothesis, see Ausloos, “Angel” (see n. 31).

72

David P. Wright

just this,51 a consequence for theories of pentateuchal development is that this would point to the existence of a rather extensive pre-Priestly narrative about national origins. It might be possible to argue – against my observation earlier – that CC along with CCA was originally an independent composition, where CCA operated as a conclusion that encapsulated traditions about conquest. This work was then incorporated into a developing pentateuchal narrative and the other messenger or vanguard passages were added, either when CC and CCA were added or later over time, as analysis of those other passages dictates. A problem with a solution along the foregoing lines is that the various vanguard and messenger passages in the Pentateuch do not all cohere with the picture of divine guidance in CCA. That is, they do not simply flesh out CCAʼs portrayal. In particular, the instance in Exod 33:2, whose language is closest to that in CCA (‫יְבוּסי‬ ִ ‫ת־ה ְכּנַ ֲענִ י ָה ֱאמ ִֹרי וְ ַה ִח ִתּי וְ ַה ְפּ ִרזִּ י ַה ִחוִּ י וְ ַה‬ ַ ‫וְ ָשׁ ַל ְח ִתּי ְל ָפנֶ יָך ַמ ְלאְָך וְ גֵ ַר ְשׁ ִתּי ֶא‬, a combination of wording from 23:20, 23, and 28), is presented in a negative light, a punishment that takes the place of a positive and fundamental picture of the deity himself going in the midst of the people (see 33:1–6, 12–23; and loosely 34:2–9). This contradicts CCA, where the advance messenger is the ideal in guidance and care. There are therefore two models for divine guidance, going before and going among. It looks like two competing descriptions have been combined, and this points to a documentary solution. Admittedly this solution is not without its problems. Not all cases of the messenger or vanguard motif can be solved by source attribution. The vanguard passages associable with CCA fit in the narrative proposed for E by Neo-Documentary analysis.52 They chart a story in which, right after the revelation of CC+CCA and the covenant ceremony that established it (Exod 23:3–8), the people ask Aaron, in ironic contravention of the promise made in CCA, to make the golden calf to represent the deity that will go before them (‫ֹלהים ֲא ֶשׁר יֵ ְלכוּ ְל ָפנֵ ינוּ‬ ִ ‫;א‬ ֱ 32:1, 23).53 After this, Moses is told to start the journey to the land, during which the messenger will indeed go before the people (‫ִהנֵּ ה‬ ‫אָכי יֵ ֵלְך ְל ָפנֶ יָך‬ ִ ‫מ ְל‬, ַ 32:34, a combination of wording from 23:20 and 23). Later in the wilderness narrative, the people make reference to the deityʼs sending a messenger to lead them out of Egypt (‫וַ נִּ ְצ ַעק ֶאל־יְ הוָ ה וַ יִּ ְשׁ ַמע ק ֵֹלנוּ וַ יִּ ְשׁ ַלח ַמ ְלאְָך וַ יּ ִֹצ ֵאנוּ‬

 See Neef in n. 50.  For source analysis for the various passages in Exodus, see the text indices in Baden, Composition (see n. 28); idem, J, E, and the Redaction of the Pentateuch (FAT 68; Tübingen: Mohr Siebeck, 2009). For the different pictures of divine guidance in J and E, see B aden , Composition, 67–81; idem, “On Exodus 33.1–11,” ZAW 124 (2012), 329–340. 53  Some have suggested that the obedience clause in CCA in Exod 23:21–22 (and one can add CCʼs 23:13) operates in the narrative as a setup to the golden calf story (Exod 32). This has also been argued for the prohibition against making gods of gold at the very beginning of CC (20:23). 51

52

The Covenant Code Appendix and Neo-Assyrian Sources

73

‫מ ִמּ ְצ ָריִ ם‬, ִ Num 20:16).54 The instances of the vanguard motif in Deut 31:1–8 are probably D, though they build on the E motif.55 At the end of the journey, Moses tells the people that in the coming conquest YHWH will go before them (‫יְ הוָ ה‬ ‫ת־הגֹּויִ ם ָה ֵא ֶלּה ִמ ְלּ ָפנֶ יָך וִ ִיר ְשׁ ָתּם‬ ַ ‫ֹלהיָך הוּא ע ֵֹבר ְל ָפנֶ יָך הוּא־יַ ְשׁ ִמיד ֶא‬ ֶ ‫א‬, ֱ Deut 31:3). He restates this to Joshua before the people (‫וַ יהוָ ה הוּא ַהה ֵֹלְך ְל ָפנֶ יָך הוּא יִ ְהיֶ ה ִע ָמְּך‬, Deut 31:8; cf. v. 6, ‫ֹלהיָך הוּא ַהה ֵֹלְך ִע ָמְּך‬ ֶ ‫)א‬. ֱ In contrast, J describes the deityʼs presence more complexly as a pillar of cloud and fire that goes before but that is also understood as representing the deityʼs presence in the midst of the people (Exod 13:21–22; 14:14, 19, 24; 17:7; 33:1, 3–6bα, 12–23; 34:2–3, 4*, 5aβ–9; Num 14:14).56 Just prior to the miracle of the sea – hence quite a bit before CCA in the extant Pentateuch – it is announced that the deity will go before the people by day in a cloud column and by night in a column of fire (‫] וְ ַליְ ָלה ְבּ ַעמּוּד ֵאשׁ‬. . .[ ‫יֹומם ְבּ ַעמּוּד ָענָ ן‬ ָ ‫יהם‬ ֶ ֵ‫וַ יהוָ ה ה ֵֹלְך ִל ְפנ‬, Exod 13:21). In the sea event, the cloud column moves from before them to the rearguard, as here in 14:19 (cf. vv. 14, 24):  Blum, Studien (see n. 31), 119–120, n. 75 (cf. 377, n. 64), argues that the ‫ ַמ ְלאְָך‬in Num 20:16 is Moses (see also idem, “Der kompositionelle Knoten” [see n. 48], 192–193; Schmid, Genesis [see n. 28], 67, n. 110). Blum says that this instance is not connected to the theme of march to the land, comparing the parallel wording of 1 Sam 12:8, which uses similar language to say that YHWH sent Moses and Aaron to bring the people from Egypt, and that the term elsewhere is used of prophetic figures (Hag 1:13; Mal 1:1; 2 Chr 36:15–16; Ps 151:4 LXX; one may also compare Hos 12:14). It is possible, however, that the Samuel verse is interpretive (if it is dependent). It can be noted here that it is rather clear that the ‫ ַמ ְלאְָך‬in CCA is a supernatural figure. The messenger goes in front and is thus separate from the people (vv. 20, 23). It keeps the people safe, not clearly the capability of a single human (v. 20). It forgives sins (v. 21). The terror and “hornet” are supernatural and correlates of the messenger, with similar language describing their position (“before you”) and dispatch (“send”) (vv. 27–28). The vanguard motif in royal inscriptions, which appears to be influential here, has to do with supernatural entities (the gods or their emblems) and not with the king. Vanguard-related motifs in the Bible elsewhere and particularly in the pentateuchal texts refer to either a pillar of cloud/fire or the deity as in front and leading, and they use similar language (see below). Moses does not in fact bring the people into the land as described in 23:20. Judg 2:1–5, which depends on CCA (see n. 48), views the ‫ ַמ ְלאְָך‬as supernatural (it certainly is not Moses). Exod 32:34 and 33:2 portray the messenger as distinct from Moses. For discussion and literature, see Ausloos, “Angel” (see n. 31), 8–10. 55  For source analysis of Deut 31, see Baden, Composition (see n. 28), 137–139, 146–148, along with the contextual description of E versus D in J. Stackert, A Prophet Like Moses: Prophecy, Law, and the Israelite Religion (New York: Oxford University Press, 2014), 70, 117, 118 n. 114, 133, 140. Compare R. E. Friedman, The Bible with Sources Revealed (New York: Harper, 2003), 358–368. B. J. Schwartz, “A Bible Scholarʼs Simchat Torah – The Joy of Unraveling the Torahʼs Mysteries,” http://thetorah.com/a-bible-scholars-simchat-torah/ (accessed 2/23/15) has a rather different analysis of Deut 31–34, in which he assigns Deut 31:1–8 to E. 56  Jʼs portrayal of the arkʼs traveling before the people is relatable to divine guidance (Num 10:33 [cf. vv. 34–35]; cf. 14:44). 54

74

David P. Wright

Exodus 14:19

‫יהם‬ ֶ ‫אַח ֵר‬ ֲ ‫ֹלהים ַהה ֵֹלְך ִל ְפנֵ י ַמ ֲחנֵ ה יִ ְשׂ ָר ֵאל וַ יֵּ ֶלְך ֵמ‬ ִ ‫וַ יִּ ַסּע ַמ ְלאְַך ָה ֱא‬ ‫יהם‬ ֶ ‫אַח ֵר‬ ֲ ‫יהם וַ יַּ ֲעמֹד ֵמ‬ ֶ ֵ‫וַ יִּ ַסּע ַעמּוּד ֶה ָענָ ן ִמ ְפּנ‬

We will have reason to question whether v. 19a, which refers to a messenger, is original or belongs to J. As the narrative continues, after some sin, perhaps hinted at in Exod 32:25 (not the golden calf, as otherwise described in Exod 32), YHWH says that he will not go among the people, whereupon Moses pleads for him to reconsider and the deity relents (Exod 33:1, 3–6bα, 12–23; 34:2–3, 4*, 5aβ–9; I will return to 33:2). Later in the wilderness quail story, the deity criticizes the people for rejecting the deity who is among them (‫אַס ֶתּם ֶאת־‬ ְ ‫י־מ‬ ְ ‫ִכּ‬ ‫יְ הוָ ה ֲא ֶשׁר ְבּ ִק ְר ְבּ ֶכם‬, Num 11:20). In the later spy episode, Moses again bargains with YHWH and associates the cloud and fire column with the deityʼs being in the midst of the people. YHWH has to protect his reputation. The nations around have heard: Numbers 14:14 ‫אַתּה ה ֵֹלְך‬ ָ ‫וּב ַע ֻמּד ָענָ ן‬ ְ ‫אַתּה יְ הוָ ה וַ ֲענָ נְ ָך ע ֵֹמד ֲע ֵל ֶהם‬ ָ ‫ר־עיִ ן ְבּ ַעיִ ן נִ ְראָה‬ ַ ‫אַתּה יְ הוָ ה ְבּ ֶק ֶרב ָה ָעם ַהזֶּ ה ֲא ֶשׁ‬ ָ ‫׃‬ ‫וּב ַעמּוּד ֵאשׁ ָליְ ָלה‬ ְ ‫יֹומם‬ ָ ‫יהם‬ ֶ ֵ‫ִל ְפנ‬

This ties the presence of the deity among the people with the cloud and fire phenomena: the default situation is that cloud is over the people, hence “among” them, but it moves to the fore to lead them as they travel.57 If this mixes phenomenologies and shows signs of development, it is within the context of J itself. Two of the descriptions of the vanguard motif, which appear in larger Exodus J contexts, may be secondary – a very circumscribed ‫מ ְלאְָך‬-Schicht. ַ 14:19, cited just above, seems problematic in having a double description of the phenomenon. Since v. 19b accords with 13:21 and because v. 19a has ‫מ ְלאְָך‬, ַ not found in 13:21, it may be that v. 19a is an expansion.58 As an addition, v. 19a would seek to harmonize the portrayals of divine guidance in E and J.  Besides the shared theme of bargaining, the J spy story is integrally thematically tied to Exod 33–34* in the motifs of the divine glory (from a non-Priestly perspective; Exod 33:18, 22; Num 14:21–22), request for forgiveness (Exod 34:9; Num 14:19), epithets for the deity (Exod 34:6–7; Num 14:18), promise of the land to the patriarchs (Exod 33:1; Num 14:16, 23), and the contrast of commanding then blocking the peopleʼs arrival in the land that YHWH has promised to give (Exod 33:1, 3; Num 14:22–23). 58  B aden , Composition (see n. 28), 310, n. 28, says that v. 19a should be considered J. Note that Exod 3:2 (J) has a divine messenger (appearing in the flaming bush). It is possible, however, that v. 19a and perhaps other material in the environment that appears to intrude (or may be deleted to give a more streamlined story; i.e., vv. 20aα, 25a) are fragments of E. This would be intriguing in that it would parallel the apparent preoccupation with the divine name in the narrative ahead of CC (Exod 3:14–15). This concern is central to CC and generated by the use of LH. See W right, Inventing (see n. 1), 334–335. The problem with attribution of 14:19a to E is that CCA seems to announce the messenger for the first time in E. Moreover, v. 19a would drop the motif into E without any preparation. One would have to argue that E had some sort of setup for this, similar to Jʼs setup in 13:21–22, now lost. 57

The Covenant Code Appendix and Neo-Assyrian Sources

75

Similarly, the reference to the divine presence in 33:2 may be an addition. As Baden has argued, v. 3 flows from v. 1 and continues with the definition and description of the land.59 Exodus 33:2 in particular looks like it seeks to resolve the contradiction resulting from the juxtaposition of 32:34–35 (E) to 33:1, 3–6a (J). That the Pentateuch compiler has relocated the E tent of meeting passage (33:6b–11, which describes the disposition of the divine cloud pillar over the tent) to this environment from its place after the second set of tablets in Exod 34:1, 4*, 5aα, 28*–29* and before the E story of seventy elders in Num 11:11–12, 14–17, 24b–3060 makes 33:2 look like an addition from the compiler of pentateuchal sources himself, who is trying to harmonize competing portrayals of divine accompaniment and presence. One of the questions that the various examples of the vanguard motif in the Pentateuch raise is the extent to which traditional motifs and expressions have influenced CCA. If descriptions of the deity in the vanguard in J – such as in Exod 13:21 and 14:19b and especially in the J story of Abraham sending a servant to get a wife for Isaac, which refers to YHWH sending a messenger before him (‫הוּא יִ ְשׁ ַלח ַמ ְלאָכֹו ְל ָפנֶ יָך‬, Gen 24:7; v. 40, ‫ יִ ְשׁ ַלח ַמ ְלאָכֹו ִא ָתְּך‬. . . ‫ – )יְ הוָ ה‬predate the Covenant Codeʼs narrative (which is E or a proto-E), then they show that such notions were in CCAʼs native atmosphere and presumably influenced the formulation of CCA. These native traditions would be simply an auxiliary set of sources that have been brought into play in formulating CCA. The employment of native sources or traditions is found in CC proper.61 Other underlying native traditions that were presumably influential for CCA would include the effect of fear that befalls an enemy (note especially ‫ימ ָתה וָ ַפ ַחד‬ ָ ‫יהם ֵא‬ ֶ ‫ ִתּפֹּל ֲע ֵל‬in Exod 15:16)62 and the deity as a warrior who fights the enemy.63

 Baden, “On Exodus 33.1–11” (see n. 52), 331–333.  See Stackert, Prophet Like Moses (see n. 55), 83–92. See his reconstructed sequence of E on p. 92. 61  For example, the participial laws in Exod 21:12, 15–17; 22:17–19 appear based in part on a native tradition or source (see Wright , Inventing [see n. 1], 159–163, 192–204), and the apodictic laws in Exod 20:23–26 and 22:20–23:19 reflect native perspectives (Wright, Inventing, 286–321, 358). 62  See also Deut 2:25; 11:25; Josh 2:9; Ps 105:38. A ster , Unbearable Light (see n. 18), 255–257, 356, argues that ‫נֹורא ְת ִהֹלּת‬ ָ in Exod 15:11 corresponds to pulḫuti melamme, but he does not discuss ‫ימ ָתה‬ ָ ‫ ֵא‬in v. 16. If there is influence from royal inscriptions in the poem, the latter term may arise from this influence. Note that the motif of the enemiesʼ panic in v. 15 can be tied to royal inscriptions. Because the Song of the Sea can be analyzed as belonging to J (Baden, Composition [see n. 28], 275 n. 121), the date and relationship of E and J, discussed next, may bear on the relationship of ‫ימה‬ ָ ‫ ֵא‬in Exod 15:16 and 23:27. 63  For God as warrior, see, for example, Exod 14:14, 25; 15:1–18; 17:16; Deut 3:22; Josh 10 (esp. vv. 11, 14, 42); Judg 5:4, 20; 1 Sam 7:10–13; 17:47; 25:28; Ps 144:1; Prov 21:31; 2 Chr 32:8. This notion is portrayed in several other passages under discussion reflecting the vanguard motif. 59 60

76

David P. Wright

But one has to entertain the possibility that Jʼs descriptions of divine accompaniment are posterior to CCA. J appears to conflate two motifs that are distinct in E. The latter describes (1) a cloud (not involving fire) on the inaugural mountain of revelation and later at the tent of meeting outside the camp (Exod 33:7–11; Num 11:16, 24–27, 30; 12:4–5, 9–15; Deut 31:15) and (2) the vanguard motif of a messenger, terror, and “hornets” that guide the Israelites to their land and ensure conquest (CCA and the other E passages listed above).64 Jʼs portrayal blends and expands these two phenomenologies. The column is a cloud by day and fire by night and stands over the people, but it is maneuverable so that it can lead them and, in the miracle at the sea, even move protectively behind them. This complements other evidence or considerations that I have raised that J is secondary to and dependent on CCʼs narrative.65 If J should turn out not only to be posterior to E but also to respond to E, Exod 33:2 may well be original to J and seek to counter Eʼs notion of mere external divine guidance. Though many questions remain, this analysis puts us in a position to see the significance of the Neo-Assyrian period after 740 BCE, and more specifically from 700 BCE onward, as a matrix for the production of significant portions of pentateuchal and other biblical literature.66 It indicates that a lengthy narrative about national origins predates the P work. In terms of methodology for the historical analysis of the text, the recognition of the use of source materials in these and other texts can help explain the complexity in textual composition and help provide a control and anchor for redactional analyses.67 Appendix 1: The Future-King Passage from the Epilogue of the Laws of Hammurabi (cols. 48:59–51:91; curses summarized; my translation) Exhortation: In the future at any time, may any king who appears in the land keep the just commands (words) that I have written on my stela. May he not alter the law of the land that I have set down or the verdicts of the land that I have rendered, may he not remove my ordinances. If that man has understanding and he is able to provide justice for his land, let him give heed to the words that I have written on my stela. May this stela reveal to him the way, behavior, the law of the land that I have set down, and the verdicts of the land that I have rendered. May he secure justice for humankind. May he set down their

64  The cloud and messenger motifs in E may not be totally unrelated. In Exod 33:10 the people show reverence with respect to the appearance of the cloud at the revelation tent. It is a simple next step to correlate this cloud with the messenger. 65  See Wright, Inventing (see n. 1), 358–359. 66  For some consideration of the dating of non-Priestly/Holiness sources in this context, see Stackert, Prophet Like Moses (see n. 55), 31–35. 67  I thank my student Dustin Ray for his help preparing the manuscript for this essay.

The Covenant Code Appendix and Neo-Assyrian Sources

law. May he render their verdicts. May he root up the evil and wicked from his land. May he promote the well-being of his people.

Blessing: I am Hammurabi, king of justice, on whom Shamash has bestowed truth. My words are choice; my deeds are without equal. They are vanity to the fool, but to the wise they are objects of praise. If that man gives heed to my words that I have written on my stela and he does not remove my law, he does not overthrow my words, and he does not alter my ordinances – may Shamash lengthen the scepter of that man as he did for me, the king of justice. May he shepherd his people in justice.

Curses: If that man does not give heed to my words that I have written on my stela, and ignores my curses and does not fear the curses of the gods, and he (thus) upsets the law I have set down, overthrows my words, alters my ordinances, effaces my inscribed name and writes his name – or because of these curses has another person (do this) – that man, whether king, lord, ruler, or anyone whatsoever – § 1: May the great Anu, father of the gods, who has designated my reign, remove the splendor (ME.LÁM) of his kingship, break his staff, and curse his fate. Other curses follow from the following gods: § 2 Enlil, § 3 Ninlil, § 4 Ea, § 5 Shamash, § 6 Sin, § 7 Adad, § 8 Zababa, § 9 Ishtar, § 10 Nergal, § 11 Nintu, § 12 Ninkarrak, § 13 great gods of heaven, § 14 Enlil. (See excerpts from some of these curses in tables 3 and 4, above.)

Appendix 2: Sample Passages and Motifs from Neo-Assyrian Royal Inscriptions Index of motifs exemplified in the following inscriptions i. Campaign against enemies (pervasive in inscriptions, examples not cited here) ii. Vanguard motif: the gods or divine standard going before the king and army in battle (variant, god going or standing at the side) iiia. Terror/fear (puluḫtu, pulḫu, ḫattu, rašubbatu) or splendor (melammu, namarrutu, namrīru), sometimes combined (e.g., puluḫti melamme), that overwhelm (saḫāpu, tabāku) the enemy iiib. Royal ṣiriḫtu “anger” or ṣarāḫu (N) “be angry” (but see n. 22, above) iv. Destroying (e.g., smashing, šubburu) gods or temples or pillaging cult items v. Lists of enemies (including gentilic formulation) vi. Motifs in future-prince passages: blessings and curses (e.g., famine, political destruction, extinction), the royal name, obedience (possibly also treaty)

77

78

David P. Wright

A. Adad-Nirari II [911–891 BCE], RIMA 2 A.0.99.2. ii. God going before: (97) “By the command of Aššur, the great lord, my lord, and the goddess Ištar, mistress of battle and strife, who goes at the head of my extensive army (a-lik-at pa-na-at ERIN. ḪI.A.MEŠ-ia DAGAL.MEŠ)” vi. Future prince, blessing and curse (royal name, political destruction, destruction of posterity): (131b–133) “May a later prince restore it (the temple) (and) return my inscribed name to its place. (Then) Aššur and the goddess Gula will listen to his prayers. As for the one who removes my inscriptions and my name: may Aššur and the goddess Gula overthrow his sovereignty (and) destroy his name (and) his seed from the land.” B. Aššur-Naṣirpal II [883–859 BCE], RIMA 2 A.0.101.17. ii. Divine standard going before: (ii 89–92) “With the supreme might of the divine standard which goes before me (ina Á.MEŠ MAḪ.MEŠ (90) šá ⸢d⸣Ù[RI.GAL] DU ⸢IGI⸣-ia) I fought with them, brought about their defeat, (and) broke up their band.” iiia. Melammu of god overwhelming: (i 79) “While I was in Mount Kirruru the radiance of Aššur, my lord, overwhelmed the Ḫubušku and the Gilzānu (me-lam-me ⸢šá aš-šur⸣ EN-a ⸢ìs⸣-ḫup-šú-nu).” (iv 89) “I unleashed against them my lordly radiance (me-lam-me (90) EN-ti-a UGUšú-nu at-bu-ku).” vi. Future prince and curse (royal name, blessings of political security and abundance, disobedience, national and political destruction, famine, military conflict): (v 24b–45a) “May a later prince restore its weakened (portions and) restore my inscribed name to its place. (Then) Aššur will listen to his prayers. He must not forsake my mighty palace, my royal residence, of Calah, nor abandon (it) in the face of enemies. He must not remove the doors, beams, (or) knobbed nails (v 30) of bronze from it (and) put them in another city (in) another place. He must not smash its beams. He must not tear out its drain pipes. [Other architectural considerations continue here in a similar way.]” (v 45b–54a) “As for the one who acts according to this inscription of mine (and) does not alter the ordinances of my text: may Aššur, the god Enlil, (and) the great gods who make my sovereignty supreme make his dominion supreme in all lands. (v 50) May they constantly lead him in victory, might, and heroism. May they grant the tribute of the four quarters as his portion. May they establish in his land plenty, affluence, and abundance (nu-uḫ-šu ṭù-uḫ-du ù ḫe-gal-lu ina KUR-šu lu-kín-nu).” (v 55–95) “As for one who does not act according to (v 55) this inscription of mine (but) alters the ordinances of my text; (who) destroys this monument, discards (it), covers it with oil, buries it in dust, (v 60) burns it with fire, throws it in water, puts it in the path of beasts or the track of animals; (who) prevents scholars from seeing and

The Covenant Code Appendix and Neo-Assyrian Sources

reading the ordinances of my inscription, bars anyone access (v 65) to my inscription in order that it might not be seen and read; (who), because of these curses, instructs and incites a stranger, a foreigner, a malignant enemy, a prisoner, or any living (v 70) being so that he destroys, chisels away, changes its wording to something else; (v 75) (who) makes up his mind and decides to destroy this monument of mine and to alter my ordinances and (therefore) commands a scribe or diviner or anyone else, ‘Destroy this monument! Its dictates are not to be observed!’ (v 80) and whoever heeds his statements; (who) conceives anything injurious and orders (it to be done) to my works and my monument; (who) says, ‘I know nothing (of this)’ and during his sovereignty diverts his attention elsewhere (v 85) with the result that this monument is destroyed and smashed (or) the wording of its text altered; or (who) seeks (to do) evil against this monument of mine: may Aššur, the great lord, the Assyrian god, (v 90) lord of destinies, curse his destiny; may he remove his works; may he pronounce an evil curse for the uproot of the foundations of his sovereignty and the destruction of his people; may he inflict his land with distress, famine, hunger, (v 95) and want (su-um-qa bu-bu-ta ⸢ù ni-ib⸣-[r]i-⸢tú⸣ ù ḫu-šá-aḫ-ḫa a-na KUR-ti-šú li-du-ú).” (v 96b–103) “As for the one who becomes angry with this monument of mine and says, ‘What is this?’: may the gods Anu, Enlil, and Ea, the gods who approve of me, (v 100) decree by their weighty edict his unhappiness; may they establish a truceless war without terms, strife, conflict, (and) battle in his land.” C. Shalmaneser III [858–824 BCE], RIMA 3 A.0.102.2. ii. Divine standard going before: (i 44–45) “With the exalted might of the divine standard which goes before me (ina Á.MEŠ ṣi-ra-a-ti šá dÙRI.GAL a-lik IGI-ia) (and) with the fierce weapons which Aššur my lord gave to me, I fought (and) defeated them.” (ii 70) “By the command of Aššur, the great lord, my lord, and the divine standard which goes before me (u dÙRI.GAL a-lik IGI-ia), I approached Mount Šītamrat [. . .].” (ii 96–97) “With the supreme forces which Aššur, my lord, had given to me (and) with the mighty weapons which the divine standard, which goes before me (ÙRI.GAL a-lik IGI-ia), had granted me I fought with them.” iiia. Fearful melammu that overpowers: (i 22–23) “Overwhelmed by fear of the radiance of Aššur, my lord (púl-ḫi me-lam-me šá aš-šur EN-ia is-ḫu-pu-šu-nu-ti), they came down (and) submitted to me.” (i 30–31) “Overwhelmed by fear of the radiance of Aššur, my lord (pu-ul-ḫi me-lam-me šá aš-šur EN-ia is-ḫu-pu-šú-nu), they fled upstream/higher to save their lives.” (ii 74) “Fear of the brilliance of Aššur, my lord, overwhelmed them (pu-ul-ḫi melam-me šá aš-šur EN-ia is-ḫu-pu-šú-nu) (and) they came down (and) submitted to me.” Compare the related formulations: (i 21) “(they) became frightened in the face of the flash of my weapons (TA pa-an na-mur-rat GIŠ.TUKUL.MEŠ-ia ip-la-ḫu-ma).” [In the context of i 22–23, cited above.]

79

80

David P. Wright

(ii 68) “He (Aḫunu of Bīt-Adini) became afraid in the face of the flash of my weapons (and) my lordly brilliance (TA IGI na-mur-rat GIŠ.TUKUL.MEŠ-ia me-lam-me [šá] EN-ti-ia ip-laḫ-ma).” (ii 76) “They became frightened in the face of the flash of my mighty weapons and my tumultuous onslaught (TA IGI na-mur-rat GIŠ.TUKUL.MEŠ-ia KAL.MEŠ u MÈ-ia šit-mu-ri ip-la-ḫu-ma).” [Near ii 74, cited above.] (ii 79) “They were frightened of my lordly fearfulness (and) the flash of my fierce weapons (púl-ḫa-at EN-ti-ia na-mur-rat GIŠ.TUKUL.MEŠ-ia ez-zu-te ip-la-ḫu-ma).” v. Lists of enemies (including gentilic formulaton): A list of rebellious kings with descriptions of nationality appears in ii 83–84: “I received tribute from kings on the opposite bank of the Euphrates, from Sangara the Carchemishite, from Kundašpu the Kummuḫite, from Aramu, the man of Bīt-Abūsi, from Lalla the Melidite, from Ḫaiiānu the man of Bīt-Gabbari, from Qalparuda the Patinean, (and) from Qalparuda the Gurgumite.” The text continues with another list in ii 89–95, which includes Ahab the Israelite (i.e., m a-ḫa-ab-bu KUR sir-ʼa-la-a-a). Here the Assyrian king took chariots, cavalry, troops, and animals from “. . . Hadad-ezer (Adad-idri), the Damascene; [. . .] Irḫulēnu, the Ḫamatite; [. . .] Ahab (Aḫabbu) the Israelite (Sirʾalāia); the land Irqanatu; [. . .] Matinu-ba‍ʾal of the city Arvad; [. . .] the land Usanātu; [. . .] Adunu-baʿal of the land Šianu; [. . .] Gindibu of the Arabs; [. . .] Ba‍ʾasa, the man of Bīt-Ruḫubi, the Ammonite.” [Just after this the text refers to battle and has the vanguard motif, cited above, ii 96–97.] D. Sargon II [721–705 BCE], Eighth Campaign, “Letter to Aššur”; W. Mayer, Assyrien und Urartu I: Der Achte Feldzug Sargon II. im Jahr 714 v.Chr. (AOAT 395/1; Münster: UgaritVerlag, 2013). Translation here adapted from Mayer. ii. Divine standard going before: (line 14) “Against Zikirtu and Andia I guided the yoke of the standard wagon of Nergal and Adad, who go before me (šá dÙRI.GAL dIŠKUR ú-rí-gal-li a-li-kut maḫ-ri-a).” (line 319): “with the valuable consent of the hero Šamaš, who caused reliable omens to be written on the liver, (that indicated that) he would go at my side (ša a-lak i-di-ia).” iiia. Fearful melammu covering, falling on enemy: (line 69) “The fear of my splendor fell upon them, terror befell them in their land (pu-luḫ-ti me-lam-me-ia ik-tùm-šu-nu-ti i-na qé-reb KUR-šu-nu im-qut-su-nu ḫat-tu).” (Compare line 420; § 176.) iv. Spoiling gods and temple: (lines 367–368, 405) “My eunuchs (and) soldiers for the campaign I sent into the temple of Haldi. Haldi, his god, and Bagbartu, his goddess, together with uncountable possessions from his temple, as much as was present: [a long list is given here] . . . All his great wealth, which cannot be enumerated, I carried forth.”

The Covenant Code Appendix and Neo-Assyrian Sources

81

(lines 408–409) “The property of the palace of Urzana and (the god) Haldi, with its enormous wealth, which I carried off from Munasir city, I loaded on my numerous armies, in their masses, and had them transport it into Assyria.” (lines 423–425) “From Urzana of Musasir I carried off Haldi, his god, Bagbartu, his goddess, along with the great property of his temple [. . .].” E. Sennacherib [704–681 BCE], Bavian Inscription H 3, D. D. L uckenbill , Annals of Sennacherib (OIP 2; Chicago: University of Chicago Press, 1924), 78–85 (see E. F rahm , Einleitung in die Sanherib-Inschriften [AfO Beiheft 26; Vienna: Institüt für Orientalistik der Universität Wien, 1997], 151–154, text T 122); translation of the primary part of the text by M. C ogan , in COS 2.119E. (The campaign against Babylon, described in the inscription, occurred in 689 BCE.) iiia. Fear poured out on enemies: (41, Cogan) “Perhaps Sennacherib king of Assyria will become angry and will set out for Elam once again. Fear and terror (ḫat-tu pu-luḫ-tu) were poured out (it-ta-bi-ik-ma) over all of Elam and they left their land to save their lives.” iv. Divine images and cult places destroyed: (47–48, 51–53, Cogan) “I handed out the wealth of that city [Babylon] – silver, gold, precious stones, property and goods – to my people and they made it their own. My men took the (images of the) gods who dwell there and smashed them (ik-šú-su-nuti-ma. . . ú-šab-bi-ru-ma) [. . .]. I destroyed and tore down and burned with fire the city (and) its houses, from its foundations to its parapets. I tore out (as-suḫ-ma) the inner and outer walls, temples, the ziggurat (É.MEŠ DINGIR.MEŠ zig-gur-at) of brick and earth, as many as there were, and threw them into the Arahtu river. I dug canals through the city and flooded its place with water, destroying the structure of its foundation. I made its devastation greater than that of ‘the Flood’ (a-bu-bu).” vi. Future prince and curses (treaty?, divine names, political destruction): (57–60, Luckenbill) “If ever there is a future prince among the kings, my sons, who destroys the work which I have done, (and) breaks the covenant [or: dismantles the canal system; see n. 29, above] I have (hereby) made with him (rik-sa-te ar-ku-su i-paṭṭa-ru), diverts the course of the waters of those canals from the plain of Nineveh, may the great gods, all whose names are named in these stelas, by the word of their mouth, a holy decree which cannot fail, curse him with an evil curse, and overthrow his rule.” F. Esarhaddon [680–669 BCE], RINAP 4.1 (“Nineveh A”; from various hexagonal prisms from Nineveh, Aššur, and Susa; 673 and 672 BCE, with later copies). ii. God standing at the side: (i 74–75) “The goddess Ištar, the lady of war and battle, who loves my priestly duties, stood at my side (i-da-a-a ta-zi-iz-ma), broke their bows, (and) she split open their tight battle ranks.”

82

David P. Wright

iiia. Fear and fearful splendor overwhelming enemies: (i 72) “Fear of the great gods, my lords, overwhelmed them (pu-luḫ-ti DINGIR.MEŠ GAL.MEŠ EN.MEŠ-ia is-ḫup-šú-nu-ti-ma), (and when) they saw my mighty battle array, they became like crazed women.” [This is followed by i 74–75, about the god going at the side.] (iv 37–39) “The awesome fear of the god Aššur, my lord, overwhelmed them (pu-ul-ḫi me-lam-me ša daš-šur EN-ia is-ḫup-šu-nu-ti) (and) they brought to Nineveh, my capital city, large thoroughbreds (and) blocks of lapis lazuli [. . .] and they kissed my feet.” iiib. Kingʼs anger with verb ṣarāḫu (N): (i 57) “I, Esarhaddon, who with the help of the great gods, his lords, does not turn back in the heat of battle, quickly heard of their evil deeds. I said ‘Woe!’ and rent my princely garment. I cried out in mourning, I raged like a lion, and my mood became furious (lab-biš an-na-dir-ma iṣ-ṣa-ri-iḫ ka-bat-ti).” [The “going” of the gods in battle is found in i 62 and overwhelming “fear” appears just a few lines later in i 71, cited in motif iiia, above.] (ii 51) “I heard of his evil deeds (while) in Nineveh; my heart became angry and my liver was inflamed (iṣ-ṣa-ri-iḫ ka-bat-ti).” iv. Gods plundered (and returned): (iv 1–16) “(As for) the city Adumtutu, the fortress of the Arabs, which Sennacherib, king of Assyria, (my) father, who engendered me, conquered and whose goods, possessions, (and) gods, together with Apkallatu, the queen of the Arabs, he plundered and brought to Assyria – Hazael, the king of the Arabs, came to Nineveh, my capital city, with his heavy audience gift and kissed my feet. He implored me to give (back) his gods, and I had pity on him. I refurbished the gods Atar-samayin, Dāya, Nuḫāya, Ruldāwu, Abirillu, (and) Atar-qurumâ, the gods of the Arabs, and I inscribed the might of the god Aššur, my lord, and (an inscription) written in my name on them and gave (them) back to him. I placed the lady Tabūa, who was raised in the palace of my father, as ruler over them and returned her to her land with her gods.” vi. Future king (royal name, blessing): (vi 65–74) “In the future, may one of the kings, my descendants, whom the god Aššur and the goddess Ištar name to rule the land and people, renovate the dilapidated section(s) of that palace when it becomes old and dilapidated. Just as I placed an inscription written in the name of the king, (my) father, who engendered me, beside an inscription written in my name, so you (too) should be like me (and) read an inscription written in my name, anoint (it) with oil, make an offering, and place (it) beside an inscription written in your name. The god Aššur and the goddess Ištar will (then) hear your prayers.”

The Covenant Code Appendix and Neo-Assyrian Sources

83

G. Esarhaddon, RINAP 4.98 (“Esarhaddonʼs Monument A”; stela from Zinçirli; post671 BCE). ii. God going at the side: (obv. 9) (Various gods listed, including:) “the goddess Ištar, lady of war and battle, who goes at my side (a-li-kàt i-di-ia).” iiia. King clothed in splendor (namurratu): (obv. 20–24) “[. . .] the merciless, the one who curbs the insolent ones, the one who is clothed in splen[dor] (la-biš na-mur-[ra-ti]), fearless in battle, per[fect] warrior [. . .].” iv. Restoration of gods and cult: (obv. 34–rev. 7) “The one who provides provisions for the great gods, kn[ows] how to revere the gods and goddesses, [. . .] [the one who (re)construct]ed the temple of the god Aššur, completed its ornaments, (re)built Esagil and Babylon, restored the rites, (and) who returned the plundered gods of the lands to their (proper) place from the city Aššur; the king whose food offerings the great gods love and whose priestly service they established forever [in the tem]ples.” vi. Future prince (royal name, military defeat): (rev. 50–57) “I had a stele written in my name made and I had inscribed upon it the renown (and) heroism of the god Aššur, my lord, (and) the might of my deeds which I had done with the help of the god Aššur, my lord, and my victory (and) triumph. I set (it) up for all time for the admiration of all (my) enemies. Whoever takes away this stele from its place and erases my inscribed name and writes his name, covers (it) with dirt, throws (it) into water, burns (it) with fire, or puts (it) in a place where (it) cannot be seen, may the goddess Ištar, lady of war and battle, change him from a man into a woman, and may she seat him, bound, at the feet of his enemy. May a future ruler (NUN-u EGIR-u) look upon a stele written in my name, read (it) aloud (while standing) in front of it, anoint (it) with oil, make an offering, (and) praise the name of the god Aššur, my lord.” H. Assurbanipal [668–631 BCE], Rassam Decagon/Cylinder; H. C. R awlinson and T. G. Pinches, A Selection from the Miscellaneous Inscriptions of Assyria and Babylonia (The Cuneiform Inscriptions of Western Asia 5; London: Oriental Antiquities, British Museum, 1880) = Pinches V R, pl. 1–10; M. S treck , Assurbanipal, II. Teil: Texte (Vorderasiatische Bibliothek; Leipzig: J. C. Hinrichsʼsche Buchhandlung, 1916) (by column/line); ARAB, 290– 323 (by § ). ii. Gods going before, going at the side: (iv 49; § 794) “Assur, Sin, Shamash, Adad, Bêl, Nabû, Ishtar of Nineveh, the queen of Kidmuri, Ishtar of Arbela, Urta, Nergal (and) Nusku, who march before me (ša ina maḫ-ri-ia il-li-ku), slaying my foes, cast Shamash-shum-ukîn, my hostile brother, who became my enemy, into the burning flames of a conflagration and destroyed him.” (v 100; § 807) “Ishtar, who dwells in Arbela, in the night time revealed a dream to my armies, thus she addressed them: ‘I will go before Assurbanipal (a-na-ku al-lak in ma-ḫar IAN.ŠÁR-DÙ-A), the king whom my hands have formed.’ ”

84

David P. Wright

(v 29; § 803): “Assur and Ištar, who go at my side and cause me to stand upon (the neck) of my foes, looked into the heart of Tammaritu, the wicked rebel. [. . .] They brought him [. . .] into submission at my feet.” iiia. Splendor or terror overcoming, covering: (i 84–85; § 771) “The terrible splendor of Assur and Ishtar overcame him and he went mad. The glory of my majesty, with which the gods of heaven and earth have crowned (adorned) me, overpowered (lit., covered) him (nam-ri-ri AN.ŠÁR u dXV is-ḫu-pu-šú-ma il-li-ka maḫ-ri-taš me-lam-me LUGAL-u-ti-ia ik-ta-mu-šú-ma).” (ii 20–21; § 775) “As for Tarkû, in the place to which he had fled, the terror of the weapon of Assur, my lord, overwhelmed him (a-šar in-nab-tu ra-šub-bat GIŠ.TUKUL AN.ŠÁR EN-ia is-ḫu-up-šú-ma) and the night of death (lit., the fate of his night) overtook him.” (iv 119–220; § 800) “[. . .] I marched upon Elam. The terrible brilliance of Assur and Ishtar, my lords, (and) the fear of my majesty, overwhelmed them (nam-ri-ri AN.ŠÁR d XV EN.MEŠ-ia pu-luḫ-ti LUGAL-ti-ia is-ḫu-up-šú-nu-ti).” (vii 53–54; § 816) “Paʼe, who exercised the rulership over Elam in place of Ummanaldash, reflected upon the fury of the terrible weapons of Assur and Ishtar, which had been poured out over Elam (na-mur-rat GIŠ.TUKUL.MEŠ AN.ŠÁR u dXV ez-zu-ti ša [. . .] it-bu-ku UGU KUR.NIM.MA.KI).” (vii 75–76; § 816) “[. . .] these people, who had made Mount Salatri their stronghold, the splendor of Assur and Ishtar, my lords, overwhelmed them (nam-ri-ri AN.ŠÁR u dXV EN.MEŠ-ia (76) is-ḫu-up-šu-nu-ti).” (Compare ix 79–81; § 829.) iiib. Royal ṣiriḫtu “anger”: (v 37; § 803) “With my heart enraged (burning) (i-na ṣi-ri-iḫ-ti ŠÀ-ia) at these crimes which Tammaritu the faithless has committed, in the power and might of the great gods, my lords, I advanced victoriously into the midst of Elam, and all over it.” [The godsʼ going at the side appears a few lines earlier (v 29) and, just after, a list of twenty-eight cities captured (v 43–57; § 804), followed by description of city destruction and god pillaging (v 59–62).] i 64 § 770: “At these deeds my heart became enraged, my soul burned (ŠÀ-bi e-gug-ma iṣ-ṣa-ru-uḫ ka-bit-ti).” iv. Gods smashed and pillaged, temple destroyed: (v 117–123; § 808) “[The city] Bashimu and the cities of its environs I destroyed, I devastated. I struck down the people living therein. I smashed their gods, and pacified the divine heart of the lord of lords. His gods, his goddesses, his property, his goods, his people, great and small, I carried off to Assyria” ((119) ú-šab-bir DINGIR.MEŠ-šú-un (122) ú-šap-ši-iḫ DINGIR ka-bit-ti EN EN.EN (cf. CAD K 12b at 2.2′) (121) DINGIR.MEŠ-šú dXV.MEŠ-šú NÍG-šú NÍG.GA-šú (123) UN.MEŠ DUMU u GAL áš-lu-la a-na KUR.AN.ŠÁR.KI).

The Covenant Code Appendix and Neo-Assyrian Sources

(vi 27, 62–64; § 810) “The zikkurat of Susa, which was built of enameled bricks, I destroyed.” [Followed by description of pillaging of numerous gods and other objects.] “The sanctuaries of Elam I destroyed totally (lit., to nonexistence). Its gods (and) goddesses I turned into ghosts.” [For the last phrase, see CAD M/1 226; Z 59, 60]. v. List of enemies: See v 43–57; § 804 for a list of twenty-eight cities captured and destroyed. vi. Future prince (royal name, blessing of political stability, general curse): (x 108–120; §§ 838–839) “In days to come, may the one among the kings, my sons, whom Assur and Ishtar shall call to rule land and people, when that bît-ridûti shall become old and fall into ruins, restore its ruins. Let him look upon (or, perhaps, look for) the memorial with the inscription of my name, (that of) my father (and) my fatherʼs father – an endless line of royalty, let him anoint it with oil, offer sacrifices, and set it up alongside the memorial inscribed with his name. And the great gods, all whose (names) are written upon this memorial, will surely grant him, as they did me, power and might. Whoever destroys the memorial inscribed with my name, (that of) my father (and) my fatherʼs father, and does not set it up alongside his memorial, may Assur, Sin, Shamash, Adad, Bêl, Nabû, Ishtar of Nineveh, the queen of Kidmuri, Ishtar of Arbela, Urta, Nergal, and Nusku, pass judgment upon him in my name.” [For the same list of gods, see ix 61–64; § 828 and x 33–36; § 833.]

85

Data to Inform Ongoing Debates about the Formation of the Pentateuch From Documented Cases of Transmission History to a Survey of Rabbinic Exegesis David M. Carr The last few years have seen an explosion of work on documented cases of textual revision in Israel and surrounding areas. In 2011 I published a book on the formation of the Hebrew Bible that built on a number of previous studies and summarized insights I had derived from my own decade-long survey of primary-text data regarding documented cases of transmission history in Mesopotamia, the Bible itself, and Qumran finds.1 In the years after I completed the book (in 2010), important works on documented cases of ancient scribal revision that can substantially inform study of the formation of the Pentateuch have appeared, including contributions by Sara Milstein, Russell Hobson, Molly Zahn, Paul Delnero, Juha Pakkala and Reinhard Müller, and others.2 This essay, based on my contributions to the 2013 and 2014 Jerusalem Pentateuch conferences, summarizes several ways in which study of documented cases – so-called empirical evidence, – can contribute to Pentateuchal study.

 D. M. C arr , The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011). The vast majority of this survey work, in the form of parallels between editions, lists of divergences, and other notes, was not published in the book. This essay continues to synthesize insights from that survey work. 2  S. Milstein, “Reworking Ancient Texts: Revision through Introduction in Biblical and Mesopotamian Literature” (PhD diss., New York University, 2010; being prepared for publication); R. Hobson, Transforming Literature into Scripture: Texts as Cult Objects at Nineveh and Qumran (Sheffield: Equinox, 2012); M. M. Z ahn , Rethinking Rewritten Scripture: Composition and Exegesis in the 4QReworked Pentateuch Manuscripts (STDJ 95; Leiden: Brill, 2011); P. Delnero, “Memorization and the Transmission of Sumerian Literature,” JNES 71 (2012), 189–208; J. Pakkala, Godʼs Word Omitted: Omissions in the Transmission of the Hebrew Bible (Göttingen: Vandenhoeck & Ruprecht, 2013); J. Pakkala and R. M üller , Evidence of Editing: Growth and Change of Texts in the Hebrew Bible (Atlanta: Scholars Press, 2014). 1

88

David M. Carr

Fluidity in Scribal Transmission of Textual Traditions Ancient scribes were far more fluid in the way they reproduced ancient written traditions than many scholars have presupposed heretofore. Working in an oral-written environment where they had memorized many texts verbatim, scribes often substituted words and phrases that they considered semantically equivalent to the known text. The resulting phenomenon in the textual evidence, where semantically equivalent words, grammatical expressions, and phrases are exchanged, minor particles are added or subtracted, shifts in order are introduced, etc., is what I have termed “memory variants.”3 For example, even though there is substantial evidence for some scribesʼ preference for Yahweh or Elohim, such as in Gen 20–21 or in parts of the book of Psalms, there is also evidence in manuscript transmission that scribes sometimes felt that those divine designations were equivalent and switched them. Similarly, we have documentation for the switching of the terms ‫ שפחה‬and ‫אמה‬, terms that are often used to distinguish J and E.4 The issue here, of course, is not just the handful of cases where such variation in divine designations or other phraseology is actually documented in existing manuscript witnesses. Rather, the problem is that early manuscripts document a probable broader process of undetectable, fluid switching of terms, addition and elimination of particles, and order shifts that are used as evidence for transmission-historical arguments that depend too much on the presence of specific language elements in a given text.5 Insofar as the case for J and E depends on such arguments, that case suffers particularly from the fact that divine designations and other terms were transmitted in such a fluid environment. In contrast, the theory distinguishing P from non-P is so well established by doublets and contradictions, combined with extensive phraseological and ideological differences, that this theory is less threatened than the J/E hypothesis by the fact that some terminological differences between P and non-P were blurred by subsequent scribes.  J. Tigay, The Evolution of the Gilgamesh Epic (Philadelphia: University of Pennsylvania Press, 1982), 58–68, 88, 218–222; Carr, Formation of the Hebrew Bible (see n. 1), 41–42, 49–50, 54–55, 57–65, 105–110; and now Delnero, “Memorization and Transmission” (see n. 2). 4  Carr, The Formation of the Hebrew Bible (see n. 1), 107–108. See also the much more extensive study in Hobson, Literature Into Scripture (see n. 2), and cf. the different evaluation of the data in E. Tov , “The Source of Source Criticism: The Relevance of Non-Masoretic Textual Witnesses,” in Text – Textgeschichte – Textwirkung: Festschrift Siegfried Kreuzer (ed. Jonathan Robker; Münster: Ugarit Verlag, 2015), 283–301. I am not as convinced as Tov that all the divergences from the MT in non-MT text witnesses on divine designations and other terms can be explained as harmonizations and other clearly post-MT developments. 5  Therefore I believe the data from early nonmasoretic witnesses is more relevant for the documentary and other transmission-historical hypotheses than argued in Tov , “Source of Source Criticism” (see n. 4). 3

Data to Inform Ongoing Debates about the Formation of the Pentateuch

89

This phenomenon of frequently fluid transmission of written traditions in the Second Temple period has implications for questions of linguistic dating discussed elsewhere in this volume:6 it could contribute to a distinction between (a) the reliability of linguistic dating in establishing an early date for a given biblical text and (b) the reliability of linguistic dating in establishing a late date for a given text. On the one hand, most would agree that our earliest manuscripts of biblical texts (third–first centuries BCE) date from a period when scribes could not successfully write large swaths of texts in perfect classical-Hebrew idiom. This is a point well made by Jan Joosten in his studies of pseudoarchaisms, including in his contribution to this volume.7 Insofar as this is true, texts that happen to have come down to us in relatively well-preserved classical Hebrew are likely to predate the disintegration of the classical Hebrew dialect. This shift hardly occurred at one time in all loci, but the disintegration appears to have started in the exilic period and probably was universal by the middle of the Persian period. Present biblical texts judged to have an overall profile of classical Hebrew need not feature perfect classical Hebrew, given the likelihood of fluid transmission up through the earliest manuscripts, but the widespread lack of late linguistic features in substantial swaths of a given biblical text would seem to be a good indicator of potential exilic or preexilic origins of at least the textʼs core (aside from framing elements, later memory variants, and limited conscious updates). In contrast, since all of our manuscripts, including manuscripts of texts presumed to be of early origin, were transmitted across centuries in a frequently fluid environment dominated by Aramaic and postclassical Hebrew, even the earliest of these texts could be infused with late linguistic features by copyists. This seems to have been particularly true of manuscripts of non-Torah texts, such as Song of Songs, which were relatively marginal in use and authority in the latter half of the Second Temple period.8 This would suggest that the presence of late linguistic features in a biblical book is a relatively unreliable guide to the probable lateness of that book. Those features could point to its late origins, or they could point to high levels of contamination of an originally early (exilic or preexilic) book in the process of its transmission into the late Second Temple period (or even to the fact that an early form of the book was not written in the classical Hebrew literary dialect).

6  See in particular the essays of Jan Joosten, “Diachronic Linguistics and the Date of the Pentateuch,” and Erhard Blum, “The Linguistic Dating of Biblical Texts,” in this volume. 7  The initial classic study is J. Joosten, “Pseudo-Classicisms in Late Biblical Hebrew,” in Sirach, Scrolls and Sages (ed. T. E. Muraoka and J. Elwolde; Leiden: Brill, 1999), 146–159; and see his essay in this volume, “Diachronic Linguistics and the Date of the Pentateuch.” 8  For one good example of infusion of late language into a manuscript (4QCantb), see I. Young, “Notes on the Language of 4QCantb,” JJS 52 (2001), 122–131.

90

David M. Carr

In sum, frequent fluid transmission of biblical texts, particularly of non-Torah texts in the Second Temple period, means that study of linguistic features may be more helpful in identifying relatively early texts (by the relative absence of late linguistic features, despite fluid transmission) than in identifying late texts (by the presence of linguistic features diverging from classical biblical Hebrew, of potentially diverse origins). To be sure, it appears that the Pentateuch was preserved more stably than other Hebrew Bible traditions and thus may have been relatively protected from linguistic contamination in transmission. Nevertheless, the Pentateuchʼs general linguistic profile, despite occasional possible late linguistic features here and there (themselves possible reflections of episodic linguistic contamination), does not conform well with a dating of large swaths of the pentateuchal text well into the Persian or Hellenistic periods.

Secondary Scribal Coordination Much ancient textual fluidity was not random. Some of the best-documented tendencies of scribal processing of earlier traditions are gradual blending, harmonization, and other forms of coordination of compositional elements, whether of elements secondarily combined through conflation or of different episodes within a unified story. Usually scholars have classified such scribal operations as harmonization, but Zahn has done a good job of problematizing this somewhat broad characterization.9 Scribes coordinated written traditions with each other in a variety of ways, from copying material from another location in order to conform one textual locus with another, to other sorts of revisions (such as adding characters thought to be missing [e.g., Aaron]). Secondary scribal coordination is a phrase that better encompasses the variety of such scribal interventions than the oft-used harmonization. One thing that emerges from the study of secondary scribal coordination is that ancient scribes typically engaged in one of two types of coordination. First, we see a number of cases of harmonization of closely parallel documents with each other. For example, the multiple manuscript traditions for Samuel–Kings and Chronicles, along with the ketiv-(versus often more fluidly preserved) qere tradition for those books, document ongoing coordination of the parallel reports in those books.10 The other major type of scribal coordination is the accommodation of one episode in a given literary work to another episode in the  Zahn, Rewritten Scripture (see n. 2), esp. 17–19.  S. L. M c K enzie , The Chroniclerʼs Use of the Deuteronomistic History (HSM 33; Atlanta: Scholars, 1984), esp. 94, 152–153, 160; Y. Z akovitch , “Assimilation in Biblical Narratives,” in Empirical Models for Biblical Criticism (ed. J. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 175–196; S. J aphet , 1 and 2 Chronicles (OTL; Louisville: Westminster John Knox, 1993), 162; Carr, Formation of the Hebrew Bible (see n. 1), 96–97. 9

10

Data to Inform Ongoing Debates about the Formation of the Pentateuch

91

same literary work. Jerrold Cooperʼs classic early study of Gilgameshʼs dreams, for example, showed how these dreams within the same epic were harmonized in later editions of that epic.11 Likewise, Hans Jürgen Tertel found several examples of similar secondary scribal coordination in the Atrahasis, Etana, and Anzu epics.12 These and other examples document that secondary coordination of different parts of a literary composition with each other was one of the most common types of scribal revision in the ancient Near East. The manuscript traditions for the Hebrew Bible confirm this conclusion. Hermann-Josef Stipp and others have documented similar harmonizing interventions of different parts of Jeremiah in its proto-MT edition.13 Juha Pakkala has argued for the presence of several proto-MT pluses that eliminate a pre-Solomon temple in 2 Sam 5:8, 7:11, and 15:25, along with 2 Sam 7:15–16. These pluses appear to have aimed to coordinate earlier materials that mentioned a temple associated with David with the narratives about Solomon building a temple in 1 Kgs 5:15–8:66.14 These examples join a long and expanding list of scribal coordination of different parts of the Pentateuch with each other, interventions documented in the early manuscript traditions for the Pentateuch (LXX, Samaritan, proto-MT, etc.) and semipentateuchal compositions (e.g., 4QRP). 15 In the case of pentateuchal and semipentateuchal compositions, a substantial proportion of additions appear to be focused on bridging the divide between There are even more examples of secondary coordination in the Greek manuscript tradition for the Christian Gospels, admittedly a chronologically distant comparison point. 11  J. Cooper, “Gilgamesh Dreams of Enkidu: The Evolution and Dilution of Narrative,” in Essays on the Ancient Near East in Memory of Jacob Joel Finkelstein (ed. M. Ellis; Hamden, CT: Archon Books, 1977), 39–44. 12  H. J. Tertel, Text and Transmission: An Empirical Model for the Literary Development of Old Testament Narratives (Berlin: de Gruyter, 1994), 33–36. 13  H.-J. S tipp , Das masoretische und alexandrinische Sondergut des Jeremiabuches: Textgeschichtlicher Rang, Eigenarten, Triebkräfte (OBO 136; Freiburg: Vandenhoeck & Ruprecht, 1994), esp. 137–140 and 146–151. 14  Pakkala, Godʼs Word Omitted (see n. 2), 214–220. 15  L. P rijs , Jüdische Tradition in der Septuaginta (Leiden: Brill, 1948), 93–99; E. Tov , “The Nature and Background of Harmonizations in Biblical Manuscripts,” JSOT 31 (1985), 3–29; idem, “Rewritten Bible Compositions and Biblical Manuscripts, with Special Attention to the Samaritan Pentateuch,” DSD 5 (1998), 334–354; idem, “Textual Harmonizations in the Ancient Texts of Deuteronomy,” in Mishneh Todah: Studies in Deuteronomy and Its Cultural Environment in Honor of Jeffrey H. Tigay (ed. N. S. Fox et al.; Winona Lake, IN: Eisenbrauns, 2009), 15–28; R. Hendel, The Text of Genesis 1–11: Textual Studies and Critical Edition (New York: Oxford University Press, 1998), esp. 36–38, 49–56, 75, 85–92; E. Eshel, “Harmonistic Editing in the Pentateuch in the Second Temple Period [Hebrew],” (masters thesis, Hebrew University, 1999); idem, with H. Eshel, “Dating the Samaritan Pentateuchʼs Compilation in Light of the Qumran Biblical Scrolls,” in Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls in Honor of Emanuel Tov (ed. S. M. Paul et al.; VTSup 94; Leiden: Brill, 2003), 215–240, and Carr, Formation of the Hebrew Bible (see n. 1), 91–95.

92

David M. Carr

reports of events in the Moses story in Exodus and Numbers on the one hand and back-references to events in the Moses story in Deuteronomy on the other. One thing that unites this list of coordinating pluses documented in diverse manuscript traditions for the Bible is that they generally modify one part of a given biblical composition so that it coordinates with another part of what is understood to be the same composition. They are innercompositional coordinating revisions. As such, they are important indicators of the evolving composition consciousness of the scribes who introduced them. The innercompositional character of these revisions gains significance because it points to scribal efforts to bridge nonoverlapping compositions that have been combined into a new literary whole. The most prominent example of innercompositional coordinating revision is the above-referenced extensive set of documented revisions conforming Exodus and Numbers to Deuteronomy, revisions that are particularly prominent in (though not exclusive to) proto-Samaritan and similar manuscript traditions. These secondary scribal attempts to bridge the Tetrateuch and Deuteronomy16 are among the important data supporting the idea that Deuteronomy was not originally written as a continuation of the preceding materials. The secondary scribal accommodations of tetrateuchal materials to Deuteronomy (and, less often, vice versa)17 are examples of the sort of gradual scribal revision that can occur when divergent literary works are brought together. Once together, these works then become subject to innercompositional scribal coordination, a coordination that reflects an emergent scribal composition consciousness that Deuteronomy and Exodus–Numbers are now part of one literary work. There are other possible marks of secondary scribal coordination attesting to the grouping of originally unconnected works. The material anticipating the exodus in Gen 15:13–16, for example, marked by resumptive repetition in 15:12 and 17, is a good candidate for such late coordination of major parts of the Pentateuch with each other (this time the ancestral narratives of Genesis with the Moses story that now follows them). In this case, the secondary character of the coordination is not documented by divergent textual traditions and has been so artfully done that some would dispute whether Gen 15:13–16 is secondary at

16  J. W ellhausen , Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (4th ed.; Berlin: de Gruyter, 1963 [1st ed., 1876]), 193; E. B lum , “Pentateuch-Hexateuch-Enneateuch? Oder: Woran erkennt Man ein literarischer Werk in der Hebräischen Bibel?,” in Les Dernières Rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (ed. T. Römer and K. Schmid; Leuven: Leuven University Press, 2007), 67–97; Carr, Formation of the Hebrew Bible (see n. 1), 256 (incl. n. 10). 17  On this tendency toward coordination of narrations of episodes with reviews of those episodes occurring later in the same text, rather than the reverse, see Tov, “Nature and Background of Harmonizations” (see n. 15), 8, and C arr , Formation of the Hebrew Bible (see n. 1), 92 (incl. n. 97).

Data to Inform Ongoing Debates about the Formation of the Pentateuch

93

all.18 Nevertheless, the exceptional character of the forward reference to Exodus in Gen 15:13–16 and the resumptive repetition in 15:12 and 17 are substantial reasons to hypothesize that Gen 15:13–16 is a secondary scribal attempt to link Abraham and his heirs with a Moses-exodus narrative that was not initially considered the sequel to the ancestral stories of Genesis.19 Again, this sort of secondary revision was prompted by these ancestral and Moses-exodus narratives beginning to be considered part of the same overall literary work and likely being transmitted (at least for a time) on the same written medium (probably a scroll).20 We see a similar likely scribal coordination of the Moses story with the ancestral narrative in several secondary additions to non-P Moses story materials that explicitly refer back to the promise of the land, by oath – e.g., Exod 32:13 (as part of added 32:10–14; including a reference to Abraham, Isaac and Jacob); Num 11:12 (as part of added Num 11:11–12, 14–17, 24b–30); and Num 14:16, 23 (as part of added Num 14:11–25*).21 As in the case of Gen 15:13–16, these potential secondary scribal additions, each of which coordinates the ancestral narrative with the Moses-exodus account, are not documented through divergent manuscript traditions. Nevertheless, these likely scribal additions in the Moses story conform to documented patterns elsewhere of scribal coordination of different episodes in the same literary work, including – in the case of Deuter18  J. H a , Genesis 15: A Theological Compendium of Pentateuchal History (BZAW 181; Berlin: de Gruyter, 1989), 39–62 (esp. 52–56 dealing with Gen 15:13–16 and the links between 15:12 and 15:17); K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 175–176; B. Ziemer, AbramAbraham: Kompositionsgeschichtliche Untersuchungen zu Genesis (BZAW 350; Berlin: de Gruyter, 2005), 174. 19  For a thorough synthesis of various arguments against the secondary character of Gen 15:13–16 and for the post-Priestly character of Gen 15 as a whole, see S chmid , Erzväter und Exodus (see n. 18), 172–186; English translation, Genesis and the Moses Story (trans. James Nogalski; Winona Lake, IN: Eisenbrauns, 2009), 158–171. I respond at more length to these and other arguments in a forthcoming article, D. M. Carr, “Strong and Weak Cases and Criteria for Establishing the Post-Priestly Character of Hexateuchal Material,” in The PostPriestly Pentateuch (ed. F. Giuntoli and K. Schmid; FAT 101; Tübingen: Mohr Siebeck, 2015), 19–34. For a recent synthesis of arguments for the secondary character of Gen 15:13–16, see M. Köckert, “Gen 15: Vom ‘Urgestein der Väterüberlieferung’ zum theologischen Programmtext der späten Perserzeit,” ZAW 125 (2013), 39–40. 20  In Carr, Formation of the Hebrew Bible (see n. 1), 275–277, I argue that this linkage occurred in a very late, “post-D” stage of formation of non-P materials prior to their combination with P. For other perspectives, see J. C. G ertz et al., eds., Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (BZAW 315; Berlin: de Gruyter, 2002), and T. B. Dozeman and K. Schmid (eds.), A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (SBLSymS 34; Atlanta: Society of Biblical Literature, 2006). 21  I discuss these and other examples further in Carr, Formation of the Hebrew Bible (see n. 1), 260–268, 275.

94

David M. Carr

onomy and preceding books – coordination of episodes in works that were once separate and were then secondarily brought together into a larger literary work.

Priestification – A Particular Type of Late-Persian/ Hellenistic Revision The above-discussed secondary scribal coordinations often have a somewhat mechanical feel to them. Though there certainly are theological and ideological dimensions to the probable additions mentioned above, such as Gen 15:13–16, Exod 32:10–14 and Num 14:11–21, many of these scribal coordinations merely copy material from one part of the Pentateuch, say, Mosesʼs mention of defeating Og of Bashan in Deut 3:1–3, into another part of the Pentateuch, in this case Num 21:33–35.22 Yet there are a number of documented and probable additions to the Hexateuch that adapt non-Priestly materials, often seamlessly, in Priestly directions. By this I mean a tendency of middle–Second Temple tradents to creatively build on and beyond Priestly precedents in transmitting and modifying non-Priestly traditions. Non-Priestly stories receive what might be described as a “Priestly wash,” a term that indicates the often subtle and even undetectable character of many such P-like scribal interventions. This is, of course, a phenomenon already well documented in the often P-like Chronistic expansion and revision of a non-Priestly composition about Judah and Israelʼs kings close (but not identical) in contents to large sections of our present books of Samuel–Kings.23 As Julius Wellhausen classically demonstrated, the whole of Chronicles can be viewed as creatively priestifying the Samuel–Kings narrative. Note, for example, in the parallels below how the tent of meeting and bronze altar are added by the Chronicler to the 1 Kgs 3 story of Solomonʼs sacrifice at Gibeon.

 Discussed in Carr, Formation of the Hebrew Bible (see n. 1), 260.  By now I and others are clear that Chronicles did not work with the proto-MT of Samuel–Kings nor any other exact literary witness presently available. Earlier, in D. M. Carr, “Empirische Perspektiven auf das deuteronomistische Geschichtswerk,” in Das deuteronomistische Geschichtswerk: Redaktions- und religionsgeschichtliche Perspektiven zur ‘Deuteronomismus’-Diskussion in Tora und vorderen Propheten (ed. M. Witte et al.; BZAW 365; Berlin: de Gruyter, 2006), 1–17, I played with a model of “shared source” advocated by Auld (e.g., A. G. Auld, Kings without Privilege: David and Moses in the Story of the Bibleʼs Kings [Edinburgh: T&T Clark, 1994] and many other works), where Chronicles might have been based on a source more like the shared material between Samuel–Kings and Chronicles (an Old Testament “Q”) rather than a composition resembling Samuel–Kings. Nevertheless, further detailed study of omissions in Chronicles led me to withdraw much of my support for that model (see Carr, Formation of the Hebrew Bible [see n. 1], 73–75). 22

23

‫‪95‬‬

‫‪Data to Inform Ongoing Debates about the Formation of the Pentateuch‬‬

‫‪1 Kings 3:4‬‬ ‫וילך המלך‬ ‫גבענה לזבח שם‬ ‫כי היא הבמה הגדולה‬

‫אלף עלות יעלה שלמה על המזבח ההוא‬

‫‪2 Chronicles 1:3–6‬‬ ‫וילכו שלמה וכל הקהל עמו‬ ‫לבמה אשר בגבענה‬ ‫כי שם היה אהל מועד האלהים אשר עשה‬ ‫משה עבד יהוה במדבר אבל ארון האלהים‬ ‫העלה דויד מקרית יערים בהכין לו דויד‬ ‫כי נטה לו אהל בירושלם ומזבח הנחשת‬ ‫אשר עשה בצלאל בן אורי בן חור שם‬ ‫לפני משכן יהוה וידרשהו שלמה והקהל‬ ‫ויעל שלמה שם על מזבח הנחשת לפני‬ ‫יהוה אשר לאהל מועד‬ ‫ויעל עליו עלות אלף‬

‫‪Similarly, the full array of properly dressed priests and Levites are specified by‬‬ ‫‪the Chronicler as part of the assembly prior to the divine cloud filling the temple‬‬ ‫‪and Solomonʼs dedicatory prayer.‬‬ ‫‪1 Kings 8:10‬‬ ‫ויהי בצאת הכהנים מן הקדש‬

‫והענן מלא את בית יהוה‬

‫‪2 Chronicles 5:11–13‬‬ ‫ויהי בצאת הכהנים מן הקדש‬ ‫כי כל הכהנים הנמצאים התקדשו אין לשמור‬ ‫למחלקות והלוים המשררים לכלם לאסף‬ ‫להימן לידתון ולבניהם ולאחיהם מלבשים‬ ‫בוץ במצלתים ובנבלים וכנרות עמדים‬ ‫מזרח למזבח ועמהם כהנים למאה ועשרים‬ ‫מחצררים בחצצרות ויהי כאחד למחצצרים‬ ‫ולמשררים להשמיע קול אחד להלל ולהדות‬ ‫ליהוה וכהרים קול בחצצרות ובמצלתים ובכלי‬ ‫השיר ובהלל ליהוה כי טוב כי לעולם חסדו‬ ‫והבית מלא ענן בית יהוה‬

‫‪Similarly, Josiahʼs singular Passover in 2 Kgs 23:21–23 is given a P cast through‬‬ ‫‪several modifications in 2 Chr 35.‬‬ ‫‪2 Kings 23:21–23‬‬ ‫ויצו המלך את כל העם לאמר‬ ‫עשו פסח ליהוה אלהיכם‬ ‫ככתוב על ספר הברית הזה‬

‫‪2 Chronicles 35:1–2, 17–19‬‬ ‫ויעש יאשיהו בירושלם פסח ליהוה‬ ‫וישחטו הפסח בארבעה עשר לחדש‬ ‫הראשון‬ ‫ויעמד הכהנים על משמרותם ויחזקם‬ ‫לעבודת בית יהוה‬ ‫]‪[large plus Priestly Passover 35:3–16‬‬

96

David M. Carr

2 Chronicles 35:1–2, 17–19 (cont.) ‫ויעשו בני ישראל הנמצאים את הפסח‬ ‫בעת ההיא ואת חג המצות שבעת ימים‬ ‫ולא נעשה פסח כמהו בישראל‬ ‫מימי שמואל הנביא‬ ‫וכל מלכי ישראל‬ ‫לא עשו כפסח אשר עשה יאשיהו‬ ‫והכהנים והלוים וכל יהודה וישראל‬ ‫הנמצא ויושבי ירושלם‬ ‫בשמונה עשרה שנה למלכות יאשיהו‬ ‫נעשה הפסח הזה‬

2 Kings 23:21–23 (cont.)

‫כי לא נעשה כפסח הזה‬ ‫מימי השפטים אשר שפטו את ישראל‬ ‫וכל ימי מלכי ישראל ומלכי יהודה‬

‫כי אם בשמנה עשרה שנה למלך יאשיהו‬ ‫נעשה הפסח הזה ליהוה בירושלם‬

These are just three concrete examples of a much broader phenomenon of Chronistic priestification of the Samuel–Kings narrative alongside numerous other shifts reflecting the Chroniclerʼs specific post-Priestly ideology.24 This overall orientation toward the Priestly tradition is also evident in the later stages of the transmission of the Pentateuch. We see it, for example, in the multiple late levels of harmonization of the Priestly tabernacle instruction and construction accounts, the likely addition of Priestly elements of the spy story (Num 14:31) to Deut 1:39 (elements missing in the LXX), or the addition of the tabernacle to Joshuaʼs covenant in a late plus to the LXX of Josh 24:25. Some of this priestification, to be sure, represents ongoing harmonization of P and non-P strands in the Second Temple period, a well-documented trend of scribal coordination that – as mentioned above – also combines Deuteronomic and tetrateuchal traditions and other once-separate tradition streams. Nevertheless, there seems to be an overall tendency in the Second Temple period to privilege Priestly figures and Priestly conceptuality, a tendency that corresponds well to the fact that Hebrew literary textuality during this period was connected particularly to Second Temple Priestly groups, whether ones connected to the Jerusalem temple or groups outside that complex. These examples of priestification, both in Chronicles and in divergences in early manuscript traditions for the Hexateuch, show not only that Second Temple scribes often added a Priestly cast to non-Priestly texts but also that some of these documented changes would have been largely undetectable if we could 24  By post-Priestly I mean that the Chroniclerʼs ideology is not Priestly itself but presupposes it. For more examples and specification, the discussion in Wellhausenʼs Prolegomena remains a classic. See J. Wellhausen, Geschichte Israels: Erster Band – Prolegomena zur Geschichte Israels (Berlin: G. Reimer, 1883), 166–223; English translation, Prolegomena to the History of Ancient Israel (trans. J. Sutherland Black and Allan Menzies; Cleveland: Meridian Books, 1957), 171–227, which argues that the difference between Samuel–Kings and Chronicles a few hundred years later “beruht auf dem inzwischen eingetretenen Einfluß des Priesterkodex” (“is caused by the intervening influence of the Priestly source,” 166). For specific discussions, see in particular 177–180, 184–185, and 207.

Data to Inform Ongoing Debates about the Formation of the Pentateuch

97

not compare Chronicles with Samuel–Kings or early manuscript traditions with each other. Sometimes, of course, there are indicators of secondary growth such as the resumptive repetition marking Chroniclesʼ addition to Solomonʼs second epiphany in 2 Chr 8, the doubling of children in Deut 1:39, or the extended ‫כי‬ clause interrupting the movement from the priests going out in 2 Chr 5:11a and the cloud filling the sanctuary in 5:13b. I would argue that we have similar slight indicators of undocumented secondary harmonization in the temporary deviation to third-person language in Josh 24:6–7aα. This mention of the Reed Sea event (marked below in square brackets) is marked as a likely secondary addition by its duplication of Josh 24:5 and the divergence of this portion of Joshuaʼs review from its context in speaking about Israelʼs exodus generation as “your fathers” (24:6) rather than just as “you” (24:5, 8, 9, 10, 11, etc.). Joshua 24:5–7

*‫ואשלח את משה ואת אהרן‬ †‫ואגף את מצרים כאשר עשיתי בקרבו ואחר הוצאתי אתכם‬ ‫[ואוציא את אבותיכם ממצרים ותבאו הימה וירדפו מצרים אחרי אבותיכם ברכב ובפרשים ים סוף‬ ‫ויצעקו אל יהוה וישם מאפל ביניכם ובין המצרים ויבא עליו את הים ויכסהו] ותראינה עיניכם את‬ ‡‫אשר עשיתי במצרים ותשבו במדבר ימים רבים‬ * missing in LXX † note 2nd person plural here and surrounding ‡ note 2nd person plural again

In other cases, such as the above-discussed addition of multiple elements to Solomonʼs dream at Gibeon, the probable addition of Aaron and Moses to Josh 24:5a in a plus absent from the LXX, and the addition of the tent of meeting to the end of Josh 24:25 in the Old Greek, we would have little to go on to identify documented priestifying additions to the Hexateuch if we did not have the divergent manuscript traditions. In the case given above of Josiahʼs Passover in 2 Chr 35, we could tell that additions had happened, but the indicators might mislead us as to exactly what happened.

Use of Works Like Ezra–Nehemiah and Chronicles to Determine the Earlier Scope of Literary Works This discussion takes us to treatment of the scope of works used by Ezra–Nehemiah and Chronicles and what this can tell us about models for the development of the Pentateuch. Notably, both Ezra–Nehemiah and Chronicles have clear connections not only to the Pentateuch but also to a Hexateuch that includes Joshua. The Ezra narrative models the return of the golah community on the entry of the Israelites into Canaan and in the process repeatedly invokes the entry into the land and lists of displaced peoples that are part of the Joshua narrative. Chronicles has a complex but clearly demonstrable relationship with

98

David M. Carr

parts of Joshua, especially 1 Chr 1–9. In this sense, both Ezra–Nehemiah and Chronicles betray a Hexateuchal orientation that contrasts with the resolutely pentateuchal focus of later Second Temple works such as Sirach, Wisdom of Solomon, Philo, etc. When it comes to questions of formation, I am ever more firmly convinced that the Hexateuch is the proper scope of our discussion and not the Pentateuch. This is confirmed, incidentally, by the fact that we see P elements in the book of Joshua, including a P redactional element at its very end, but no clear P elements or priestification beyond that point. Whatever work the Priestly tradents worked with, it did not extend beyond Joshua. Notably, the clearest links connecting Ezra–Nehemiah and Chronicles to the first books of the Hebrew Bible extend only through Joshua and do not clearly include Judges. To be sure, 1 Chr 17:10 preserves a reference from 2 Sam 7:11 to a premonarchal time of the “judges,” but this hardly requires the specific book of Judges that we now have. In addition, though some have seen echoes of the judges period in the core of the speech of the prophet Azariah to King Asa that refers to a time when Israel lacked a “true God, a priest or a torah” (2 Chr 15:3–6), this speech, built around echoes of Hosea and Zechariah, does not conform well to our existing book of Judges.25 It speaks of an overall lack in Israel of a “true God,” international unrest affecting all countries, and nations crushing nations. Compare this in the parallel chart below with the much clearer reflection of Judges in Neh 9, one of the last layers of Ezra–Nehemiah,26 which echoes language in Judges. God is described as delivering (‫ )נתן ביד‬the Israelites into the hand of enemies, they “cry” out to God, and God gives them deliverers (verbal resonances between Judges and Neh 9:26–28 are underlined in the table below): Judges 2:12–14, 16; 3:9

Nehemiah 9:26–28

‫ויעזבו את יהוה אלהי אבותם‬ ‫המוציא אותם מארץ מצרים‬ ‫וילכו אחרי אלהים אחרים מאלהי‬ ‫העמים אשר סביבותיהם וישתחוו‬ ‫להם ויכעסו את יהוה ויעזבו את‬ ‫יהוה ויעבדו לבעל ולעשתרות‬ ‫ויחר אף יהוה בישראל‬

‫וימרו וימרדו בך וישלכו את‬ ‫תורתך אחרי גום ואת נביאיך‬ ‫הרגו אשר העידו בם להשיבם‬ ‫אליך ויעשו נאצות גדולת‬

2 Chronicles 15:3–5 ‫וימים רבים לישראל‬ ‫ללא אלהי אמת וללא כהן‬ ‫מורה וללא תורה‬ ‫וישב בצר לו על יהוה‬ ‫אלהי ישראל ויבקשהו‬ ‫וימצא להם‬

 Sara Japhet, who was very helpful in helping with references for this part of the essay, notes in her commentary (Japhet, 1 and 2 Chronicles [see n. 10], 718) that this text in 2 Chr 15 represents “what may be the closest the Chronicler comes to a ‘historical review.’ ” 26  On the especially late character of Neh 9 and related materials, see D. B öhler , Die heilige Stadt in Esdras α und Esra-Nehemia: Zwei Konzeptionen der Wiederherstellung Israels (OBO 158; Göttingen: Vandenhoeck & Ruprecht, 1997), esp. 385–391. 25

Data to Inform Ongoing Debates about the Formation of the Pentateuch

Judges 2:12–14, 16; 3:9 (cont.)

Nehemiah 9:26–28 (cont.)

‫ויתנם ביד שסים וישסו אותם‬ ‫וימכרם ביד אויביהם מסביב ולא‬ . . . ‫יכלו עוד לעמד לפני אויביהם‬ ‫ויקם יהוה שפטים ויושיעום מיד‬ ‫שסיהם‬

‫ותתנם ביד צריהם ויצר להם‬ ‫ובעת צרתם יצעקו אליך‬ ‫ואתה משמים תשמע וכרחמיך‬ ‫הרבים תתן להם מושיעים‬ ‫ויושיעום מיד צריהם‬ ‫וכנוח להם ישובו לעשות רע‬ ‫לפניך ותעזבם ביד איביהם‬ ‫וירדו בהם וישובו ויזעקוך‬ ‫ואתה משמים תשמע ותצילם‬ ‫כרחמיך רבות עתים‬

‫ויזעקו בני ישראל אל יהוה ויקם‬ ‫יהוה מושיע לבני ישראל ויושיעם‬ ‫את עתניאל בן קנז אחי כלב‬ ‫הקטן ממנו‬

99

2 Chronicles 15:3–5 (cont.) ‫ובעתים ההם אין שלום‬ ‫ליוצא ולבא כי מהומת‬ ‫רבות על כל יושבי‬ ‫הארצות‬

The language is not precisely the same in Judges and Neh 9, but there are enough terminological connections between them and the conceptuality of cycles is similar enough in both passages that it seems quite likely that Neh 9 has the book of Judges in view. No such clear evidence is present in 2 Chr 15. Finally, the genealogies in 1 Chr 1–9 lack a single clear instance of dependence on data found in Judges. Chronicles mentions some figures seen also in Judges, e.g., Ehud in 1 Chr 7:18, without any apparent awareness of material about him in Judges and offers competing pictures of the genealogies of others, such as Jair son of Segub (1 Chr 2:21–22), again without any evident critique of or integration with competing pictures of the same figureʼs genealogy in Judg 10:3–5. In one case, Ralph Klein even suggests dependence of Judges on Chronicles, arguing that a minor judge, Tolah son of Puah, in Judg 10:1, “owes his existence” to a creative rereading of the genealogy of Issacharʼs sons in Num 26:23–24 along with 1 Chr 7:1.27 To be sure, one might read this evidence as suggesting a conscious “silencing” of the book of Judges in Chronicles, perhaps because of concerns about the negative picture of Israel there, but then we might expect clearer links to the Judgesʼ depiction of the premonarchic period in the negative picture of Israelʼs premonarchic period in the Chronicles version of Azariahʼs prophecy. Moreover, we might also expect Chronicles to at least occasionally connect with genealogical data found in Judges as source material for its own genealogies. As it is, however, there is little to indicate that Chronicles has any relationship with our book of Judges. In sum, Chronicles can be seen, in part, as a rewriting of Samuel–Kings from the perspective of a priestified Hexateuch (non-P material reframed by P material), including adding a prologue (1Chr 1–9) that draws genealogical and other information from the P and non-P parts of the Hexateuch to the Samuel–Kings 27

218.

 R. W. Klein, 1 Chronicles: A Commentary (Hermeneia; Minneapolis: Fortress, 2006),

100

David M. Carr

story. Linking back to the theme of priestification discussed above, we see such priestification take several different forms in the works under discussion here: we only possess a Hexateuch in priestified form – its non-Priestly sources are only accessible by way of literary criticism. Chronicles represents a priestified form of Samuel–Kings that now stands alongside a version of its non-P Samuel– Kings source. Judges does not seem to have undergone such priestification at all, containing a likely early Ephraimite core with a much later Judahite frame. The relatively early core of Judges would explain its basic classical Hebrew profile, i.e., the fact that it – so far as I know – lacks many linguistic features that would identify the book as a late text overall. Nevertheless, I am suggesting that the book as we presently have it, framed as part of a larger enneateuchal sequence, is quite late. The present book of Judges likely postdates the priestification of Israelite narrative literature (both the Hexateuch and Samuel–Kings), is not reflected clearly in Chronicles, and is only reflected in one of the latest layers of Ezra–Nehemiah, namely Neh 9. An additional observation: though Chronicles certainly knows a form of Samuel–Kings and draws on the P and non-P Hexateuch, it treats Samuel–Kings and the Hexateuch in decisively different ways, rewriting one work while drawing on the other one. In this sense, Chronicles might be seen as a model for Deuteronomyʼs quite different orientation to the Moses-wilderness and ancestral traditions. Whereas Deuteronomy extensively reviews materials from the non-P Moses-wilderness story, it only glancingly refers to Genesis by way of references to the divine oath promise of land to ancestors (e.g., Deut 1:8, 35; 6:10, 18, 23; 7:8, 12; with Abraham, Isaac and Jacob mentioned in 30:20 and [in reference to becoming a God to them] 29:12). Even if some of these references to the promise in Deuteronomy are early,28 the general character of Deuteronomyʼs references to the promise may point to the fact that the ancestral story of Genesis is still understood in layers of Deuteronomy to be a separate work different in character from the Moses story that Deuteronomy more extensively reviews. In sum: as Chronicles is to the Hexateuch versus Samuel–Kings, so Deuteronomy may be to the ancestral story versus the Moses story.

Partial Preservation of Precursor Materials and Other Problems for Reconstruction One thing to which my study, Pakkalaʼs work, and Pakkala and Müllerʼs work point in particular is the way that documented cases of transmission history only 28  For the essential parameters of the debate, see T. Römer, Israels Väter: Untersuchungen zur Väterthematik im Deuteronomium und in der Deuteronomistischen Tradition (OBO 99; Freiburg: Universitätsverlag, 1990) on the one hand and N. L ohfink , Die Väter Israels im Deuteronomium (OBO 111; Freiburg: Universitätsverlag, 1991) on the other.

Data to Inform Ongoing Debates about the Formation of the Pentateuch

101

inconsistently preserve in later forms of biblical texts data that would be crucial to the reconstruction of earlier stages. Sometimes, of course, documented cases of transmission history include cases where the final form of a given text preserves marks, such as resumptive repetition, of the secondary character of a given element in only one manuscript stream. In other cases, such as probable priestifying additions in Josh 24:5 or 25, it is harder to argue that the final form of the priestified text would have allowed a scholar lacking other manuscript forms to reconstruct the exact contours of later scribal revision. Elsewhere I have argued that certain types of scribal work, such as the combination of originally separate texts, are more easily recognized than the more broadly documented phenomenon of scribal expansion-adaptation (because of cross-currents produced by the different profiles of the originally independent texts).29 But even here analysis of documented cases of combinations of separate texts should give scholars pause about the level of precision that is achievable in methodologically controlled, cross-school reconstruction of probable stages leading to the present Pentateuch or Hexateuch. In particular, as pointed out in parts of my own work and in an entire book by Pakkala, scribes often omitted large portions of their precursor documents.30 Deuteronomy only preserves themes and a few words from the Covenant Code, to which it responds in part. Chronicles does not preserve all of whatever Samuel–Kings source it used. The Temple Scroll quite incompletely draws on elements of its precursor Torah traditions. The authors of Matthew and Luke only preserved portions of their probable Markan and Q source documents. These and other documented examples from the ancient Near East show that ancient scribes did not preserve complete versions of their precursor documents except in cases where they were reproducing basically unchanged exemplars of those precursor documents. When they were drawing on a document as a “source,” they drew on it selectively. Arguments to the contrary seem based more on methodological wishful thinking (the hope that [undocumented] source documents might be fully reconstructable) than on reliable evidence from ancient Israel and surrounding scribal contexts. This takes us to a particular type of argument that has an old pedigree but has been used increasingly over recent years and has contributed unhelpfully, I believe, to a plethora of pentateuchal theories of all kinds. I term these “arguments from potential continuity.” Sometimes such arguments relate to places where a distinctive text-block truly interrupts the flow of a text. Joel Baden mentions, for instance, the example of how Exod 24:3–8 interrupts the progression from 24:1–2 to 24:9–11.31 More often, however, there is no real textual discontinuity  Carr, Formation of the Hebrew Bible (see n. 1), 105, 145–147.  Carr, Formation of the Hebrew Bible (see n. 1), 111–115, and (on the topic of loss of crucial indicators through omission) 146–147; Pakkala, Godʼs Word Omitted (see n. 2). 31  J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven, CT: Yale University Press, 2012), 19. 29

30

102

David M. Carr

in evidence, but a scholar simply notes that a text seems to read as well without a given text block as with it. The problem with reliance on such arguments is that they are so easily made in so many different ways, as can easily be shown through a history of pentateuchal scholarship, especially recent pentateuchal scholarship. Once you set a group of smart people loose on a text, they will find many ways it would read better if this or that section was omitted. If the group who heard this presentation in Jerusalem was set loose on this essay, I have no doubt that many could find supposed secondary sections, the elimination of which would make this essay both shorter and better. I would maintain that we all would be better served if such arguments from potential continuity were reduced to a minor and supporting role in broader cases based primarily on other criteria. Generally, arguments for potential continuity are notoriously unreliable. Show me a pentateuchal theory of any kind based heavily on arguments from discontinuity and the greater readability of a reconstructed text, and Iʼll show you a theory whose real impact will be limited to the advocate of that theory and their school. There is one case in which arguments from potential continuity can be helpful – in establishing a likelihood that a given stratum (established primarily on other grounds) once existed independently. For example, the relative readability of parallel P and non-P pentateuchal strata, reconstructed on the basis of other criteria, makes unlikely arguments that either the P layer (as argued by some particularly in the late twentieth century) or the non-P layer (as argued by many more recently) was composed, from the outset, as a redactional or compositional extension of earlier works. To be sure, this argument also has been taken too far in past scholarship. Martin Noth, for example, made the very influential argument that P was preserved virtually intact by the pentateuchal redactor “im allgemeinen das vollständige Erhaltensein des ursprüngliche Bestandes und daher ein lückenloser Zusammenhang der ausgeschiedenen Elemente.”32 This line of argument by Noth has misled several generations of later scholars into arguing that P was a “redaction” or compositional extension of non-P works, largely on the basis of possible gaps in P.33 In recent years this has been flipped and applied to non-Priestly  “[P]reserved in its original extent and therefore identified P elements connect seamlessly with each other.” Quote from p. 17 of broader discussion in M. Noth, Überlieferungsgeschichte des Pentateuch (2nd ed.; Darmstadt: Kohlhammer, 1960 [1st ed., 1948]), 11–17. 33  This focus on gaps in P as an argument for its redactional and/or compositional character was particularly prominent in the foundational arguments of F. M. Cross in Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), 306–307, 317–321, and had already appeared in brief a few years previously in C. M achholz , “Israel und das Land: Vorarbeiten zu einem Vergleich zwischen Priesterschrift und deuteronomistischem Geschichtswerk” (Habilitationsschrift, Heidelberg University, 1969), 38–39. The idea has been prominent in a number of subsequent studies as well. 32

Data to Inform Ongoing Debates about the Formation of the Pentateuch

103

materials, such as the argument that the non-P flood narrative is a redactional extension of P because the non-P flood narrative lacks elements like Godʼs order for Noah to build an ark.34 Yet we lack a single documented example of a conflator completely preserving the entire text of combined source documents. Our few potential examples of combination of prior documents all feature major omissions of source materials, indeed, omissions that surpass those posited for either P or non-P.35 Perhaps this exceptional character of preservation of P and non-P source materials (in comparison to most cases of conflation) contributed to Nothʼs and othersʼ perceptions that P was preserved so completely, even as this conception (of virtually perfect preservation of conflated source documents) has proven to be one of the most misleading ideas to be introduced into pentateuchal scholarship of the last century.

Concluding Reflection on Achieving More Methodological Control Given the skeptical tone of the last paragraphs, I turn finally to reflections on how we might achieve more methodological control, finding ways to create 34  This started in the mid-1990ʼs with new interpretations of long-observed gaps in J in the flood as evidence that J was an expansion of P. In subsequent years, this redactional approach to J has expanded to other arguments. See J. B lenkinsopp , The Pentateuch: An Introduction to the First Five Books of the Bible (ABRL; New York: Doubleday, 1992), 78; J. L. S ka , “El relato del diluvio: Un relato sacerdotal y algunos fragmento redaccionales posteriores,” EstBib 52 (1994), 52; D. L. P etersen , “The Formation of the Pentateuch,” in Old Testament Interpretation: Past, Present, and Future; Essays in Honor of Gene M. Tucker (ed. D. L. Petersen et al.; Nashville: Abingdon, 1995), 42–43; T. Krüger, “Das menschliche Herz und die Weisung Gottes: Elemente einer Diskussion über Möglichkeiten und Grenzen der Tora-Rezeption im Alten Testament,” in Rezeption und Auslegung im Alten Testament und in seinem Umfeld: Ein Symposium aus Anlass des 60. Geburtstags von Odil Hannes Steck (ed. R. G. Kratz and T. Krüger; OBO 153; Freiburg: Universitätsverlag, 1997), 74; R. G. Kratz, Die Komposition der erzählenden Bücher des Alten Testaments: Grundwissen der Bibelkritik (UTB 2137; Göttingen: Vandenhoeck & Ruprecht, 2000), 259. 35  Conflation is documented relatively rarely, especially if one excludes cases of combination of nonoverlapping prior documents, such as the integration of the flood narrative into the Standard Babylonian Gilgamesh Epic. The best example is probably still the conflation of canonical Gospels in Tatianʼs Diatessaron, a case distant culturally and chronologically from the formation of the Pentateuch, but there are smaller-scale examples of conflation (depending often on oneʼs interpretation of the evidence) in parts of the Temple Scroll and some 4QRP manuscripts, 2 Sam 16–18, Ezra–Nehemiah, and the so-called proto-Samaritan and other expansionist pentateuchal manuscripts. (Tov, “Source of Source Criticism” [see n. 4], 295, adds the example of the joining of Deut 27 and Josh 8:35–40 in m. Sotah 7 as well). The statements here about conflation are based on this range of near and far documented examples of the phenomenon.

104

David M. Carr

theories that will have a shelf life outlasting one or another particular school of pentateuchal scholarship. If some of our arguments from potential continuity, textual dependence,36 or terminology are relatively weak (at least on their own), where might we find stronger ones? Where would we find an audience of close readers of the Torah who demonstrably have no investment in one model or the other and would just read the text and say what parts of it are unintelligible as they stand? Where would we find such an impartial focus group? I have an idea, based on frequent use of Jewish biblical interpretation in historical-critical work by many Jewish colleagues already.37 I think we could find such a focus group among premodern Jewish commentators on the Torah, whether a later synthesizer of rabbinic interpretation like Rashi or pəšaṭ interpreters like Ibn Ezra, Rashbam, or Radak. There are, of course, close readers of the text who are not Jewish in the earlier history of the biblical text, and they could be useful too. Nevertheless, Christian readers of the Torah often have such different agendas as they work through the text that their interpretations provide less material, in my general experience, than do close readings by classical Jewish commentators. Jean Louis Skaʼs essay in a recent Genesis volume on Radakʼs identification of chronological problems in Genesis is the closest recent example to what I am proposing here,38 but I am suggesting something more global. Furthermore, what I am suggesting is somewhat distinct from the frequent past use of interpretive suggestions by Jewish commentators to deflect proposals about textual growth, e.g., the attempt by some to use the rabbinic distinction between ‫ יהוה‬as gracious and ‫ אלהים‬as judge to counter use of divine designation as an indicator of diverse sources. Rather, I am suggesting a broad survey of a clearly circumscribed commentary tradition, e.g., Rashiʼs synthesis of Jewish interpretation, to see where Rashi (or another early interpreter) attempts to interpret a textual feature that transmission-historical models would explain otherwise through particular diachronic models. I personally do not presume to be particularly well qualified for this sort of study. Nevertheless, my initial sounding, just surveying Rashiʼs commentary on Genesis, found many more examples of places where he struggled with features probably caused by joining P and non-P than I found of him struggling with features correlated with other proposed models for pentateuchal formation. To 36  This area was a feature of my 2013 contribution to the Jerusalem conference, but my primary concerns are already covered in Carr , Formation of the Hebrew Bible (see n. 1), 137–144. 37  Some prominent examples include Jacob Milgrom, Jeffrey Tigay, Nahum Sarna, Michael Fishbane, and Moshe Greenberg, but there are many others. I am indebted to Professor Robbie Harris of Jewish Theological Seminary for adding some precision to my comments in this section of the essay. 38  J. L. Ska, “The Study of the Book of Genesis: The Beginning of Critical Reading,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; Leiden: Brill, 2012), 3–26.

Data to Inform Ongoing Debates about the Formation of the Pentateuch

105

be sure, he does work to explain the problem of Midianites and Ishmaelites in Gen 37:27–28, 36 and 39:1, a problem that has been attributed to a conflation of J and E source documents or to scribal revision in Gen 37:28. So also, Rashi notes the parallel between well-digging and controversy between Isaac and Abimelek in Gen 26:18 (often attributed to J) and the earlier description of Abrahamʼs similar interactions with the same characters in Gen 21 (often attributed to E). In addition, Rashi notes a couple of features of the Joseph story – the apparent presupposition in Josephʼs dream that Rachel is still alive (comment on Gen 37:10) and that Jacob had many daughters (comment on Gen 37:35) – that do not square well with the biblical Jacob story that the Joseph story now follows. These latter indicators can be read as signs of the secondary combination of the Jacob and Joseph stories with each other. Finally, Rashi deals with various inner-Priestly issues, such as the divergent data about Esauʼs family in Gen 26:34, 28:9, and 36:2–3 or some repetitions in Gen 17 (comments on 17:13 and 19), and he notes in comments on Gen 38:1 and 39:1 the disruption caused by the apparent insertion of the Tamar story. All these can be linked with diverse theories about pentateuchal formation advocated today, from the J/E hypothesis for the formation of the non-Priestly Tetrateuch to what I now like to term the “sources and supplements” model for the joining of diverse written compositions and their expansion by compositional layers.39 Nevertheless, my initial survey suggests that Rashi struggles far more often with features that can be linked to the combination of P and non-P. He works on explaining, for example, elements of the divergent flood chronologies in Gen 8:3 and with the statements in Gen 6:19 and 7:9 that the animals were to enter two by two (P), rather than an extra six pairs of clean animals as specified in Gen 7:2 (the non-P flood narrative). He notes the strangeness of Isaac coming back to the tent of his mother in Gen 24:67 (a non-P text), when she has already died in the previous chapter (Gen 23; P). He also deals with numerous chronological issues in the ancestral narratives caused by combining P age notices with non-P narratives, such as the implication that Hagar carried a sixteen-year-old Ishmael on her shoulder (comment on Gen 21:24).

 I proposed this term “sources and supplements” originally in an oral comment made on November 25, 2013, in response to a panel discussion “Convergence and Divergence in Pentateuchal Theory” in the Pentateuch section of the annual meeting of the Society of Biblical Literature (Baltimore, MD). I thought and think this phrase better characterizes new alternative approaches to the Pentateuch than do the terms “European approach” or “non-Documentarian approach” sometimes used in recent discussions. The term sources here encompasses both a cross-pentateuchal P source and earlier (non-Priestly) compositions of more limited scope, say a Jacob or a Moses story. The term supplements in this phrase connects back to classical divisions of pentateuchal theories into fragment, source, or supplement approaches, even as many contemporary scholars would prefer to term such supplements redactions or compositional layers. 39

106

David M. Carr

This is just an initial sounding done by an admitted amateur at analysis of rabbinic commentary. There are many more qualified than I who could do this work better, combining thorough knowledge of contemporary pentateuchal approaches to texts like Genesis with fluency in interpretation of Rashi and other classical Jewish commentators. It is not, by the way, that I believe each pentateuchal theory requires confirmation in all details by Jewish or other commentators, and I believe it would be a mistake to start oneʼs transmission-historical work through working only with indicators mined from Rashi or another commentator. They were doing something quite different from study of pentateuchal formation. Nevertheless, precisely because these close readers of the biblical text were doing something different than transmission history, they can provide a good secondary focus group to check whether problems observed by contemporary scholars were ever seen by anyone else in centuries of interpretation. I suggest that a more systematic survey of commentators like Rashi might represent another sort of “empirical” approach that might help us measure how much diverse theories are supported by actual problems in the text that we did not just manufacture but that were noted by earlier readers who were not biased by one particular approach to diachronic study or another. The focus in such a survey would not be on the solutions offered by early Jewish commentators to the problems that they found, since the differences between Rashi, Radak, and other commentators in the tradition are often immense. Instead, the focus of such a survey would just be on which features of the text were deemed problematic enough by these commentators to deserve any kind of explanation, particularly features that are interpreted diachronically in various pentateuchal theories. My concluding point is more general. It is based both on years of poring over documented examples of transmission history and on decades of reading older pentateuchal scholarship and watching the field develop. If there is anything I have learned from this, it is the following: we know far less than we think we do about the formation of these texts. Put another way, I am ever more struck with just how fraught and difficult it is for us to know anything secure and detailed about the undocumented prehistory of any text. The field is littered with the carcasses of dead theories by once-prominent pentateuchal scholars, and I suspect that many theories advanced today will fare no better. I would not have attended the Jerusalem conferences and written this essay if I didnʼt think that we can form plausible theories about the Pentateuchʼs prehistory. Nevertheless, I think that all of us, whatever our models for pentateuchal formation, could use more humility about our reconstructive abilities and modesty about our respective research results.

Innerbiblical Exegesis The View from beyond the Bible Molly M. Zahn In the last quarter-century, the idea of “innerbiblical exegesis” has taken root widely in the study of the Pentateuch.1 For a field marked in recent times largely by the absence of consensus, innerbiblical exegesis constitutes a rare point of convergence between Israeli, European, and North American scholarship, even if the term sometimes means different things to different people. Whatever the precise compositional models involved, it now appears to be generally acknowledged that exegetical revision and reapplication of earlier textual traditions played a substantial role in the development of the Pentateuch as we know it. This recognition bears theoretical and methodological significance both for biblical studies and for the broader discipline of religious studies, insofar as it breaks down the old boundary between Scripture and tradition or Scripture and interpretation.2 As Levinson puts it, “Interpretation is constitutive of the canon; it is not secondary to the canon in terms of either chronology or significance.”3 1  In English-speaking scholarship, the term is usually associated with the groundbreaking work of M. F ishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985). As two recent bibliographical essays indicate, however, Fishbaneʼs work had important precedents, and a parallel movement in Europe (associated especially with the work of O. H. Steck and his students) has helped the phenomenon gain widespread recognition. See K. S chmid, “Innerbiblische Schriftauslegung: Aspekte der Forschungsgeschichte,” in Schriftauslegung in der Schrift (ed. R. G. Kratz et al.; BZAW 300; Berlin: de Gruyter, 2000), 1–22; B. M. Levinson, Legal Revision and Religious Renewal in Ancient Israel (Cambridge: Cambridge University Press, 2008), 95–181. 2  The dynamics of Scripture and interpretation remain somewhat understudied from a comparative perspective, despite the seminal works of W. A. G raham , Beyond the Written Word: Oral Aspects of Scripture in the History of Religion (Cambridge: Cambridge University Press, 1987) and W. C. S mith , What Is Scripture? A Comparative Approach (Minneapolis: Fortress, 1993). Nevertheless, some relatively recent publications from the broader field of religious studies do attend explicitly to the ways in which (versions of) scriptural texts can embed interpretation or in which interpretive texts can themselves become scriptural. See, for example, J. B. Henderson, Scripture, Canon, and Commentary: A Comparison of Confucian and Western Exegesis (Princeton: Princeton University Press, 1991); P. Richman (ed.), Many Ramayanas: The Diversity of a Narrative Tradition in South Asia (Berkeley: University of California Press, 1991). 3  Levinson, Legal Revision (see n. 1), 18.

108

Molly M. Zahn

Despite the considerable importance of the concept in crystallizing a new way of thinking about the development of scriptural texts, I would like to propose in what follows that it is time to move away from use of the term innerbiblical exegesis itself. My argument is rooted in new insights into scribal practice derived from the study of Second Temple materials, the Qumran texts in particular. On the one hand, the countless documented cases of scribal revision and reuse attested in Second Temple manuscripts provide support for the idea of such activities in earlier, nondocumented stages of the text.4 On the other hand, study of this material has demonstrated the extent to which traditional ways of thinking – and talking – about scriptural texts and scribal activities impose anachronistic or otherwise inappropriate frameworks.5 In several ways, the term innerbiblical exegesis implicates its users in these traditional ways of thinking, and as such it is beginning to hinder our understanding of the texts. Two major difficulties emerge from the way the concept of innerbiblical exegesis leads us to construe the data constituted by scriptural texts. The first pertains to the variety of ways scriptural exegesis can be presented textually and the various textual contexts in which it occurs. More specifically, I will argue that speaking of innerbiblical exegesis as a broad category obscures the difference between exegesis that takes place in the course of production of a new copy or edition of a work and that which stems from the reuse of earlier materials in a new composition. The second difficulty concerns the distinction the term implies between innerbiblical and extrabiblical interpretation – a distinction that is not meaningful for the period in which the Pentateuch was formed. In exploring these difficulties, I hope not only to offer a terminological correction but also to indicate avenues for continued study of the relevance of Second Temple models to pentateuchal theory.

1 Revision and Reuse The concept of innerbiblical exegesis has always encompassed a wide variety of textual phenomena, as indicated by even a glance at Fishbaneʼs magisterial book. These range from minor glosses to extensive new compositions. Yet close attention to the variety of examples shows that, despite this diversity, all innerbiblical exegesis occurs in one of two basic situations in terms of the relationship 4  See in particular D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011), 13–149; also M. M. Zahn, “ ‘Editing’ and the Composition of Scripture: The Significance of the Qumran Evidence,” HBAI 3 (2014), 298–316. 5  E. Ulrich, “Methodological Reflections on Determining Scriptural Status in First Century Judaism,” in Rediscovering the Dead Sea Scrolls (ed. M. L. Grossman; Grand Rapids, MI: Eerdmans, 2010), 145–161; D. A. T eeter , “The Hebrew Bible and/as Second Temple Literature: Methodological Reflections,” DSD 20 (2013), 349–377.

Innerbiblical Exegesis

109

of the new material to the text it interprets. On the one hand, it can take the form of exegetical changes made to the interpreted text itself, a technique I refer to as revision. On the other hand, it can involve what I call reuse: redeployment of the interpreted text, with exegetical modifications, in a new composition. Some elaboration on these two modes of presenting interpretation will help clarify the distinction between them. Revision involves secondary additions to or alterations of a given textual unit, embedding interpretation within the very text being interpreted. To give just one minor example, Fishbane identifies clause B in the following presentation of Lev 25:8 as a secondary gloss meant to clarify that ‫ שבת‬in this context means a seven-day period:6 Leviticus 25:8

‫ וספרת לך שבע שבתת שנים‬A ‫ שבע שנים שבע פעמים‬B ‫ והיו לך ימי שבע שבתת השנים תשע וארבעים שנה‬C A You shall count for yourself seven Sabbaths of years – B seven years, seven times – C so that the days of the seven Sabbaths of years come to forty-nine years.

It is important to note that, strictly speaking, the identification of clause B as a later insertion is hypothetical. It is based not on manuscript evidence but on the redundancy it creates vis-à-vis clauses A and C.7 Of course, this is just one of myriad cases where scholars have identified secondary material in the Pentateuch (as represented by the Masoretic Text) on the basis of literary, stylistic, or linguistic evidence alone. But there is also plenty of manuscript evidence to support the idea that texts were revised and expanded in the course of their transmission. Copies of biblical books preserved at Qumran, as well as other ancient versions such as the LXX and Samaritan Pentateuch (SP), preserve numerous examples of such revisions. The defining feature of this type of innerbiblical exegesis is that it is inserted into the text being interpreted. That is, even though there is no manuscript evidence for Fishbaneʼs example presented above, he implicitly suggests that at one point a copy of Leviticus existed with a version of Lev 25:8 that lacked clause B. The other situation in which innerbiblical exegesis can occur is in the context of a new composition rather than a copy or edition of an existing work. Here, instead of interpretive revision of the base text, the text is reused (recycled, redeployed, alluded to) in a different work. The many instances in which Deuteronomy updates the laws of the Covenant Code all constitute examples

 Fishbane, Biblical Interpretation (see n. 1), 166–167.  F ishbane , Biblical Interpretation (see n. 1), 167: “The abrupt syntax and redundant specifications suggest that section (b) constitutes a secondary elucidation [. . .].” 6 7

110

Molly M. Zahn

of this type of innerbiblical exegesis. Consider, for example, Deuteronomyʼs transformation of the manumission laws in Exod 21:2–11:8 Exodus 21:2–11 ‫׃ אם בגפו יבא בגפו יצא אם בעל‬ ‫כי תקנה עבד עברי שש שנים יעבד ובשבעת יצא לחפשי חנם‬ ‫׃ אם אדניו יתן לו אשה וילדה לו בנים או בנות האשה וילדיה תהיה‬ ‫אשה הוא ויצאה אשתו עמו‬ ‫׃‬ ‫׃ ואם אמר יאמר העבד אהבתי את אדני את אשתי ואת בני לא אצא חפשי‬ ‫לאדניה והוא יצא בגפו‬ ‫׃‬ ‫והגישו אדניו אל האלהים והגישו אל הדלת או אל המזוזה ורצע אדניו את אזנו במרצע ועבדו לעלם‬ . . . ‫וכי ימכר איש את בתו לאמה לא תצא כצאת העבדים‬ If you purchase a Hebrew slave, he shall serve for six years and in the seventh he shall go out free, without debt. If he comes alone he shall go out alone; if he has a wife, then his wife shall go out with him. If his master gives him a wife and she bears him sons or daughters, the woman and her children shall belong to her master, and he shall go out alone. If the slave should say, “I love my master, my wife, and my children; I will not go free!” then his master shall bring him before God; he shall bring him to the door or the doorpost, and his master shall pierce his ear with an awl, and he shall serve him forever. If a man sells his daughter as a female slave, she shall not go out as the male slaves go out. [. . .] Deuteronomy 15:12–17 ‫׃ וכי‬ ‫כי ימכר לך אחיך העברי או העבריה ועבדך שש שנים ובשנה השביעת תשלחנו חפשי מעמך‬ ‫׃ העניק תעניק לו מצאנך ומגרנך ומיקבך אשר ברכך יהוה‬ ‫תשלחנו חפשי מעמך לא תשלחנו ריקם‬ ‫׃ וזכרת כי עבד היית בארץ מצרים ויפדך יהוה אלהיך על כן אנכי מצוך את הדבר‬ ‫אלהיך תתן לו‬ ‫׃ ולקחת את‬ ‫׃ והיה כי יאמר אליך לא אצא מעמך כי אהבך ואת ביתך כי טוב לו עמך‬ ‫הזה היום‬ ‫׃‬ ‫המרצע ונתתה באזנו ובדלת והיה לך עבד עולם ואף לאמתך תעשה כן‬ If one of your fellow Hebrews, man or woman, sells himself to you, he shall serve you for six years, and in the seventh year you shall send him out from you free. And when you send him out free from you, do not send him out emptyhanded. You shall surely provide for him from your flock and from your threshing floor and from your winepress; you shall give to him some of that with which YHWH your God has blessed you. Remember that you were a slave in the land of Egypt and YHWH your God redeemed you; therefore I am giving you this commandment today. If it should happen that he says to you, “I shall not go out from you,” because he loves you and your household because he is well off with you, then you shall take an awl and put it through his ear and into the door, and he will be your perpetual slave. Even for your female slave you shall do likewise.

In contrast to the cases of revision mentioned above, where the interpretive element is integrated directly into the text being interpreted, here the interpreted text is not changed directly. Rather, the interpretation is located in a separate composition. One could imagine a scribe revising the Covenant Codeʼs manumission law (say, by making it apply equally to men and women) in the course of making a new copy of the book of Exodus, just as Fishbane proposes that a scribe revised the beginning of the Jubilee legislation of Lev 25:8. But in this case, the 8  For a thorough discussion of the texts and defense of the argument that Deuteronomy uses the Covenant Code, see B. M. Levinson, “The Manumission of Hermeneutics: The Slave Laws of the Pentateuch as a Challenge to Contemporary Pentateuchal Theory,” in Congress Volume: Leiden, 2004 (ed. A. Lemaire; VTSup 109; Leiden: Brill, 2006), 281–324.

Innerbiblical Exegesis

111

law in Exodus remains unchanged, because the new, reinterpreted version of the law is located in a separate work (i.e., in Deuteronomy). In other words, the Covenant Codeʼs manumission law is reused in the context of Deuteronomy and altered in the course of that reuse. Unlike cases of revision, which require multiple manuscripts of the same work in order to be documented with certainty, many cases of reuse can be identified by looking at the text of the MT Pentateuch alone – precisely because the Pentateuch has preserved distinct compositional units of varying origin alongside one another (in particular the legal material of the Covenant Code, Holiness Code, and Deut 12–26).9 Of course, study of such reuse does face the difficulty of determining which of two parallel passages is the later, rewriting text and which is the earlier version.10 This distinction between the two types of innerbiblical exegesis, which I have dubbed revision and reuse, perhaps becomes clearer in light of the way scholars have distinguished these two forms of presenting interpretation in the context of study of the Qumran scrolls. What I have referred to as revision – that is, the introduction of changes in the course of production of a new manuscript copy of a text – lies behind the idea, popularized by Eugene Ulrich, of “variant literary editions.”11 This is the idea that scriptural texts (as well as other texts) would have been successively revised as they were transmitted. Thus for Ulrich, the Old Greek of Exodus, the MT of Exodus, and the version represented by 4QpaleoExodm (and SP) all constitute variant literary editions of the book of Exodus.12 Ulrichʼs model raises questions about what kinds and what volume of changes are necessary to constitute a new “edition,” but the salient point here is the ongoing revision of manuscripts as they are transmitted.13 The hypothetical identification of examples of innerbiblical exegesis such as Fishbaneʼs analysis of Lev 25:8 presented above extends this same observation back beyond the witness of our earliest manuscripts and presumes (rightly, in my view) that similar kinds of revisions must have occurred in earlier stages.  For my purposes here it is not necessary to resolve the question whether the later of these corpora ever circulated independently or were intended from the start to accompany the earlier ones; the point is that the interpretation is embedded in a new textual unit rather than in a new version of the text being interpreted. 10  See D. M. Carr, “Method in Determination of Direction of Dependence: An Empirical Test of Criteria Applied to Exodus 34,11–26 and Its Parallels,” in Gottes Volk am Sinai (ed. M. Köckert and E. Blum; VWGTh 18; Gütersloh: Gütersloher Verlag, 2001), 107–140. 11  E. U lrich , The Dead Sea Scrolls and the Origins of the Bible (Grand Rapids, MI: Eerdmans, 1999), 17–50, 99–120. 12  Ulrich, Origins (see n. 11), 39. 13  This basic insight was articulated already by Talmon, who was one of the first scholars to integrate evidence from Qumran into study of the transmission history of the biblical text; see S. Talmon, Text and Canon of the Hebrew Bible: Collected Studies (Winona Lake, IN: Eisenbrauns, 2010). 9

112

Molly M. Zahn

Innerbiblical exegesis of the type I identified as reuse – that is, the redeployment, with alterations, of an existing text in the context of another composition – corresponds to the phenomenon generally known as Rewritten Scripture.14 Classic examples of Rewritten Scripture, such as the Temple Scroll and the book of Jubilees, were given this moniker in an attempt to highlight their extensive interpretive reuse of earlier scriptural compositions in the context of a new literary work. Indeed, it has long been recognized that works inside the Bible, in particular Chronicles and Deuteronomy, are just as deserving of the label Rewritten Scripture as are extrabiblical compositions. Recent years have witnessed extensive debate about the suitability and proper definition of Rewritten Scripture, but the term remains widely recognized as a reference to compositions constituted substantially through redeployment of existing scriptural texts.15 A few points of clarification are necessary at this juncture. First, the actual means by which the base text is reconfigured in the two types of innerbiblical exegesis outlined above are more or less identical: additions, rearrangements, replacements, and so forth.16 It is not the nature of the change that is significant in this context but the textual situation in which the changed text is presented, either in a new manuscript copy of an existing work (revision) or in a new work (reuse). Nor does the source of the change impact the distinction between revision and reuse. Frequently, revision involves an attempt to better coordinate a passage with related texts, which may involve importing material from these parallel texts into the interpreted text. For example, Deuteronomyʼs version of the Decalogue underwent many attempts to harmonize it with the Exodus version, which took place via insertion of material from Exod 20 into Deut 5.17 One might be tempted to view such harmonistic insertions as reuse of the parallel texts; after all, the Exodus text is being redeployed in a new literary context. But insofar as the insertions respond to perceived gaps in Deuteronomy, regarding them as reuse would be incorrect according to this model. Since it is the text of Deuteronomy that is being interpreted and the changes (the harmonistic addi For an overview, see M. M. Zahn , “Rewritten Scripture,” in The Oxford Handbook of the Dead Sea Scrolls (ed. T. H. Lim and J. J. Collins; Oxford: Oxford University Press, 2010), 323–336. 15  For a recent overview of the debates, see the essays by A. K. Petersen (“Textual Fidelity, Elaboration, Supersession or Encroachment? Typological Reflections on the Phenomenon of Rewritten Scripture”) and J. G. C ampbell (“Rewritten Bible: A Terminological Reassessment”) in Rewritten Bible after Fifty Years: Texts, Terms, or Techniques? (ed. J. Zsengellér; JSJSup 166; Leiden: Brill, 2014), 13–48 and 49–81. 16  For examples, see M. M. Z ahn , Rethinking Rewritten Scripture: Composition and Exegesis in the 4QReworked Pentateuch Manuscripts (STDJ 95; Leiden: Brill, 2011). 17  For the various degrees of harmonization, see E. E shel and H. E shel , “Dating the Samaritan Pentateuchʼs Compilation in Light of the Qumran Biblical Scrolls,” in Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov (ed. S. M. Paul et al.; Leiden: Brill, 2003), 215–240, here 232–235. 14

Innerbiblical Exegesis

113

tions) are made directly to the text of Deuteronomy itself, the changes constitute revisions. In other words, the terminology of revision vs. reuse is intended to pertain to the techniques by which the exegetical rewriting of a given passage (in this case, the Deuteronomy Decalogue) is accomplished, without regard for the sources of that rewriting. (By contrast, the summary recounting of the Decalogue in Josephusʼs Antiquities – a new composition – constitutes reuse.)18 Finally, these two alternative settings for exegesis each encompass considerable variety in content and in scope. A new composition may involve systematic reuse of older material, as in the Temple Scroll or Jubilees, or may consist primarily of new material with only a small amount of reuse (as, arguably, 1 Enoch, Ben Sira, or the Qumran text titled Apocryphon of Jeremiah C). Likewise, activity within a new copy or edition of a text could be as extensive as a comprehensive revision or substantial redactional layer or as isolated as a single gloss or insertion. But this diversity within each type should not obscure the fundamental literary distinction between a new work and a copy of an existing work – that is, in the situation in which new interpretive material is presented vis-à-vis the text being interpreted. There are a number of reasons why I have laid so much stress on this distinction between revision and reuse. First, it allows us to speak more precisely about forms of textual development in ancient Judaism. In a context where interpretation could be embedded directly in a given text, choosing to embed interpretation in a new work, or to compose a new work substantially dependent upon an existing work, represents to my mind a conscious decision that should be interrogated as such.19 Furthermore, despite the fact that the same interpretive moves and the same scribal techniques can appear in both types of presentation, it may be that certain techniques or attitudes toward the source text tend to appear in one type rather than the other. For example, it appears that new compositions, even those that draw extensively on an earlier source, tend to be more likely than new editions to omit significant amounts of material. New editions, on the other hand, seem generally to follow what David Carr has termed the “trend towards expansion.”20 Second, part of the reason for calling attention to these two different forms of presentation of innerbiblical exegesis is to counter a trend in which various scholarly camps have tended to focus on one type or the other depending on 18  Ant. 3.91–92. Since Josephus is recounting the events at Sinai, the direct source of the reuse in this case is Exod 20 rather than Deut 5. Also, without entering into debates over its original literary setting, it can be noted that the presentation of the Decalogue for a second time in the Pentateuch (either Exod 20 redeploying Deut 5 or vice versa) itself constitutes reuse. 19  For an elaboration of this argument, see M. M. Zahn, “Genre and Rewritten Scripture: A Reassessment,” JBL 131 (2012), 271–288. 20  C arr , Formation (see n. 4), 65–90. On the issue of omissions, see now J. Pakkala , Godʼs Word Omitted: Omissions in the Transmission of the Hebrew Bible (FRLANT 251; Göttingen: Vandenhoeck & Ruprecht, 2013).

114

Molly M. Zahn

their own views of the composition of the Pentateuch. As Jeffrey Stackert has recently pointed out, “Non-Documentarians” who posit a significant contribution from various redactors tend to identify innerbiblical exegesis with redactional revision and supplementation.21 That is, scholars who emphasize the role of redaction in the composition of the Pentateuch construe innerbiblical exegesis as the interpretive encounter with, and shaping of, earlier versions of the text by later redactors – revision, in the terms outlined above. Reinhard Kratz, for instance, goes so far as to say that “[i]nnerbiblische Exegese und die Redaktionsgeschichte der biblischen Bücher sind eins” (“innerbiblical exegesis and the redaction history of the biblical books are one”).22 On the other hand, “Neo-Documentarians” such as Stackert maintain that very little of the text of the MT Pentateuch should be attributed to redaction: the (single) redactorʼs goal was to bring previously independent sources together into a single narrative, and this redactor only intervened “to harmonize glaring contradictions at the level of the narrative.”23 Since very little material, in this view, can be assigned to redactors, and what little there is functions only to allow for a basic narrative coherence, innerbiblical exegesis cannot be equated with redaction. Instead, innerbiblical exegesis in the Pentateuch is restricted to the interpretive reuse of earlier sources by later sources, for example Deuteronomyʼs use of the Covenant Code or the Holiness Codeʼs use of Deuteronomy.24 In a way, the idea of innerbiblical exegesis has become a pawn in the struggle over how best to understand the Pentateuchʼs composition: each side construes the term as befits their theoretical framework. To me (as a relative outsider to these debates), the ample documented cases of interpretive rewriting in both modes – revision and reuse – at Qumran and in other Second Temple contexts would seem to suggest a more balanced approach. We clearly have cases within the Pentateuch where a distinct literary work reuses earlier material still preserved elsewhere in the Pentateuch (e.g., Deuteronomyʼs use of the Covenant Code); though direction of dependence can be debated, the fact that some sort of reuse took place is indisputable. On the other hand, the presence of redactional supplementations and revisions in the MT Pentateuch is in a sense more easily questioned, since their identification depends on the use of literary or stylistic clues to detect diachrony within a single text. Nevertheless, the extensive doc21  J. Stackert, “Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction: Leviticus 26 as a Test Case,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 369–386, here 372. 22  R. G. Kratz, Das Judentum im Zeitalter des Zweiten Tempels (FAT 42; Tübingen: Mohr Siebeck, 2004), 156. Despite this strong formulation and his general commitment to redaction-critical approaches, Kratz offers a broad and nuanced discussion of the various forms by which earlier traditions were interpreted and reconfigured in Second Temple texts. 23  Stackert, “Distinguishing Innerbiblical Exegesis” (see n. 21), 372 (emphasis in original). 24  Stackert, “Distinguishing Innerbiblical Exegesis” (see n. 21), 372.

Innerbiblical Exegesis

115

umented evidence for revision in later stages of transmission (i.e., manuscript evidence from Qumran and elsewhere) suggests that similar revisions in earlier stages are a priori likely. The evidence does not necessarily support particular redactional reconstructions but does suggest that revisions of the type often identified by redaction critics are to be expected. Without committing to particulars, we can say that the composition of the Pentateuch very likely involved both reuse and revision.

2 Innerbiblical vs. Extrabiblical: An Artificial Distinction As suggested by the reference to Second Temple manuscript finds in the previous section, my proposal to distinguish between revision of a text in the context of a new copy and reuse in the context of a new composition also has the benefit of highlighting points of continuity between so-called biblical and extrabiblical material. This leads to my second main point, the artificial and arbitrary distinction that talking about innerbiblical exegesis imposes. The problem, of course, is with the term innerbiblical. As significant as the idea of innerbiblical exegesis has been for breaking down the artificial barrier between Scripture and interpretation, its use now in a way threatens to reify that very distinction. The sole unifying factor behind all the various examples of innerbiblical exegesis that Fishbane presented, and all the additional examples collected by others both before and after Fishbane, is that they occur in the Bible. But this unifying factor is completely inappropriate as a descriptor of the textual situation of ancient Judaism prior to 70 CE, at the earliest. Simply put, there was no Bible, no fixed list of scriptural texts, at that time.25 There were certainly authoritative texts, whose importance was confirmed by their subjection to revision, updating, and transformation.26 Some of these texts – but certainly not all – later ended up in the (Jewish) Bible, others in various Christian canons; others were lost until the Qumran discoveries. To speak of “innerbiblical exegesis” is to imply that the textual processes affecting those works that did end up in the Bible belong in a separate category from those affecting other works. But to do so means reading the boundaries of the canon back into the Second Temple period and thus creating categories that would not have existed in the minds of Second Temple writers and audiences.

 See, e.g., J. C. VanderKam, “The Wording of Biblical Citations in Some Rewritten Scriptural Works,” in The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries (ed. E. Herbert and E. Tov; London: British Library, 2002), 41–56. 26  G. J. Brooke, “Between Authority and Canon: The Significance of Reworking the Bible for Understanding the Canonical Process,” in Reworking the Bible: Apocryphal and Related Texts at Qumran (ed. E. Chazon et al.; STDJ 58; Leiden: Brill, 2005), 85–104, here 95. 25

116

Molly M. Zahn

The following points should make clear that this is not just an issue of anachronistic terminology but actually causes confusion in how we talk and think about the texts. One of my goals in my work with the Qumran materials is to illustrate how, in the late Second Temple period, a more-or-less shared set of processes of textual rewriting was operative in the composition and development of a wide range of texts. These include copies and translations of biblical books, Rewritten Scripture, pseudepigraphic works of various kinds, and classic sectarian texts. When we talk about innerbiblical exegesis as a distinct phenomenon, we obscure the fact that – as far as we can tell – the books of the Bible were composed and transmitted in essentially the same ways as extrabiblical compositions. A few more specific observations will illustrate this point. Compositions produced by means of reconfiguring an older work – that is, extended examples of what is here referred to as reuse – are, as mentioned above, reflected both inside and outside the canon. Deuteronomyʼs rewriting of the Covenant Code and Chroniclesʼ rewriting of Samuel–Kings are precisely analogous in a functional sense to Jubileesʼ rewriting of Genesis and the Temple Scrollʼs rewriting of Deuteronomy. And there is good evidence that some people would have considered the Temple Scroll and Jubilees to be at least as authoritative as Deuteronomy or Chronicles.27 But only two of these works would be considered innerbiblical exegesis (or three, if you take the Ethiopic canon into account). In other words, the label innerbiblical exegesis tells us nothing about the goals, strategies, compositional techniques, or reception of these texts in the Second Temple period: it simply reflects a later decision about the scope of the canon. Furthermore, it is becoming clear that composition via reuse was not limited to the reuse of biblical texts. Reinhard Kratz and Annette Steudel have recently argued that the Damascus Document known from Qumran and the Cairo Geniza should also be considered Rewritten Scripture, in the sense that it rewrites the Community Rule.28 Rewriting, in other words, was not limited to biblical texts. Other Qumran evidence, including for example the reuse of  For Jubilees, see J. C. VanderKam, “Authoritative Literature in the Dead Sea Scrolls,” DSD 5 (1998), 382–402, here 396–400. The case of the Temple Scroll is more difficult, but its presence at Qumran in one, perhaps two, de luxe-format manuscript(s) strongly indicates that it was held in high regard there. On de luxe manuscripts, see E. Tov, Scribal Practices and Approaches Reflected in the Texts from the Judean Desert (STDJ 54; Leiden: Brill, 2004), 126–129. In contrast, the same Qumranites who seem to have regarded Jubilees and the Temple Scroll as authoritative appear to have been reluctant to accept the authority of Chronicles, perhaps because of perceived connections to the Hasmoneans; see G. J. Brooke, “The Books of Chronicles and the Scrolls from Qumran,” in Reflection and Refraction: Studies in Biblical Historiography in Honour of A. Graeme Auld (ed. R. Rezetko et al.; VTSup 113; Leiden: Brill, 2007), 35–48. 28  R. G. Kratz, “Der ‘Penal Code’ und das Verhältnis von Serekh ha-Yahad (S) und Damas­ kusschrift (D),” RevQ 25 (2011), 199–227; A. Steudel, “The Damascus Document (D) as a Rewriting of the Community Rule (S),” RevQ 25 (2012), 605–620. 27

Innerbiblical Exegesis

117

Jubilees in 4Q390, points in the same direction.29 If all of these rewritten texts are doing essentially the same thing, surely we stand to gain the most if we allow all of them to inform one another, irrespective of their canonical status and without introducing distinctions based on later stages in Jewish and Christian history. Similar connections across canonical boundaries pertain to examples of revision, i.e., cases where the interpreted text is itself directly changed. The first boundary to consider is that between the MT as textus receptus and other manuscripts and versions. I have already indicated that revisions of the type illustrated above or postulated by redaction critics are well attested in manuscripts and translations of biblical books. But for whatever reason – perhaps an implicit overreliance on the MT or the disciplinary disconnects between pentateuchal studies, text criticism, and Qumran studies – such documented cases have not always been seen as examples of innerbiblical exegesis. There seems to be a sense that these cases belong rather to the category of “empirical models” – that they are somehow different from “the Bible.” Yair Zakovitch and Kratz both reference this general impression in perceptive recent essays on innerbiblical exegesis. Kratz emphasizes that the “interpretation and reception” of biblical texts witnessed in the versions and in Qumran manuscripts constitutes a continuation of compositional and redactional processes. In this context, he observes that “[t]extual transmission and translation are not extrabiblical analogies in the strict sense but rather examples of innerbiblical exegesis, although those with external evidence.”30 Similarly, Zakovitch uses the pluriform witness to the biblical text as evidence of the fluid boundary “between innerbiblical and extrabiblical interpretation,” asking, “[i]s exegesis that is incorporated into these different versions innerbiblical or extrabiblical exegesis?”31 It is critical to recognize here that the only difference between cases of revision attested in manuscripts and versions and those detected within MT (and therefore considered examples of innerbiblical exegesis) is, as Kratz notes, precisely the fact that in the former cases we happen to have manuscript evidence for the revision. It is not a true difference in the nature or setting of the change. That is, we are dealing here not with different categories but with a single category within which the nature of the evidence for specific examples differs. Distinguishing empirically attested examples as something different from innerbiblical exegesis only makes sense if by Bible we really mean MT. Granted, in pentateuchal scholarship down to recent times, Bible often did tacitly mean the Hebrew text in its Masoretic form. But the Qumran 29  See T. R. Hanneken, “The Status and Interpretation of Jubilees in 4Q390,” in A Teacher for All Generations (ed. E. Mason et al.; 2 vols.; JSJSup 153; Leiden: Brill, 2012), 1:407–428. 30  Kratz, Das Judentum im Zeitalter (see n. 22), 148 (my translation). 31  Y. Zakovitch, “Inner-Biblical Interpretation,” in A Companion to Biblical Interpretation in Early Judaism (ed. M. Henze; Grand Rapids, MI: Eerdmans, 2012), 27–63, here 35.

118

Molly M. Zahn

evidence has demonstrated spectacularly the inadequacy of this traditional identification. As indicated above, until at least the end of the first century CE, the Masoretic form of the various biblical books was only one of multiple text forms in circulation.32 One could argue that all this means is that we need to expand our definition of Bible: recognize that even in the late Second Temple period the text of the Bible “noch im Werden ist,” as Kratz puts it,33 and adjust our understanding of what counts as innerbiblical exegesis accordingly. Such a move, however, still leaves a conceptual boundary in place between the textual histories of works that were later included in the canon (innerbiblical exegesis, however broadly defined) and those of nonbiblical compositions. This distinction is difficult to justify from a methodological perspective. Empirical evidence from Qumran and elsewhere suggests that many different literary texts developed over time through a process of revision and expansion, which could involve intentional and unintentional as well as systematic and unsystematic elements.34 Perhaps the clearest Qumran example is the multiple extant versions of the Community Rule: comparison of 1QS with the numerous Cave 4 copies indicates that the Community Rule existed in at least three different versions, with differences encompassing the presence or absence of major sections as well as alternative forms of individual passages.35 Given the broad attestation of revision in the course of textual transmission, there would seem to be nothing to gain from distinguishing certain examples of revision from others based on the canonical status of the work in which they appear. Taking into account the above discussion about the versional evidence, all of the additional material in, for example, MT Jeremiah vis-à-vis the shorter LXX version, or Greek Daniel vis-à-vis the MT, could be classified as innerbiblical exegesis. But what then do we call functionally equivalent revisions in the Cave 1 copy of the Community Rule vis-à-vis the shorter manuscripts from Cave 4? Just as in the case of new “rewritten” compositions, the real downside of employing the label innerbiblical exegesis for textual revision and expansion is that fundamental similarities between biblical and nonbiblical texts are obscured.  See S. W. Crawford, “ ‘Biblical’ Text – Yes or No? The Hebrew Bible,” in What is Bible? (ed. K. Finsterbusch and A. Lange; CBET 67; Leuven: Peeters, 2012), 113–119; E. Ulrich, “The Text of the Hebrew Scriptures at the Time of Hillel and Jesus,” in Congress Volume: Basel, 2001 (ed. A. Lemaire; VTSup 92; Leiden: Brill, 2002), 85–108. 33  Kratz, Das Judentum im Zeitalter (see n. 22), 145. 34  Carr, Formation (see n. 4), 13–101. 35  See the methodologically astute analysis by C. Hempel, “Sources and Redaction in the Dead Sea Scrolls: The Growth of Ancient Texts,” in Rediscovering the Dead Sea Scrolls (ed. M. Grossman; Grand Rapids, MI: Eerdmans, 2010), 162–181, and the literature cited there. Signs of textual development are also found in the manuscript traditions of the Damascus Document, the War Scroll, the Hodayot, and 4QPseudo-Ezekiel, among others. 32

Innerbiblical Exegesis

119

3 Conclusions The following chart attempts to represent the main points of my argument in graphic terms. The two categories separated by the vertical dividing line – new compositions and new copies or editions – represent a significant distinction between two textual formats for presenting interpretation, one that I have argued could play a meaningful role in further discussions of textual composition and transmission in early Judaism. The distinction between texts traditionally labeled biblical and nonbiblical/postbiblical, represented by the horizontal line, is not significant and in fact is misleading, insofar as it implies a difference between cases of reuse and revision that occur within and outside of the canon as represented by MT. The difficulty with the term innerbiblical exegesis is that it lays stress on the false distinction represented by the horizontal line while obscuring the useful distinction represented by the vertical line.

“nonbiblical” or “postbiblical”

Temple Scroll Jubilees 1 Enoch 4QPseudo-Ezekiel etc.

Greek Daniel/Esther Targumim 1QS Samaritan Pentateuch 4QReworked Pentateuch 11QPsalmsa etc.

“the Bible” (= MT)

The term innerbiblical exegesis . . .

1–2 Chronicles Deuteronomy (e.g., Levinson) Holiness Code (e.g., Stackert) Isaiah 40–66 (Sommer) etc.

MT Jeremiah MT Ezekiel MT Sinai Pericope MT Deuteronomy (e.g., P/post-P additions) etc.

New compositions

New copies/editions



. . . obscures this line (which is important) . . . imposes this line (which is not) Figure 1. The Problematic Implications of “Innerbiblical Exegesis”36

 The specific examples offered in each quadrant are not universally agreed upon, of course; they are there for heuristic purposes only. The scholarsʼ names in parentheses in the lower lefthand quadrant refer to their specific construals of these texts as incorporating substantial reuse of earlier material; see B. M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (Oxford: Oxford University Press, 1997); J. S tackert , Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007); B. D. Sommer, A Prophet Reads Scripture: Allusion in Isaiah 40–66 (Stanford: Stanford University Press, 1998). 36

120

Molly M. Zahn

Moving away from use of the term innerbiblical exegesis constitutes one step in the ongoing process of fully integrating Second Temple materials into discussions about the development of the Pentateuch. We have learned a great deal – from the study of the Qumran manuscripts in particular – about the nature of text production and transmission in the late Second Temple period. These insights have begun to be felt in pentateuchal studies, but there is much more to be done. We can use Qumran texts as models to understand the Pentateuch as an authoritative text comprising (at least in part) previously independent authoritative texts but also to recognize how many different scribal authors likely contributed to the growth of the Pentateuch as a text, in stages empirically attested and in those that are not. The rise of the concept of innerbiblical exegesis was a critical step in appreciating the nature of the Pentateuch and the nature of its growth, but it is time now to develop language and models that more closely reflect the textual and cultural realities of early Judaism.

From Many to One Some Thoughts on the Hebrew Textual History of the Torah Armin Lange Until the discovery of the Dead Sea Scrolls, the Hebrew text of the Torah was almost exclusively known through the medieval Masoretic text. Only the Nash Papyrus, the Severus Scroll variant list,1 the variant readings attributed to Rabbi Meir,2 and biblical quotations in the ancient Jewish and rabbinic literature3 provided rare insights into the ancient and late ancient history of the Hebrew text of the Bible. In addition, a limited number of manuscripts and other witnesses survived from late antiquity and the early Middle Ages (see below, 1.2), which remained largely unknown outside the small group of text-critical specialists. Since the discovery of the Dead Sea Scrolls, this situation changed dramatically. The biblical Dead Sea Scrolls illuminate the early textual history of the Torah This paper has a long history. It merges two presentations that I gave at the Institute for Advanced Studies in Jerusalem in the years 2013 and 2015 with two articles that I published: A. Lange, “The Dead Sea Scrolls and the Date of the Final Stage of the Pentateuch,” in On Stone and Scroll: Essays in Honour of Graham Ivor Davies (ed. J. Aitken et al.; BZAW 420; Berlin: de Gruyter, 2011), 287–304; and idem, “1.2.2 Ancient and Late Ancient Hebrew and Aramaic Texts,” in The Hebrew Bible, Vol. 1 of Textual History of the Bible (ed. A. Lange and E. Tov; Leiden: Brill, 2015), http://referenceworks.brillonline.com/entries/textual-historyof-the-bible/122-ancient-and-late-ancient-hebrew-and-aramaic-jewish-texts-COM_00000007 (accessed 3/3/16). The latter two papers are only included insofar as they are relevant to the Hebrew textual history of the Torah. I am much obliged to Andrea Colella (Vienna), who adapted the style of this paper for this volume, and to my friend and colleague Bernard M. Levinson, whose kind invitation and patient reminders made this paper possible. 1  For the Severus Scroll variant list, see below, n. 25. 2  See A. Lange, “Rabbi Meir and the Severus Scroll,” in“Let the Wise Listen and Add to Their Learning” (Prov 1:5): Festschrift for Günter Stemberger on the Occasion of His 75th Birthday (ed. C. Cordoni and G. Langer; SJ 90; Berlin: de Gruyter, 2016), 53–76. 3  See A. L ange , “1.7.1 Jewish Quotations and Allusions,” in The Hebrew Bible, Vol. 1 of Textual History of the Bible (ed. A. Lange and E. Tov; Leiden: Brill, 2015), http://referenceworks.brillonline.com/entries/textual-history-of-the-bible/171-jewish-quotations-and-allusions-COM_000707 (accessed 3/3/16); and A. R osen -Z vi , “Rabbinic Literature,” in Textual History of the Bible (ed. A. Lange; Brill Online, 2016) http://referenceworks.brillonline.com/entries/textual-history-of-the-bible/216rabbinic-literature-COM_00000424 (accessed 3/3/16).

122

Armin Lange

and leave little doubt that the consonantal text of MT was just one among many different Torah texts. In this article, I will try to trace the textual history of the Torah from its textual plurality in the Second Temple period until the time of the master codices from the Ben Asher family. Because after the Second Temple period this article is concerned only with the textual history of the (proto)-Masoretic text, I will not engage with the Samaritan Pentateuch, the Septuagint, or any of the other translations of the Hebrew Bible beyond the point where they provide information about the textual history of MT. At the end of this article, I will draw some conclusions for the higher criticism of the Torah from my observations. I will begin by surveying the existing manuscript witnesses to the Hebrew text of the Torah.

1 Hebrew Manuscripts of the Torah 1.1 Masoretic Manuscripts The medieval Masoretic manuscripts are witnesses to an ancient consonantal text of the highest importance. The most important of these manuscripts are those that are either attributed or related to the Ben Asher family, namely, MTA, MTL, and MTB.4 MTA Codex Aleppo, Israel Museum (ca. 930 CE; most of the Torah is missing; the following is preserved on photographs or in the original: Gen 26:34–27:30; Exod 8:10, 17; Deut 4:38–6:3; Deut 28:17–34:12).5  Further early and important Masoretic manuscripts are described in A. H arkavy and H. L. S track , Die hebräischen Bibelhandschriften, Vol. 1 of Catalog der hebräischen und samaritanischen Handschriften der kaiserlichen öffentlichen Bibliothek in St. Petersburg (St. Petersburg: C. Ricker, 1875); R. Gottheil, “Some Hebrew Manuscripts in Cairo,” JQR 17 (1905), 608–655; P. Kahle, Masoreten des Westens (2 vols.; Texte und Untersuchungen zu vormasoretischen Grammatik des Hebräischen 1 + 4; Stuttgart: W. Kohlhammer, 1927–1930), 1:56–77; I. Y eivin , Introduction to the Tiberian Masorah (ed. and trans. E. J. Revell; SBLMass 5; Missoula, MT: Scholars Press, 1980), 15–31; and in M. Beit-Arié et al., Codices hebraicis litteris exarati quo tempore scripti fuerint exhibentes (4 vols.; Turnhout: Brepols, 1997–2006), which discuss manuscripts up to the year 1200. 5  Descriptions of Codex Aleppo can be found in Yeivin, Introduction (see n. 4), 17–18; and M. Beit-Arié et al., Codices hebraicis (see n. 4), 1:65–72. A facsimile edition of MTA was published by M. Goshen-Gottstein (ed.), The Aleppo Codex: Provided with Massoretic Notes and Pointed by Aaron ben Asher (Jerusalem: Magnes Press, 1976). Critical editions of the extant passages of Codex Aleppo were provided by M. Breuer, Torah Neviʾim Ketuvim Proofread according to Mesorah of Keter Aram Tzova and Like Manuscripts (Jerusalem: Mossad HaRav Kook, 1989), and N. Ben Z vi (ed.), Jerusalem Crown: The Bible of the Hebrew University of Jerusalem (Jerusalem: S. Karger Publishing, 2000). MTA also provides the running text of the Hebrew University Bible and is the biblical text printed in the M. Cohen (ed.), Miqraot 4

From Many to One

123

MTB London, British Library, Oriental Ms. 4445 (820–850 CE6 or 925 CE;7 folios 1–28 [Gen 1:12–39:20], 125 [Num 7:46–72], 128 [Num 9:12– 10:18], and 160–186 [Deut 1:33–34:12] were added in 1539/1540 CE; extant are Gen 39:20–Deut 1:33, except Num 7:46–73; 9:12–10:18).8 C3 MT Cairo Pentateuch Codex (Ms Gottheil 18) (tenth century; complete Pentateuch).9 L MT Codex Leningradensis, Codex EPB I B 19a of the National Library of Russia in St. Petersburg (1009 CE; complete Hebrew Bible).10 Gedolot Haketer (Ramat Gan: Bar Ilan University Press, 1992–) (Hebrew). Images of the Aleppo Codex are also available online at aleppocodex.org (accessed 3/3/16) and through Wikimedia Commons, https://commons.wikimedia.org/wiki/Category:Aleppo_Codex (accessed 3/3/16). Yeivin, Introduction (see n. 4), 17–18, reports Exod 8:10, 17 and 2 Chr 35:7–36:19 as missing. These passages were discovered in 1982 and 1987 (M. Beit-Arié, “Another Page from the Aleppo Codex,” Tarbiz 51 [1982], 171–174 [Hebrew]; Y. Y evin , “Another Page from the Aleppo Codex,” Tarbiz 51 [1982], 174–176 [Hebrew]; Y. ʿO fer , “A Fragment of the Book of Exodus from the Missing Part of the Aleppo Codex,” Peʿamim 41 [fall 1990], 41–48 [Hebr.]). For MTA, see also J. S. Penkower, New Evidence for the Pentateuch Text in the Aleppo Codex (Ramat-Gan: Bar-Ilan University Press, 1992) (Hebrew); D. Barthélemy, Studies in the Text of the Old Testament: An Introduction to the Hebrew Old Testament Text Project (Textual Criticism and the Translator 3; Winona Lake, IN: Eisenbrauns, 2012), 238–255; English translation of idem, Critique textuelle de lʼAncien Testament, 3: Ézéchiel, Daniel et les 12 Prophètes (OBO 50/3; Göttingen: Vandenhoeck & Ruprecht, 1992), vii–xviii. 6  Thus C. D. G insburg , Introduction to the Massoretico-Critical Edition of the Hebrew Bible (London: Trinitarian Bible Society, 1897; repr. New York: Ktav, 1966), 470; A. Dotan, “Reflections towards a Critical Edition of Pentateuch Codex Or. 4445,” in Proceedings of the Tenth Congress of the International Organization for Masoretic Studies (ed. E. Fernández Tejero and M. T. Ortega Monasterio; Textos y Estudios “Cardenal Cisneros” 55; Madrid: CSIC, 1993), 39–51. 7  Thus Yeivin, Introduction (see n. 4), 19. 8  The manuscript remains unpublished. A digitized version is available through British Library Digitised Manuscripts, http://www.bl.uk/manuscripts/FullDisplay.aspx?ref=Or_4445 (accessed 3/3/16). For first discussions and descriptions of MTB, see Ginsburg, Introduction (see n. 6), 469–474 (Ginsburgʼs information on which folios are secondary insertions is erroneous); Dotan, “Reflections” (see n. 6), 39–50; idem, “The London Pentateuch Codex: A New Approach,” in Studies in Ancient and Modern Hebrew in Honor of Prof. M. Z. Kaddari (ed. S. Sharvit; Ramat Gan: Bar-Ilan University Press, 1999), 209–217 (Hebrew); idem , “Babylonian Residues in the London Pentateuch Codex,” Studies in Bible and Exegesis 7 (2005), 33–40 (Hebrew); P. Cassuto, “Qeré/ketiv dans le manuscrit Londres Or. 4445,” in Proceedings of the Eleventh Congress of the International Organization for Masoretic Studies (ed. A. Dotan; Jerusalem: World Union of Jewish Studies, 1994), 15–24. 9  For descriptions of the manuscript, see Gottheil, “Some Hebrew Manuscripts in Cairo” (see n. 4), 631–634; J. S. Penkower, “A Tenth-Century Pentateuchal MS from Jerusalem (MS C3), Corrected by Mishael Ben Uzziel,” Tarbiz 58 (1988), 49–74 (Hebrew). 10  For facsimile editions and critical editions, see D. S. Loewinger, Pentateuch, Prophets and Hagiographa: Codex Leningrad B 19a: The Earliest Complete Bible Manuscript (Jerusalem: Makor Publishing, 1971); D. N. F reedman et al. (eds.), The Leningrad Codex: A

124

Armin Lange

MTL10 Codex EBP II B 10 of the National Library of Russia in St. Petersburg (first half of the tenth century;11 extant are Gen 11:29–20:2; 21:9–50:26; Exod 1:1–30:34; 32:25–40:38; Lev 1–27; Num 1:1–19:19; 22:31–36:13; Deut 1–34).12 L17 MT  Codex EBP II B 17 of the National Library of Russia in St. Petersburg (929 CE; preserved are Gen 2:5–6:3; 8:9–20:9; 21:17–25:22; 27:6–41:43; Gen 42:35–Exod 23:7; Exod 25:2–40:38; Lev 1:1–19:20; 20:8–25:17; Lev 25:44–27:34; Num 1:1–10:35; 17:6–36:13; Deut 1–34).13 M1 MT  Codex M1 in the Complutensian Library of Madrid (1280 CE; whole Hebrew Bible except Exod 9:33b–24:7b).14 S1 MT  Manuscript Sassoon 1053; National Library of Israel (tenth century; complete Hebrew Bible except Gen 1–10).15 Facsimile Edition (Grand Rapids, MI: Eerdmans, 1998); A. D otan (ed.), Biblia Hebraica Leningradensia (Peabody: Hendrickson, 2001). The earlier codicological description by Harkavy and Strack, Catalog (see n. 4), 263–274, is surpassed by Beit-Arié et al., Codices hebraicis (see n. 4), 1:114–31; see also Yeivin, Introduction (see n. 4), 18–19. 11  Kahle, Masoreten des Westens (see n. 4), 1:61–64. 12  For a description, see K ahle , Masoreten des Westens (see n. 4), 1:60–64; Y eivin , Introduction (see n. 4), 23. 13  For a description, see Beit-Arié et al., Codices hebraicis (see n. 4), 1:53–64; as well as Kahle, Masoreten des Westens (see n. 4), 1:58–59; and Yeivin, Introduction (see n. 4), 22–23. 14  For an early description, see Ginsburg, Introduction (see n. 6), 771–776. The reseach on this codex focused mainly on its Masorah. See, e.g., M. J. Azcárraga de Servert, Las masoras del libro de Números: Códice M1 de la Universidad Complutense de Madrid (Textos y Estudios “Cardenal Cisneros” 66; Madrid: CSIC, 2001); M. T. Ortega Monasterio, Las masoras del libro de Éxodo: Códice M1 de la Universidad Complutense de Madrid (Textos y Estudios “Cardenal Cisneros” 67; Madrid: CSIC, 2002); G. S eijas de los R íos , Las masoras del libro de Deuteronomio: Códice M1 de la Universidad Complutense de Madrid (Textos y Estudios “Cardenal Cisneros” 68; Madrid: CSIC, 2002); E. Fernández Tejero, Las masoras del libro de Génesis: Códice M1 de la Universidad Complutense de Madrid (Textos y Estudios “Cardenal Cisneros” 73; Madrid: CSIC, 2004); M. J. A zcárraga de S ervert , Las masoras del libro de Levíticos: Códice M1 de la Universidad Complutense de Madrid (Textos y Estudios “Cardenal Cisneros” 74; Madrid: CSIC, 2004); E. M artín Contreras, Apéndices Masoréticos: Códice M1 de la Universidad Complutense de Madrid (Textos y Estudios “Cardenal Cisneros” 72; Madrid: CSIC, 2004). Cf. also M. T. Ortega Monasterio and E. Fernández Tejero, “Distintas manos en la Masora Parva de Pentateuco del Códice M1,” in Judaísmo Hispano: Estudios en memoria de José Luis Lacave Riaño (ed. E. Romero; 2 vols.; Madrid: CSIC, 2002–2003), 1:145–161; M. J. Azcárraga de Servert, “Un apéndice ornamental a Reyes en el ms: M1 de la Universidad Complutense de Madrid,” Sefarad 64 (2004), 227–241; E. M artín C ontreras , “The Continuity of the Tradition: Masorah with Midrashic Explanations,” JSS 50 (2005), 329–339; idem, “M1ʼs Masoretic Appendices: A New Description,” JNSL 32 (2006), 65–81. 15  For descriptions of MTS1, see D. S. Sassoon, Ohel Dawid: Descriptive Catalogue of the Hebrew and Samaritan Manuscripts in the Sassoon Library, London (2 vols.; Oxford: Oxford University Press, 1932), 2:1111–1112; and Yeivin, Introduction (see n. 4), 21–22. A digitized ver-

From Many to One

125

MTS5 Manuscript Sassoon 507, Codex ms. Heb 24° 5702 of the National Library of Israel (Damascus Pentateuch; 1000 CE;16 Gen 9:26–Deut 34:12 except Exod 18:1–23).17 V MT Manuscript Vatican ebr. 448 (1100 CE; complete Pentateuch).18 An enormous number of medieval Hebrew biblical manuscripts were also discovered in the Cairo Genizah and are today part of the various collections of manuscripts from the Cairo Genizah.19 Research on these manuscripts is still in its early stages. They are best accessible through the website of the Friedberg project.20 1.2 Manuscripts and other Witnesses between the Second Jewish War and the Medieval Masoretic Codices Although the consonantal text of MT is well attested in medieval manuscripts, and while the Dead Sea Scrolls provide significant insights into the biblical text of the late Second Temple period, the same cannot be said for the time between the Second Jewish War (132–135 CE) and the medieval Masoretic codices. Only six Hebrew biblical manuscripts from this period, attesting to various parts of the Pentateuch, 1–2 Kings, and Job, are known to date. An initial list of late ancient and early medieval biblical manuscripts was published by Gentry in sion of the manuscript is available through Wikimedia Commons, https://commons.wikimedia. org/wiki/File:Tanakh-Sassoon1053-11-Jeremiah.pdf (accessed 3/3/2016). 16  Thus D. S. L oewinger and M. B eit-A rié (eds.), The Damascus Pentateuch (2 vols.; Early Hebrew Manuscripts in Facsimile 2; Copenhagen: Rosenkilde and Bagger, 1978–1982), 2:10. Yeivin, Introduction (see n. 4), 21, argues for the tenth century. Sassoon, Ohel Dawid (see n. 15), 1:23, and Loewinger and B eit-A rié , Damascus Pentateuch, 1:13, suggest the ninth century. 17  For descriptions of the manuscript, see Sassoon, Ohel Dawid (see n. 15), 1:22–23; and Yeivin, Introduction (see n. 4), 21. For a facsimile edition, see Loewinger and Beit-Arié, Damascus Pentateuch (see n. 16); a digitized version of the manuscript is available through the National Library of Israel Digitized Manuscripts from the Collection of the National Library and Other Collections, http://jnul.huji.ac.il/dl/mss/heb5702/ (accessed 3/3/16). 18  For descriptions, see Yeivin, Introduction (see n. 4), 28; and Barthélemy, Studies, 241. 19  Important publications on these manuscripts include: P. K ahle , The Cairo Geniza (Oxford: Blackwell, 21959); I. Yeivin, Geniza Bible Fragments with Babylonian Massorah and Vocalization: Including Additional Bible Fragments with Babylonian Massorah and Vocalization (5 vols.; Jerusalem: Makor, 1973) (Hebrew); E. J. R evell , Biblical Texts with Palestinian Pointing and Their Accents (SBLMasS 4; Missoula, MT: Scholars Press, 1977); M. C. D avis and B. O uthwaite , Hebrew Bible Manuscripts in the Cambridge Genizah Collections (4 vols.; Cambridge University Library Genizah Series 2/1–4; Cambridge: Cambridge University Press, 1978–2006), vols. 3 and 4; G. Khan, Karaite Bible Manuscripts from the Cairo Geniza (Cambridge University Library Genizah Series 9; Cambridge: Cambridge University Press, 1990). 20  The Friedberg Jewish Manuscript Society, https://fgp.genizah.org or http://www. jewishmanuscripts.org/ (accessed 3/3/16).

126

Armin Lange

2009.21 Most of these manuscripts are not properly edited, and photographs are often not available. To the extent that they can be classified, all of these manuscripts are proto-Masoretic in alignment. In addition, variants of the so-called Severus Scroll and variants of at least two biblical scrolls claimed to have been in the possession of Rabbi Meir are known. Furthermore, Masoretic annotations might refer to the readings of lost late ancient or early medieval master codices. Most, if not all, of these witnesses attest only to minor disagreements with the consonantal text of MTL, and they should all be regarded as witnesses to the proto-Masoretic text.22 1. T-S NS 4.3 and T-S NS 3.21 (Cambridge, Genizah Collection): One scroll with parts of Gen 4:14–17; 5:10–18; 5:32–6:7; 13:10; 14:9–22; 15:5–21; 16:5–17:2; 17:9–20. Suggested paleographic dates range from the fifth through the ninth century. 2. Ms. Heb. d. 89 (P) i (Oxford, Bodleian Library): One fragment with parts of Exod 2:23–25. Second or third century CE. 3. P 10598 (Berlin Staatliche Museen): Scroll with part parts of Exod 3:13–16, 18–22; 4:1–9. Eighth through ninth centuries. 4. MS London + MS Ashkar-Gilson Hebrew Manuscript # 2 (private collection of S. Loewentheil in New York; David M. Ruben Rare Book and Manuscript Library at Duke University): Fragments of one scroll with the complete text of Exod 9:18–13:2 and parts of Exod 9:18–16. Seventh or eighth century. 5. A Leviticus scroll from the sixth century CE or earlier was discovered inside the holy ark of the En Gedi synagogue during a controlled excavation in the 1970s. The scroll was burned, and its charred remains were unreadable until advanced 3D scanning technology allowed researchers to decipher some of its text in the summer of 2015. So far, text from Lev 1:1–8 has been deciphered. Aside from photos published in press releases,23 not much is known about the text of this important Leviticus scroll.  P. G entry , “The Text of the Old Testament,” Journal of the Evangelical Theological Society 52 (2009), 19–45, here 22. 22  For a more detailed description and discussion of these manuscripts, see A. Lange, “2.2.1 Pentateuch: Ancient Hebrew Texts; Ancient, Late Ancient, and Early Medieval Manuscript Evidence,” in The Hebrew Bible, Vol. 1 of Textual History of the Bible (ed. A. Lange and E. Tov; Leiden: Brill, 2015), http://referenceworks.brillonline.com/entries/textual-history-of-the-bible/221-ancient-late-ancient-and-early-medieval-manuscript-evidence-COM_00000038 (accessed 3/3/16). 23  W. H arder , “Sealesʼ Research Team Reveals Biblical Text from Damaged Scroll,” University of Kentucky News, http://uknow.uky.edu/content/seales-research-team-reveals-biblical-text-damaged-scroll (accessed 3/3/16). For further press releases, see “The Most Ancient Hebrew Scroll since the Dead Sea Scrolls Has Been Deciphered,” Israel Antiquities Authority, http://www.antiquities.org.il/Article_eng.aspx?sec_id=25&subj_id=240&id=4134&hist=1 (accessed 3/3/16); “Rare Ancient Hebrew Scroll Deciphered,” Israel Ministry of Foreign 21

From Many to One

127

6. Another biblical text from late antiquity or even the late Second Temple period is the so-called Severus Scroll. This Torah scroll was already lost in late antiquity, but its variant readings are preserved in a variant list, which is still extant, transmitted in the medieval Midrash Bereshit Rabbati and two illuminated Masoretic manuscripts. The Severus Scroll variant list records only thirty-three minor textual and orthographic variants for the whole Torah. It indicates these variants in comparison with another (late) ancient Torah scroll, whose text it quotes as well. It is likely that the Severus Scroll and the manuscript it was compared with in the Severus Scroll variant list were both proto-Masoretic in character.24 7. Rabbinic literature contains information about five variant readings toward MT from Gen 1:31 (Gen. Rab. 9:5); 3:21 (Gen. Rab. 20:12); 45:8 (Midrash Bereshit Rabbati 209:12); 46:23 (Gen. Rab. 94:9); and Isa 21:11 (y. Taʿan. 1:1) attributed to scrolls of Rabbi Meir.25 8. Masoretic annotations point to the existence of late ancient or early medieval western and eastern master copies that are now lost,26 such as Affairs, http://mfa.gov.il/MFA/IsraelExperience/History/Pages/Rare-ancient-Hebrew-scrolldeciphered-20-July-2015.aspx (accessed 3/3/16); and “1,500-Year-Old Text Has Been Digitally Resurrected from a Hebrew Scroll,” Smithsonian, http://www.smithsonianmag. com/science-nature/1500-year-old-text-has-been-digitally-resurrected-burned-hebrew-scroll180956031/?no-ist (accessed 3/3/16). 24  For the Severus Scroll variant list, see J. P. S iegel , The Severus Scroll and 1QIsa (SBLMasS 2; Missoula, MT: Scholars Press, 1975). Siegelʼs study of the Severus Scroll variant list is marred by mistakes. For a new edition and analysis of this list, see A. Lange, “The Severus Scroll Variant List in Light of the Dead Sea Scrolls,” in Tradition, Transmission, and Transformation from Second Temple Literature through Judaism and Christianity in Late Antiquity: Proceedings of the Thirteenth International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature Jointly Sponsored by the Hebrew University Center for the Study of Christianity, 22–24 February, 2011 (ed. M. Kister et al.; STDJ 113; Leiden: Brill, 2015), 179–207. 25  For a first exhaustive treatment of Rabbi Meirʼs variants, see Lange, “Rabbi Meir” (see n. 2). 26  For these lost master codices, see Ginsburg, Introduction (see n. 6), 429–441; Yeivin, Introduction (see n. 4), 138–139; M. T. Ortega Monasterio, “El texto de los Códices Modelo según el ʾOr Tôrah de Menaḥem de Lonzano,” in Simposio Biblico Español (Salamanca, 1982) (ed. N. Fernández Marcos et al.; Madrid: Universidad Complutense, 1984), 193–212; E. G. H irsch and I. B roydé , “Bible Manuscripts,” Jewish Encyclopedia 3 (1906), http:// www.jewishencyclopedia.com/articles/3265-bible-manuscripts (accessed 3/3/16). For codex Hilleli, see N. M. S arna , The Pentateuch: Early Spanish Manuscript (Codex Hillely) from the Collection of the Jewish Theological Seminary, New York (Jerusalem: Makhor, 1974); idem , “Introduction to the Hilleli Manuscript,” in Studies in Biblical Interpretation (Philadelphia: Jewish Publication Society, 2000), 239–251; F. Pérez Castro, “Una copia del Codex Hilleli colacionada con la primera mano del MS. B19A de Leningrado,” Sefarad 38 (1978), 13–24; G. Miletto, “Un manoscritto ‘hillelita’ della Biblioteca Palatina di Parma,” Henoch 11 (1989), 271–293. The age of these early master codices might differ significantly. Elijah

128

Armin Lange

Codex Mugah, Codex Hilleli, Codex Zambuki, Codex Jerushalmi, Codex Jericho, Codex Sinai, Codex Ezra, and the Babylonian codex, as well as a codex named the Great Maḥsor. Although the Masorot of several MT manuscripts note variant readings of these codices, there can be little doubt about their proto-Masoretic character. 9. Further information about one or more Hebrew biblical manuscripts from the early Middle Ages is preserved in annotations to one of the Theodulfian Bibles, namely the Saint-Germain Bible (manuscript lat. 11938 of the Bibliothèque nationale de France). Assisting bishop Theodulf of Orléans (ca. 750–821), an anonymous Hebraeus, noted differences between the Theodulfian Vulgate text and one or more Hebrew biblical manuscripts mostly in the margins of this Latin Bible codex. The notes are written in Latin transcription but give the Hebrew readings.27

Levita, for example, claims in Shibre Luḥot that codex Jericho would be most reliable with regard to its accents – thus indicating a relatively late time in which accents were already used – while Zacuto dates Codex Hilleli to ca. 600 CE (see Ginsburg, Introduction [see n. 6], 431). Whether Codex Hilleli was indeed copied at such an early date can at least be questioned, as it was supposed to include the whole Hebrew Bible and would thus have needed to be a codex and not a scroll. The text of Codex Hilleli might be preserved in later manuscripts that claim to be copies of it: Manuscript 44a of the New York Jewish Theological Seminary (see the book by Sarna quoted above) and manuscript Parmense 1996–1997 of the Biblioteca Palatina di Parma (= MTKenn410 and MTDeRossi413; see the article by Miletto quoted above). G insburg , Introduction (see n. 6), 437, claims that a copy of codex Ezra is in his possession (see also Ginsburg, Introduction [see n. 6], 734–743, for a description of this copy). 27  For this Hebraeus and his work, see, e.g., L. D elisle , “Les Bibles de Theodulfe,” Bibliothèque de lʼÉcole des Chartes 40 (1879), 5–47; S. B erger , Quam notitiam linguae hebraicae habuerint Christiani medii aevi temporibus in Gallia (Nancy: Berger-Levrault, 1893), 3–4; E. Power, “The Lost Ninth Century Bible of Carcassone,” Bib 5 (1924), 197–201; idem , “Corrections from the Hebrew in the Theodulfian Manuscripts of the Vulgate,” Bib 5 (1924), 233–258; R. M arsden , The Text of the Old Testament in Anglo-Saxon England (Cambridge Studies in Anglo-Saxon England 15; Cambridge: Cambridge University Press, 1995), 22; A. Candiard and C. Chevalier-Royet, “Critique textuelle et recours à lʼhébreu à lʼépoque carolingienne: Le cas exceptionnel dʼune Bible de Théodulf (Bible de Saint-Germain, ms. Paris, BnF lat. 11937),” in Études dʼexégèse médiévale offertes à Gilbert Dahan par ses élèves (ed. A. Noblesse-Rocher; Bibliothèque de lʼEcole des Hautes Etudes Sciences Religieuses 159; Brepols, 2012), 13–34. The anonymous Hebraeus was identified by A. Saltman (“Rabanus Maurus and the Pseudo-Hieronymian Quaestiones Hebraicae in Libros Regum et Paralipomenon,” HTR 66 [1973], 43–75; idem , Pseudo-Jerome Quaestiones on the Book of Samuel: Edited with an Introduction [StPB 26; Leiden: Brill, 1975], 4–11, 15–17) as the author of Pseudo-Jeromeʼs Questiones Hebraice in libros Regum et Paralipomenon (PL 23, 1391–1470), whom Hrabanus Maurus (780–856) mentioned in his commentary on the books of 1 Samuel–2 Kings.

From Many to One

129

1.3 The Dead Sea Scrolls The manuscript evidence provided by various biblical Dead Sea Scrolls and the Nash Papyrus leaves no doubt about the textual plurality of the Torah in the Second Temple period. Next to the Qumran library, copies of the Torah survive also from the sites related to the first and second Jewish wars. I have discussed and described the Torah manuscripts among the Dead Sea Scrolls in detail in individual articles for each biblical book in THB vol. 128 and in my Handbuch.29 Therefore, I will only summarize their alignments based on what I have argued earlier. In this list, I do not distinguish between excerpted and non-excerpted manuscripts, because I am interested in textual alignment only and because it remains impossible to distinguish between excerpted and complete biblical texts in many biblical Dead Sea Scrolls due to their advanced deterioration. Furthermore, I do not include any of the unpublished biblical manuscripts from the various sites around the Dead Sea. On the whole, forty-five manuscripts from Qumran attesting to parts of the Torah can still be classified according to their text typology. Thirty manuscripts from Qumran are too damaged for text-typological classification, and fifteen further manuscripts could also be remnants of biblical quotations in extrabiblical texts. – Twenty-one non-aligned Qumran manuscripts: 4QRPb,c,d,e, 4QGen f, 2QExoda, 4QExodb,c, 4QExod–Levf, 4QLev–Numa, 4QLevb, 11QpaleoLeva, 1QDeutb, 4QDeutb,c,h,j,k1,k2,n, 4QpaleoDeutr – Two pre-Samaritan Qumran manuscripts: 4QpaleoExodm, 4QNumb – Eleven Qumran manuscripts that are equally close to the text of MT and SP: 4QGenc,e,g,j, 4QLevc,e, 4QDeutd,f,i,o, 5QDeut – Two Qumran manuscripts that are close to the Hebrew Vorlage of the Septuagint:30 4QLevd, 4QDeutq – Two semi-Masoretic Qumran manuscripts: 4QGen-Exoda, 4QpaleoGen– Exodl – Seven proto-Masoretic Qumran manuscripts: 4QDeute,g, 1QIsab, 4QJera, 4QEzeka, 4QXIIe, 2QRutha – Thirty Qumran manuscripts that cannot be classified text typologically because of textual damage: 1QGen, 2QGen, 4QGend,h-title,k, 4QpaleoGenm, 4QpapGeno, 6QpaleoGen, 8QGen, 1QExod, 2QExod b, 4QExodd,e,g,h,j, 1QpaleoLev–Numa, 2QpaleoLev, 4QLevg, 11QLevb, 2QNuma,b, 1QDeuta, 2QDeutb,c, 4QDeuta,l,m,p,t,u, 11QDeut  See Lange, “2.2.1 Pentateuch” (see n. 22).  A. Lange, Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten, Vol. 1 of Handbuch der Textfunde vom Toten Meer (Tübingen: Mohr Siebeck, 2009). 30  The supralinear corrections of 5QDeut also attest to a text close the Hebrew Vorlage of the Deut-LXX (see Lange, “Pentateuch” [see n. 28]). 28 29

130

Armin Lange

– Fifteen Qumran manuscripts for which it is uncertain whether they attest to a biblical book: 4QGenh1,h2,hpara,n,p, 2QExodc, 4QExodk, 1QpaleoLevb?, 6QpaleoLev, 2QNumc,d?, 2QDeuta, 4QDeutk3, 4QpaleoDeuts, 6QpapDeut? One Qumran manuscript was wrongly identified as attesting to a biblical book (4Qpap cryptA Levh?), and another one could be a forgery (XDeut). The evidence for the sites related to the first and second Jewish wars may be summarized as follows: Eight manuscripts can still be classified according to their text typology, while seven are too destroyed for such a classification. To the latter must be added two manuscripts whose origin is uncertain. Two manuscripts of uncertain origin could also represent remnants of biblical quotations in extrabiblical texts. All classifiable manuscripts are proto-Masoretic in character. – – – –

The biblical manuscripts from Masada (First Jewish War) Three proto-Masoretic manuscripts: MasLeva,b, MasDeut One manuscript that cannot be classified text typologically: MasGen The biblical manuscripts from Naḥal Ḥever, Wadi Murabbaʿat, Wadi Sdeir, Naḥal Arugot, and other sites connected with the Second Jewish War Five proto-Masoretic manuscripts: MurGen–Exod + Numa, SdeirGen, 4QGenb, Mur/ḤevLev, XḤev/SeNumb Six manuscripts that cannot be classified text-typologically: Mur(?)Genb, ArugLev, 5/6ḤevNuma, 34ṢeNum, XḤevSeDeut, MurDeut Two manuscripts of uncertain origin that cannot be classified text typologically: ms Schøyen 4612/5, ms Schøyen 5214/2 Two manuscripts of uncertain origin for which it is uncertain whether they attest to a biblical book: ms Schøyen 4612/4, ms Schøyen 5214/1

In addition to manuscripts from the various sites around the Dead Sea, the Nash Papyrus survives from ancient Egypt as an excerpted manuscript attesting to a text that is close to the Hebrew Vorlage of the LXX. My own statistics, detailed above, therefore point to a different understanding of the alignment of the biblical Dead Sea Scrolls from Qumran from that of Tov31 and emphasize the textual plurality of the Torah even more.32 31  Tov developed and refined his classification system in a series of articles: E. Tov, “The Relationship between the Textual Witnesses of the Old Testament in the Light of the Scrolls from the Judean Desert,” Beth Miqra 77 (1979), 161–170 (Hebrew); idem, “A Modern Textual Outlook Based on the Qumran Scrolls,” HUCA 53 (1982), 11–27; idem , “Hebrew Biblical Manuscripts from the Judaean Desert: Their Contribution to Textual Criticism,” JJS 39 (1988), 1–37; idem, “The Significance of the Texts from the Judean Desert for the History of the Text of the Hebrew Bible: A New Synthesis,” in Qumran between the Old and the New Testament (ed. F. H. Cryer and T. L. Thompson; Copenhagen International Seminar 6; JSOTSup 290; Sheffield: Sheffield Academic Press, 1998), 277–309; idem , “Die biblischen Handschriften aus der Wüste Juda – Eine neue Synthese,” in Die Textfunde vom Toten Meer und der Text der Hebräischen Bibel (ed. U. Dahmen, A. Lange, and H. Lichtenberger; Neukirchen-Vluyn: Neukirchener Verlag, 2000), 1–34; idem , “The Biblical Texts from the Judaean Desert: An

131

From Many to One

Table 1: Distribution of Torah Manuscripts according to Textual Typology

Non-aligned

Pentateuch according to Lange

Pentateuch according to Tov33

52.5 percent

39.0 percent

Pre-SP

5 percent

11.0 percent

Vorlage of LXX

5 percent

2.0 percent

Equally close to MT and SP

27.5 percent

Semi-MT34

5 percent

Proto-MT

5 percent

48.0 percent

A word of caution needs to be added with regard to the Septuagint. The above statistics are calculated based only on the Hebrew manuscripts from the Qumran library. But five Greek manuscripts were found in Qumran that attest either to the Old Greek text itself or to a recension of it (4QLXXLeva-b; 4QLXXNum, 4QLXXDeut; 7QpapLXXExod). Furthermore, one scroll from Qumran attests to an Aramaic translation of Leviticus (4QtgLev). The above statistics show that with 10 percent of all Torah manuscripts, proto- and semi-Masoretic texts played a significantly smaller role during the Second Temple period than Tovʼs statistics suggest. This means that the proto-/ semi-Masoretic text was for each book of the Torah only one text that was not Overview and Analysis of the Published Texts,” in The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries (ed. E. D. Herbert and E. Tov; London: British Library, 2002), 139–166; idem, “The Biblical Text in Ancient Synagogues in Light of Judean Desert Finds,” Meghillot 1 (2003), 185–201, viii–ix (Hebrew); idem, “The Text of the Hebrew/Aramaic and Greek Bible Used in the Ancient Synagogues,” in The Ancient Synagogue: From Its Origins until 200 C.E.: Papers Presented at an International Conference at Lund University October 14–17, 2001 (ed. B. Olsson and M. Zetterholm; ConBNT 39; Stockholm: Almqvist & Wiksell International, 2003), 237–259; idem, “The Nature of the Masoretic Text in Light of the Scrolls from the Judean Desert and Rabbinic Literature,” Shnaton 16 (2004), 119–134, x–xi (Hebrew); idem , Textual Criticism of the Hebrew Bible (revised and expanded ed.; Minneapolis: Fortress Press, 2012), 107–111; idem, “The Scribal and Textual Transmission of the Torah Analyzed in Light of its Sanctity,” in Pentateuchal Traditions in the Late Second Temple Period: Proceedings of the International Workshop in Tokyo, August 28–31, 2007 (ed. A. Moriya and G. Hata; JSJSup 158; Leiden: Brill, 2012), 57–72. With the subsequent publication of the biblical Dead Sea Scrolls, Tovʼs statistics changed. In response to the growing number of biblical Dead Sea Scrolls published since the end of the last century, Tov has adjusted the numbers of his statistics continually. 32  For a more detailed comparison of my own statistics with Tovʼs, see L ange , Handschriften (see n. 29), 16–20. 33  The numbers of Tov are quoted from Textual Criticism (see n. 31), 108; see also idem, “Scribal and Textual Transmission” (see n. 31), 58–59. 34  Tov calls this group “MT-like” (Tov, Textual Criticism [see n. 31], 108).

132

Armin Lange

even common in the second half of the Second Temple period. The latter point is underlined by Tovʼs observation that “virtually all rewritten scripture compositions” are based on a text that aligns with the LXX-SP group but not with MT.35 It must be emphasized that the above classifications according to textual typology are no more than a heuristic tool. The group of nonaligned manuscripts does not identify one biblical text, textual family, or text type but summarizes a significant number of manuscripts that cannot be grouped according to texts, textual families, and text types due to their deterioration and due to their lack of textual overlap with each other. This means that in the Second Temple period not only four biblical text types existed but an unknown number of different texts, textual families, and text types. That about 50 percent of all Torah manuscripts from the Qumran library are nonaligned suggests that the nonaligned manuscripts were not the exception but the rule of the textual history of the Torah in the Second Temple period. In stark contrast to this textual plurality, the biblical manuscripts from the sites connected with the first and second Jewish wars paint a different picture. All manuscripts from these sites that can still be classified according to their textual alignment are proto-Masoretic.36 Because the Qumran manuscripts are often earlier than the ones from the other sites around the Dead Sea, it seems likely that the proto-Masoretic text became dominant only very late in the Second Temple period, that is, approximately at the time when the idea of a closed canon was born in Judaism. Furthermore, the precanonical transmission of the biblical books seems to have allowed more freedom for the scribes to change the texts they copied and seems to have granted more room for scribal errors than the transmission of the Hebrew canonical biblical text. My observations thus pose two questions: How did the textual plurality of the books of the Torah come to be, and how did the proto-Masoretic standard text of the Torah develop out of this textual plurality? To answer the first question, I will discuss ways to characterize the nature of biblical texts and the nature of their transmission as opposed to classifying them according to their textual alignment. 35  E. Tov , “2.1 Textual History of the Pentateuch,” in Textual History of the Bible (ed. A. Lange; Brill Online, 2016), http://referenceworks.brillonline.com/entries/textual-history-of-the-bible/21-textual-history-of-the-pentateuch-COM_00000036 (accessed 3/3/16). Last online update: September 2015. First appeared online: 2015. 36  Ulrich has questioned the proto-Masoretic character of the Masada manuscripts by arguing that of Leviticus, Deuteronomy, and Ezekiel no variant literary editions are known from the Qumran library. This would blur the statistical evidence from Masada, as MasLeva, MasLevb, MasDeut, and MasEzek could not have been anything but proto-Masoretic (E. Ulrich , “Two Perspectives on Two Pentateuchal Manuscripts from Masada,” in Emanuel: Studies in Hebrew Bible, Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov [ed. S. M. Paul et al.; VTSup 94; Leiden: Brill, 2003], 453–464). However, in his assessment, Ulrich ignores the distinction between semi- and proto-Masoretic manuscripts, which allows for the above statistical conclusions.

From Many to One

133

Afterward, I will come back to the question of how the proto-Masoretic text became the dominant biblical text in ancient Judaism.

2 Textual Plurality, Variant Analysis, and Types of Textual Transmission 2.1 Textual Typology The study of the textual history of the Hebrew biblical text of the Torah poses unique methodological problems related to the character of the preserved evidence. From the late Second Temple period onward, the extant manuscript evidence is clearly affiliated with only one text, the consonantal text of MT. And beginning with the tenth century, a huge amount of manuscript evidence is available for studying the textual transmission of MT in detail. Rare exceptions to this textual unity are rabbinic quotations that do not align with MT.37 The remaining variation between proto-Masoretic manuscripts does not go beyond a rather limited amount of scribal corruption. For instance, only eight of 860 differences between the Ben Asher and Ben Naphtali texts are concerned with differences in the consonantal text while all others regard differences in punctuation and accentuation.38 Since the publication of Kennicottʼs Bible edition,39 it has therefore become a common view that the consonantal text of MT goes back to one textual archetype.40 37  See Rosen-Zvi, “Rabbinic Literature” (see n. 3); Y. Maori, “The Text of the Hebrew Bible in Rabbinic Writings in Light of the Dead Sea Scrolls,” in The Dead Sea Scrolls: Forty Years of Research (ed. D. Dimant and U. Rappaport; STDJ 10; Leiden: Brill, 1992), 283–289; idem , “Rabbinic Midrash as a Witness of Textual Variants in the Hebrew Bible: The History of the Issue and Its Practical Application in the Hebrew University Bible Project,” in Moshe Goshen-Gottstein: In Memoriam, Vol. 3 of Studies in Bible and Exegesis (ed. M. Bar-Asher et al.; Ramat Gan: Bar-Ilan University Press, 1993), 267–286 (Hebrew); idem , “Rabbinic Midrash as Evidence for Textual Variants in the Hebrew Bible: History and Practice,” in Modern Scholarship in the Study of the Torah: Contributions and Limitations (ed. S. Carmy; The Orthodox Forum Series; Northvale: J. Aronson, 1996), 101–129. 38  See L. L ipschütz, “Kitab al-Khilaf: Mishael ben Uzzielʼs Treatise on the Differences between Ben Asher and Ben Naphtali,” Textus 2 (1962), 3*–57* (Hebrew); idem , Kitāb alKhilaf: Mishael Ben Uzzielʼs Treatise on the Differences between Ben Asher and Ben Naphtali (Publications of the Hebrew University Bible Project 2; Jerusalem: Magnes Press, 1965); Yeivin, Introduction (see n. 4), 141–144. 39  B. Kennicott, Vetus Testamentum hebraicum, cum variis lectionibus (2 vols.; Oxford: Clarendon Press, 1776–1780). 40  Thus first J. G. Eichhorn, Einleitung in das Alte Testament (2nd ed.; Leipzig: Weidmann und Reich, 1787), 111, 113, 203; and E. F. C. Rosenmüller, Handbuch für die Literatur der biblischen Kritik und Exegese (4 vols.; Göttingen: Vandenhoeck & Ruprecht, 1797–1800), 1:244.

134

Armin Lange

For the time before this archetype became generally accepted in late Second Temple Judaism (see below, 3.2), just enough manuscript evidence is preserved to leave no doubt about the textual plurality of the biblical text. There is enough evidence from this period to group the extant manuscripts into textual families and text types. With only few overlaps between the biblical manuscripts from Qumran and the other sites in the Judean desert, a comparative study that could identify manuscript families based on indicative errors is impossible because only few textual overlaps exist between the extant manuscripts. Tov and – based on him – Lange41 derive their typological classifications of the biblical Dead Sea Scrolls from a comparison with the three principal late ancient and medieval versions of the Bible, MT, SP, and LXX. The advantage to this approach is that it allows for a classification of rather deteriorated manuscripts. The disadvantage is that it cannot determine the genealogical relation of individual biblical manuscripts. 2.2 Variant Analysis To a certain extent, the disadvantage to the statistical approach can be compensated for by what I call “variant analysis.”42 For the Hebrew biblical texts, variant analysis should be understood as equivalent to the study of translation technique that relies on primary and secondary translations. Textual characterization of biblical manuscripts by way of variant analysis is based on the study of their secondary readings. It asks for the character of a manuscriptʼs secondary readings and thus not only is able to describe the textual character of the manuscript but also to provide information on how the textual plurality of the Second Temple period developed. At the moment I distinguish several categories of variant readings in the Hebrew texts of the Jewish scriptures. The most basic distinctions are between orthographic and textual variants and between secondary and original readings.

 See above, 1.3 and Lange, Handschriften (see n. 29).  What I discuss in this paragraph is influenced by conversations with Russel E. Fuller and by collaborative research that we are doing. I am obliged to Rus for this teamwork. A first example for the approach of variant analysis can be found in A. Lange, “4QXIIg (4Q82) as an Editorial Text” (forthcoming in Textus). Eshel published a groundbreaking study on harmonizing readings in 4QDeutn, which paved the way for variant analysis (E. Eshel, “4QDeutn – A Text That Has Undergone Harmonistic Editing,” HUCA 62 [1991], 117–154). Of similar importance with regard to abbreviative texts are E. Tov , “Three Manuscripts (Abbreviated Texts?) of Canticles from Qumran Cave 4,” JJS 46 (1995), 88–111; idem , “Excerpted and Abbreviated Biblical Texts from Qumran,” RevQ 16 (1993–1995), 581–600; and with regard to interpretative readings, D. A. T eeter , Scribal Laws: Exegetical Variation in the Textual Transmission of Biblical Law in the Late Second Temple Period (FAT 92; Tübingen: Mohr Siebeck, 2014); idem, “Textgeschichte, Fortschreibung, und Rechtshermeneutik: Das Problem der ‘profanen’ Schlachtung in Lev 17,” Hebrew Bible and Ancient Israel 2 (2013), 286–314. 41

42

From Many to One

135

In the following, I will mostly ignore the question of orthography43 and focus on secondary textual variants. These can be described as corrupt editorial linguistic adjustment and actualization harmonizing abbreviative (excerpted) expansive rewriting/redactional compilational recensional exegetical/interpretative The degree to which these types of secondary variants participate in the overall modes of text production in the ancient Near East becomes evident when the above list is compared to the modes of text production described by van der Toorn: transcription, invention, compilation, expansion, adaption, and integration.44 If the majority of the readings of a textual witness is harmonizing, corrupt, or expansive, this textual witness should be characterized as harmonizing, corrupt, or expansive; a witness may also be a combination of any of the above characterizations. A corrupt text is thus a text the majority of whose secondary readings go back to scribal error (e.g., MT-Hosea). A harmonizing text is intertextual in nature and is characterized by harmonizations of one passage to another one. An editorial text is characterized by small textual changes, usually of a single word or less, which achieve linguistic, stylistic, and orthographic alterations as well as changes that may correct inconsistencies.45 Linguistic actualizations are adjustments in the language of a given text in grammar, syntax, or vocabulary. An example can be found in 1 Sam 20:34, where MT reads ‫וַ יָּ ָקם יְ הֹונָ ָתן ֵמ ִעם‬ ‫ה ֻשּׁ ְל ָחן‬, ַ “Jonathan got up from the table,” instead of ‫ויפחז יהונתן מעל השלחן‬, “and Jonathan shot up from the table,” in 4QSama. The MT reading replaces the rare and archaic word ‫ ויפחז‬with the common ‫ וַ יָּ ָקם‬to make 1 Sam 20:34 comprehensible to the reader.46 A biblical text the majority of whose secondary readings are characterized by such linguistic updating is a linguistically actualizing text. It is often also editorial in nature. Whether a text should be described as expansive, redactional, or a rewriting is often difficult to decide. Depending on how many expansions a text includes and  For orthographic differences, see below, 3.1.2.  K. van der Toorn, Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007), 109–141. 45  For the category of editorial texts, see Lange, “4QXIIg (4Q82)” (see n. 42). 46  A. Lange, “Die Wurzel phz und ihre Konnotationen,” VT 51 (2001), 497–510, 500–501. 43

44

136

Armin Lange

the coherence of these expansions, I would characterize it either as expansive or as a redaction, that is, as a variant literary edition. If a biblical text includes many expansions that are disconnected from each other and accumulated to it over a long time, it should be characterized as expansive. If the expansions of a text and other secondary readings form a coherent layer of text for which one author or group of authors is responsible, it should be described as a variant literary edition or a redaction. The distinction between such a variant literary edition or redaction and a rewriting, otherwise of characterized as Rewritten Bible,47 is a matter of degree. At a certain point, the expansions and alterations to a base text become so significant that a new literary work is born. An example of a redaction or variant literary edition is 4QRPc, as compared to the proto-MT and pre-SP/SP texts of the Torah. Examples of rewritings would of course be the book of Jubilees and the Temple Scroll, whose textual changes to the Torah are even more extensive than the ones applied to the Torah by 4QRPc. The grey area between rewriting on the one hand and redaction or variant literary edition on the other hand is, of course, huge. How to distinguish between a rewriting and a variant literary edition/redaction also often remains a matter of how the canonical history of the Bible is reconstructed. An abbreviated text deletes individual passages from a biblical book. 4QCanta, for example, deletes Cant 4:8–6:10. An excerpted text reduces a biblical book to a collection of several select passages for liturgical or other purposes. An example is 4QDeutn, which now includes only Deut 8:5–10 and 5:1–6:1. Examples of compilational texts are various Psalms manuscripts from Qumran that include a different repertoire and sequence of Psalms, as indicated in the list below. 1. MT-Ps (for MT-Ps the reader is referred to its critical editions) 2. 11QPsa ([. . .] Ps 101–103 → [. . .] → 109 → [. . .] → 118 → 104 → 147 → 105 → 146 → 148 → [. . .] → 121–132 → 119 → 135 → 136 + 118:1, 15, 16, 8, 9, X, 29? → 145 → 154 → Plea for Deliverance → Ps 139 → 137–138 → Sir 51:13–30 → Apostrophe to Zion → Ps 93 → 141 → 133 → 144 → 155 → 142–143 → 149–150 → Hymn to the Creator → 2 Sam 23:[1–]7 → Davidʼs Compositions → Ps 140 → 134 → 151A → 151B [. . .]) 11QPsb (Ps 77 → 78; 119; 118;1, 15–16; Plea for Deliverance; Apostrophe to Zion; Ps 141 → 133 → 144) 47  I have engaged in detail with the terminological debate about the term Rewritten Bible in A. Lange, “In the Second Degree: Ancient Jewish Paratextual Literature in the Context of Graeco-Roman and Ancient Near Eastern Literature,” in In the Second Degree: Paratextual Literature in Ancient Near Eastern and Ancient Mediterranean Cultures and Its Reflections in Medieval Literature (ed. P. S. Alexander et al.; Leiden: Brill, 2010), 3–40, and would like to direct the attention of the reader to this article and the other contributions to this volume regarding the debate about the term Rewritten Bible.

From Many to One

137

4QPse (Ps 76 → 77; 78; 81; 86; 88; 89; 103 → 109; 114; 115 → 116; 118 → 104; 105 → 146; 120; 125 → 126; 129 → 130) 3. 4QPsa (Ps 5 → 6; 25; 31 → 33; 34 → 35 → 36; 38 → 71; 47; 53 → 54; 56; 62 → 63; 66 → 67; 69) 4QPsq (Ps 31 → 33 [ → 34 →] 35) 4. 4QPsb (Ps 91 [→] 92 [→] 93 [→] 94; 96; 98; 99 [→] 100; 102 → 103 [→] 112; 113; 115; 116; [117 →] 118) 5. 4QPsd (Ps 106 → 147 → 104) 6. 4QPsf (Ps 22; 107; 109 → Apostrophe to Zion; Eschatological Hymn [→] Apostrophe to Judah) 7. 4QPsk (Ps 135 [→] 99) A recensional text, or recension, adjusts one biblical text to another. The bestknown example among the Dead Sea Scrolls is 8ḤevXII gr, which adjusts the OG text of the Minor Prophets to their proto-Masoretic text. 8ḤevXII gr attests to an early version of the so-called Kaige recension. Another possible example among the Hebrew Torah manuscripts from Qumran is 5QDeut: in this manuscript, a later hand added four corrections toward the text of LXX-Deut. As these later adjustments seem to be systematic, they can be described as creating a recension toward the Hebrew Vorlage of LXX-Deut. Interpretative readings in Torah manuscripts from Qumran were recently studied in detail by Teeter.48 Two important examples of interpretative readings of 4QDeutj and 4QDeutq can be found in the Song of Moses in Deut 32. In Deut 32:8, MT reads ‫ְּב ַהנְ ֵחל ֶע ְלֹיון ֹּגויִ ם ְּב ַה ְפ ִריֹדו ְּבנֵ י ָא ָדם יַ ֵּצב ּגְ ֻבֹלת ַע ִּמים ְל ִמ ְס ַּפר ְּבנֵ י יִ ְׂש ָר ֵאל‬ When the Most High gave the nations as inheritance, when he apportioned the sons of man, he established the boundaries of the nations according to the number of the sons of Israel.

In reading ‫ּבנֵ י יִ ְׂש ָר ֵאל‬, ְ “sons of Israel,” MT is supported by SP, T, P, and V. The Septuagint and 4QDeutj read υἱῶν θεοῦ (“sons of God”) and ‫“( בני אלוהים‬sons of God” or “sons of the gods”) instead. ὅτε διεμέριζεν ὁ ὕψιστος ἔθνη, ὡς διέσπειρεν υἱοὺς Αδαμ ἔστησεν ὅρια ἐθνῶν κατὰ ἀριθμὸν υἱῶν (ἀγγέλων LXXABOCdnt)49 θεοῦ  Teeter, Scribal Laws (see n. 42).  The text of OG is debated. J. W. W evers (Deuteronomy [2nd ed.; Septuaginta Vetus Testamentum Graecum 3/2; Göttingen: Vandenhoeck & Ruprecht, 2006], 347) decided, based on P.Fouad Inv. 266, to read υἱῶν. Because this papyrus does not preserve the word θεοῦ, Hanhart treats Weverʼs reading as a conjecture and prefers to identify ἀγγέλων as the text of the Old Greek (R. Hanhart, “Die Söhne Israels, die Söhne Gottes und die Engel in der Masora, in Qumran und in der Septuaginta: Ein letztes Kapitel aus ‘Israel in hellenistischer Zeit,’ ” in Ver­gegen­wärtigung des Alten Testaments: Beiträge zur biblischen Hermeneutik: Festschrift für Rudolf Smend zum 70. Geburts­tag [ed. C. Bultmann et al.; Göttingen: Vandenhoeck & Ruprecht, 2002], 170–174). 48

49

138

Armin Lange

When the Most High was apportioning nations, as he scattered Adamʼs sons he fixed boundaries of nations according to the number of sons (angels) of God. (LXX)50 [‫בהנחי[ל עליון גוים בהפרידו בני אדם יצב גבלת עמים למספר] בני אלוהים‬ When the Most High ]gave[ the nations ]as inheri[tance, when he apportioned the sons of man, he established the boundaries of the nations according to the number of the] sons of the gods.[ (4QDeutj)

That the OG text agrees with 4QDeutj in reading ‫בני אלוהים‬, “sons of the gods,” identifies the OG reading as a literal translation of a Hebrew parent text. The meaning of the MT is somewhat garbled. It implies that God would have created as many nations as there were Israelites. The text of 4QDeutj yields a more logical version, which states that the God of Israel, as the head of a pantheon of divine sons, apportioned each of these deities its nation. Deuteronomy 32:8 would thus describe the idea of a national deity for each nation. By implication, the national deity of Israel would be the head of the pantheon of national deities. In the Second Temple period, this polytheistic message of Deuteronomy 32 was offensive to copyists. Some Septuagint copyist thus created a text that read ἀγγέλων (“angels”) instead of υἱῶν (“sons”). To avoid even the implication of angelic semi-divine beings, the scribe who was responsible for the proto-Masoretic text replaced the phrase ‫בני אלוהים‬, “sons ְ “sons of Israel,” thus avoiding any polytheistic of the gods,” with ‫ּבנֵ י יִ ְׂש ָר ֵאל‬, implication in Deut 32:8.51 That out of his general monotheistic interpretation of the Torah the consonantal text of MT introduces a monotheistic signification into the Song of Moses is also evident in Deut 32:43.

 Translation according to NETS.  See B. B arthélemy , “Les Tiqquné Sopherim et la critique textuelle de lʼAncient Testament,” in Congress Volume, Bonn: 1962 (ed. J. A. Emerton; VTSup 9; Leiden: Brill, 1963), 285–304, here 295–304; C. McCarthy, The Tiqqune Sopherim and Other Theological Corrections in the Masoretic Text of the Old Testament (OBO 36; Göttingen: Vandenhoeck & Ruprecht, 1981), 211–214; A. Rofé, Deuteronomy: Issues and Interpretation (London: T&T Clark, 2002), 47–54, here 50; E. Tov, “The Septuagint as a Source for the Literary Analysis of Hebrew Scripture,” in Exploring the Origins of the Bible: Canon Formation in Historical, Literary, and Theological Perspective (ed. C. A. Evans and E. Tov; Acadia Studies in Bible and Theology; Grand Rapids, MI: Baker, 2008), 31–56, here 43; C. McCarthy, Deuteronomy (BHQ 5; Stuttgart: Deutsche Bibelgesellschaft, 2007), 140*–141*; cf. also I. Himbaza, “Dt 32,8, une correction tardive des scribes: Essai dʼinterprétation et de datation,” Bib 83 (2002), 527–548. 50

51

139

From Many to One

Table 2: Comparison of Deuteronomy 32:4352 MT

SP

‫ַה ְרנִ ינּו ֹגויִ ם ַעֹּמו‬ ‫הרנינו גוים עמו‬ ‫ם־ע ָב ָדיו‬ ֲ ‫כי דם עבדיו יקום ִּכי ַד‬ ‫ונקם ישיב לצריו יִ ֹּקום וְ נָ ָקם יָ ִ ׁ֣שיב‬ ‫ְל ָצ ָריו וְ ִכ ֶּפר‬ ‫וכפר אדמת עמו‬ ‫ַא ְד ָמֹתו ַעֹּמו׃‬

Praise, O nations, his people, for he will avenge the blood of his servants and take vengeance on his adversaries and cleanse his land for his people.

Praise, O nations, his people, for he will avenge the blood of his servants and take vengeance on his adversaries and cleanse the land of his people.

4QDeutq ‫הרנינו שמים עמו‬ ‫והשתחוו לו כל אלהים כי‬ ‫דם בניו יקום ונקם ישיב‬ ‫לצריו למשנאיו ישלם‬ ‫ויכפר אדמת עמו‬

Praise, O heavenly beings, his people, worship him, all gods, for he will avenge the blood of his sons and take vengeance on his adversaries, he will repay those who hate him, and cleanse the land of his people.

LXX εὐφράνθητε, οὐρανοί, ἅμα αὐτῷ, καὶ ροσκυνησάτωσαν αὐτῷ πάντες υἱοὶ θεοῦ· εὐφράνθητε, ἔθνη, μετὰ τοῦ λαοῦ αὐτοῦ, καὶ ἐνισχυσάτωσαν αὐτῷ πάντες ἄγγελοι θεοῦ· ὅτι τὸ αἷμα τῶν υἱῶν αὐτοῦ ἐκδικᾶται, καὶ ἐκδικήσει καὶ ἀνταποδώσει δίκην τοῖς ἐχθροῖς καὶ τοῖς μισοῦσιν ἀνταποδώσει, καὶ ἐκκαθαριεῖ κύριος τὴν γῆν τοῦ λαοῦ αὐτοῦ. Be glad, O skies, with him, and let all divine sons do obeisance to him. Be glad, O nations, with his people, and let all angels of God prevail for him. For he will avenge the blood of his sons and take revenge and repay the enemies with a sentence, and he will repay those who hate, and the Lord shall cleanse the land of his people.

I will also not discuss the LXX, which combines the textual differences between MT and SP on the one hand, and 4QDeutq on the other hand, in a double reading. MT and SP have the same three words: ‫“( ַה ְרנִ ינּו ֹגויִ ם ַעֹּמו‬praise, O Nations, his people”). Instead of addressing the nations, 4QDeutq unfolds a polytheistic scenario reminiscent of the text of Deut 32:8 in 4QDeutj and in the Septuagint: ‫“( הרנינו שמים עמו והשתחוו לו כל אלהים‬Praise, O heavenly beings, his people, worship him, all gods”). The idea that other deities should venerate the God of Israel is not unique to Deut 32:43 but occurs also in Pss 29:1 and 97:7 (in the same wording as in Deut 32:43). It is part of a motif that occurs several times in Israelʼs psalms, namely the dominance of the God of Israel over other gods and their invalidity (e.g., Psalm 82; 95:3; 96:4–5). While it is quite likely that the Song of Moses employs a motif out of Israelʼs hymnic tradition to describe how other gods venerate the God of Israel in order to emphasize his triumph over them, I cannot imagine that in Judea of the later Second Temple period a scribe would have inserted such a polytheistic addition 52

 English translation of LXX text (fourth column) according to NETS.

140

Armin Lange

into the text of the Torah. It is much more likely that the proto-Masoretic text removed the polytheistic conception from Deut 32:43 by deleting the phrase ‫ והשתחוו לו כל אלהים‬and by exchanging ‫ שמים‬with ‫גוים‬. As in Deut 32:8, the proto-Masoretic text creates a monotheistic signification for Deut 32:43 on the basis of its general monotheistic reading of the Torah. In Deuteronomy 32, the proto-Masoretic text was thus created on the basis of an overall monotheistic interpretation of the Torah.53 2.3 Types of Textual Transmission Another question is how a harmonizing, corrupt, or expansive text is transmitted. Some books of the MT, like Hosea, suffer from scribal corruption. Nevertheless, they have been transmitted with great care and without scribal license or corruption for a very long time. The example of MT-Hosea demonstrates that the textual character of a manuscript and the character of the transmission of a biblical text need to be distinguished. That is, a corrupt textual witness was sometimes transmitted in a very conservative way. It is hence not enough to characterize a text as corrupt in the Second Temple period. A comparison with later biblical manuscripts is needed to judge the way in which such a corrupt textual witness was transmitted and regarded after the corruption occurred. On the whole, three basic types of textual transmission can be distinguished: 1. Conservative, or careful, transmission, in which a biblical text is copied accurately without scribal license or scribal corruption 2. Free, or liberal, transmission, which is characterized by an ongoing process of intentional scribal intervention 3. Corrupt, or careless, transmission, which is characterized by sloppiness in copying a text 53  Cf. P. W. Skehan, “A Fragment of the ‘Song of Moses’ (Deut. 32) from Qumran,” BASOR 136 (1954), 12–15; idem , “The Qumran Manuscripts and Textual Criticism,” in Volume du Congrès International pour lʼétude de lʼAncien Testament, Strasbourg 1956 (ed. G. W. Anderson et al.; VTSup 4; Leiden: Brill, 1957), 148–160, here 149–150; F. M. Cross, The Ancient Library of Qumrân and Modern Biblical Studies (London: Duckworth, 1958), 135–136 (3rd ed.; Sheffield: Sheffield Academic Press, 1995, 133–135); W. F. Albright, “Some Remarks on the Song of Moses in Deuteronomy xxxii,” VT 9 (1959), 338–346, here 340–341; M. M ansoor , “The Massoretic Text in the Light of Qumran,” in Congress Volume, Bonn 1962 (ed. J. A. Emerton; VTSup 9; Leiden: Brill, 1963), 305–321, here 308–309; P. W. Skehan and E. Ulrich, “44. 4QDeutq,” in Qumran Cave 4.IX: Deuteronomy, Joshua, Judges, Kings (ed. E. Ulrich et al.; DJD XIV; Oxford: Clarendon Press, 1995), 137–142, here 138; J. H. Tigay, Deuteronomy (The JPS Torah Commentary; Philadelphia: Jewish Publication Society, 1996), 516–518; Hanhart, “Söhne Israels” (see n. 49), 171; Rofé, Deuteronomy (see n. 51), 50; Tov, “Septuagint as a Source” (see n. 51), 41–43. J. Lust, “The Raised Hand of the Lord in Deut 32:40 according to MT, 4QDeutq, and Lxx,” Textus 18 (1995), 33–45; A. van der Kooij, “The Ending of the Song of Moses: On the Pre-Masoretic Version of Deut 32:43,” in Studies in Deuteronomy, Festschrift J. C. Labuschagne (F. García Martínez et al.; VTSup 53; Brill: Leiden, 1994), 93–100, here 94–99.

141

From Many to One

For texts that achieved scriptural or canonical authority, the last category of textual transmission is, of course, rarer than the first. Furthermore, in the textual history of a given biblical book, several types of textual transmission may occur. In the beginning, the textual transmission of a given biblical book might have suffered from careless transmission, while later on it was transmitted carefully. 2.4 The Example of 4QDeutn (4Q41) as an Excerpted Harmonizing Text Given the constraints of this article, I will give only one example of variant analysis and textual transmission: 4QDeutn. Among the fifty-one variant readings of 4QDeutn toward MT, SP, and LXX, Eshel54 identified seventeen harmonizing readings. These harmonizing readings are especially common in Deut 5:6–21 and adjust the Deuteronomic Decalogue to Exod 20:2–17. 4QDeutn attests to both small and extensive harmonizations. The table below summarizes the smaller harmonizations of 4QDeutn in comparison with MT, SP, and Exod 20:2–17. Table 3: Harmonizations with Exodus 20:2–17 in 4QDeutn MT

SP

4QDeutn

Exod 20:2–17 MT

Deut 5:8

‫“ ָכּל‬any”

‫“ וכול‬or any”

‫“ וכול‬or any”

‫“ וְ ָכל‬or any”

Deut 5:9

‫ל־שׁ ֵלּ ִשׁים‬ ִ ‫“ וְ ַע‬and upon the third (generation)”

‫על שלשים‬ “upon the third (generation)”

‫על שלשים‬ “upon the third (generation)”

‫ל־שׁ ֵלּ ִשׁים‬ ִ ‫ַע‬ “upon the third (generation)”

Deut 5:14

‫ל־בּ ֶה ְמ ֶתָּך‬ ְ ‫“ וְ ָכ‬and any of your cattle”

‫“ וכל בהמתך‬and any of your cattle”

‫“ ובהמתך‬and your cattle”

‫וּב ֶה ְמ ֶתָּך‬ ְ “and your cattle”

Deut 5:18, 19, 20

‫“ וְ לֹא‬and [you shall] not”

‫[“ לא‬you shall] not”

‫[“ לוא‬you shall] not”

‫[“ לֹא‬you shall] not”

Deut 5:21

‫“ וְ לֹא ַת ְחמֹד‬and you shall not covet”

‫“ לא תחמוד‬you shall not covet”

‫“ לוא תחמוד‬you shall not covet”

‫א־ת ְחמֹד‬ ַ ֹ ‫“ ל‬you shall not covet”

Deut 5:2155

‫וְ לֹא ִת ְתאַוֶּ ה ֵבּית‬ ‫“ ֵר ֶעָך‬and you shall not crave your neighborʼs house”

‫לא תחמוד בית‬ ‫“ רעך‬you shall not covet your neighborʼs house”

‫לוא תחמוד בית‬ ‫“ רעיך‬you shall not covet your neighborʼs house”

‫לֹא ַת ְחמֹד ֵבּית‬ ‫“ ֵר ֶעָך‬you shall not covet your neighborʼs house”

54  The following description of the harmonizing variants of 4QDeutn is guided by Eshel, “4QDeutn” (see n. 42), 117–154. 55  In SP and MT-Exod 20:17, ‫ל ֹא ַת ְחמֹד ֵּבית ֵר ֶעָך‬, “you shall not covet your neighborʼs house,” is the first stichos of the tenth commandment, while in MT-Deut 5:21 and 4QDeutn it is the second.

142

Armin Lange

The only larger harmonization of the Decalogue in 4QDeutn is a long addition to the Sabbath command at the end of Deut 5:15 that is taken from Exod 20:11a.56 Table 4: Deuteronomy 5:15 and Exodus 20:11a Deuteronomy 5:15 according to 4QDeutn 57

Deuteronomy 5:15 according to MT

Exodus 20:11

You shall remember that you were a servant in the land of Egypt, and the Lord your God brought you out from there with a mighty hand and an outstretched arm. Therefore the Lord your God commanded you to keep the Sabbath day to hallow it. For in six days the Lord made heaven and earth, the sea, and all that is in them and rested the seventh day; so the Lord blessed the Sabbath day to hallow it.

You shall remember that you were a servant in the land of Egypt, and the Lord your God brought you out from there with a mighty hand and an outstretched arm. Therefore the Lord your God commanded you to perform the Sabbath.

For in six days the Lord made heaven and earth, the sea, and all that is in them and rested the seventh day; so the Lord blessed the Sabbath day and hallowed it.

Because the two versions of the Decalogue give different reasons for the Sabbath command, the scribe of 4QDeutn or its Vorlage has combined them by inserting a part of the Exodus Decalogue into the Deuteronomy Decalogue. But 4QDeutn does not restrict itself to harmonizations with Exod 20:2–17. Both inside and outside the Decalogue, 4QDeutn harmonizes its Deuteronomy text with other passages in Deuteronomy or Exodus as well.58

 See, e.g., J. H. T igay , “Conflation as a Redactional Technique,” in Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 53–95, here 55–57; S. A. White , “The All Souls Deuteronomy and the Decalogue,” JBL 109 (1990), 193–206, here 200–201; S. W. C rawford , “Reading Deuteronomy in the Second Temple Period,” in Reading the Present in the Qumran Library: The Conception of the Contemporary by Means of Scriptural Interpretations (SBLSymp 30; Atlanta: Society of Biblical Literature, 2005), 127–140, here 129–130; idem , Rewriting Scripture in Second Temple Times (Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans, 2008), 31–32; Eshel, “4QDeutn” (see n. 54), 145–146; J. A. Duncan, “Excerpted Texts of Deuteronomy at Qumran,” RevQ 18 (1997–1998), 43–62, here 55–56. 57  Translation according to M. Abegg et al., The Dead Sea Scrolls Bible (Edinburgh: T&T Clark, 1999), 154. Differences between MT and 4QDeutn are marked in italics. 58  This list is compiled from Eshelʼs study (“4QDeutn”). For a complete discussion of these harmonizations, see Eshel, “4QDeutn” (see n. 55), 142–147. 56

From Many to One

143

1. ‫אלוהיכם‬, “your God,” was added to ‫יהוה‬, “Lord,” in Deut 5:5, under the influence of forty-two parallel expressions in Deuteronomy alone that speak of “the Lord, your God.”59 2. Deuteronomy 5:14 is harmonized with Exod 35:2 by adding ‫בו‬, “on it,” and with Exod 16:26 and other verses by the reading ‫וביום השביעי‬, “and on the seventh day,” instead of MTʼs ‫יעי‬ ִ ‫וְ יֹום ַה ְשּׁ ִב‬, “and the seventh day.” 3. Deuteronomy 5:15bβ is harmonized with Deut 5:12a. 4. Deuteronomy 8:6 is harmonized with “texts in Deuteronomy dealing with the general command ‘to love God’ (such as 11:22; 19:9, et al.).”60 5. Deuteronomy 8:7 is harmonized with Exod 3:8. 6. Deuteronomy 5:22 is harmonized with Deut 4:11. Because 4QDeutn is an excerpted text that currently includes only Deut 8:5–10 and 5:1–6:1, we do not know how extensive the harmonizations of its parent text were outside the two passages discussed above. Nevertheless, the preserved harmonizations demonstrate how a new text branched off a textual tradition that was close to the medieval Masoretic text, that is, how a new text developed and added to the textual plurality of the Second Temple period. The readings of 4QDeutn are thus characterized by conscious harmonization of an early Torah text. The harmonizing text of 4QDeutn should hence be described as a free or liberal text that took license to change its parent text but is not characterized by scribal corruption. Because no other manuscripts of this particular Torah text survive, we do not know whether its further transmission was conservative, liberal, or careless.

3 Textual History of the Hebrew Texts, Their Textual Character, Date, and Milieu The description of the most important extant Torah manuscripts (see 1) and the methodological approaches to analyze them (see 2) have already highlighted that the Hebrew Torah text went through several stages in its textual history. On the whole, six different stages can be distinguished, including todayʼs electronic editions: (1) the precanonical transmission of the Hebrew text of each biblical book, (2) the textual transmission of the canonical text of the Hebrew Bible before the work of the Masoretes, (3) the period of the Masoretes until the compilation of the Ben Asher master codices, (4) the period from the Ben Asher master codices until the invention of the printing press and the first printed editions of the Hebrew Bible, (5) the transmission of the Hebrew Bible in printed form and the printed critical editions, (6) the transmission of the Hebrew Bible 59 60

 Eshel, “4QDeutn” (see n. 54), 144.  Eshel, “4QDeutn” (see n. 54), 144.

144

Armin Lange

in electronic form on CD or online. Needless to say, overlaps and gray areas exist between each of these epochs. The present sketch of the Hebrew textual history of the Torah will focus only on the two periods before the work of the Masoretes began. 3.1 The Qumran Biblical Scrolls and Textual Plurality and Orthographic Plurality of the Second Temple Period In addition to the textual plurality described above, the Torah scrolls of the Second Temple period attest to a limited orthographic plurality. In principal, two main systems can be identified: the conservative orthography of MT and the baroque orthography and morphology prominent in several Dead Sea Scrolls. Furthermore, MT includes individual spellings that point to one or more additional orthographic systems not in use even in the Qumran library. In this part of my article, I will first discuss the textual plurality of the Second Temple period and then look into its orthographic plurality. 3.1.1 The Textual Plurality of the Torah in the Second Temple Period in Light of the Dead Sea Scrolls The above lists of biblical Dead Sea Scrolls and their textual affiliation (see 1.3) described the textual plurality of the Torah in the Second Temple period. They also show that by the first and second Jewish wars, the proto-Masoretic standard text of the Hebrew Bible in general and the Torah in particular existed. The Torah manuscripts from the various sites around the Dead Sea allow for conclusions as to which biblical scrolls of which affiliation were owned and most likely used by either the Qumran community or the groups occupying Masada and the refugee caves around the Dead Sea during the first and second Jewish wars. Furthermore, the biblical scrolls from Qumran indicate the affiliations of the texts that existed in Judea since the third century BCE, the time when the earliest biblical scrolls at Qumran were copied. Table 5 summarizes the evidence: Table 5: Text-Typological Chronology of the Hebrew Manuscripts of Biblical Books from the Dead Sea Semi-MT

Proto-MT

Semi-MT or pre-SP

ca. 250 BCE 200–150 BCE

Pre-SP

Vorlage of NonLXX aligned 4QExod– Levf

5QDeut

ca. 150 BCE

4QLev– Numa

150–100 BCE

4QDeutb,c

145

From Many to One

Semi-MT

Proto-MT

125–75 BCE

Semi-MT or pre-SP

Pre-SP

Vorlage of NonLXX aligned

4QDeutd

125–100 BCE

4QGen– Exoda

100–25 BCE

4QpaleoGen–Exodl

4QpaleoExodm

4QRPe, 4QpaleoDeutr

100–50 BCE

4QDeuti

75–50 BCE

4QDeutf,o

4QRPb,c,d

ca. 50 BCE

4QGeng

4QGenf, 4QLevb

50–1 BCE 50–25 BCE

4QDeutq 4QDeut

e

4QGen

30 BCE–20 CE

4QDeuth 4QExodc

e,j

4QNumb

4QLevd

4QExodb, 1QDeutb

30–1 BCE

MasLeva,b, 4QLeve MasDeut

4QDeut-

1–50 CE

4QDeutg

11QpaleoLeva

20–68 CE

4QGenc

ca. 50 CE 50–100 CE

4QDeutj 4QGenb (not from Qumran), SdeirGen, Mur/ḤevLev, XḤev/ SeNumb

50–68 CE 100–135 CE

k1,k2,n

4QLevc

2QExoda

MurGen-Exod, Numa, MurXII

Taking manuscripts of other biblical books than the Torah into consideration as well, the textual plurality of the Qumran library demonstrates that at least since the third century BCE, texts that were close to the Hebrew Vorlage of LXX were as much at home in Judea (4QSamb and 4QJerb) as were proto-Masoretic manuscripts like 4QJera and nonaligned texts like 4QExod–Levf. Because from the time before the middle of the second century BCE only very few manuscripts survive, little more can be said beyond the general recognition of the textual

146

Armin Lange

plurality of this period. Beginning with the second century BCE, texts that are affiliated loosely or closely with MT, SP, and LXX as well as various nonaligned texts are attested. Beginning in the second half of the second century BCE, biblical manuscripts in general, and Torah manuscripts in particular, of all textual affiliations begin to appear in great numbers. This situation does not change in Qumran until the destruction of the Qumran settlement in 68 CE. As for the Essene community of Qumran itself, the find of several inkwells in the Qumran settlement61 make it likely that scribes of this community copied some of the manuscripts found in the eleven caves around Qumran. This is corroborated by the fact that the hands of several scribes can be identified as having copied more than one scroll. Because the overwhelming majority of biblical scrolls cannot be attributed to specific scribes but were copied by different hands, it is also probable that most biblical scrolls were brought to Qumran from elsewhere.62 The textual plurality of the Qumran library thus reflects the textual affiliation of biblical scrolls in Judea in general and in the Qumran community in particular. From at least five Qumran scribes, more than one biblical manuscript is extant.63 The textual affiliation of these manuscripts copied by the same scribe could provide information about his textual preferences and the textual prefer61  For these inkwells, see R. de Vaux, “Fouilles au Khirbet Qumran: Rapport préliminaire sur la dernière campagne,” RB 61 (1954), 206–233, esp. 212, and plates 5, 6, 10b; idem , Archeology and the Dead Sea Scrolls: The Schweich Lectures of the British Academy 1959 (London: British Academy 1973), 29–30; S. Goranson, “Qumran: A Hub of Scribal Activity,” BAR 20 (1994), 36–39; idem, “An Inkwell from Qumran,” Michmanim 6 (1992), 37–40 (Hebrew); E. Tov, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004), 54–55; J. Saukkonen, “A Few Inkwells, Many Hands: Were There Scribes at Qumran?” in Houses Full of All Good Things: Essays in Memory of Timo Veijola (ed. J. Pakkala and M. Nissinen; Helsinki: Finnish Exegetical Society, 2008), 538–553; K. L. Rasmussen et al., “The Constituents of the Ink from a Qumran Inkwell: New Prospects for Provenancing the Ink on the Dead Sea Scrolls,” Journal of Archeological Science 39 (2012), 2956–2968. 62  How many scrolls were produced at Qumran is a matter of debate (see Tov , Scribal Practices [see n. 61], 14–16). H. Stegemannʼs idea of the Qumran settlement as a kind of Essene publishing house (H. Stegemann , Die Essener, Qumran, Johannes der Täufer und Jesus: Ein Sachbuch [4th ed.; Freiburg i.B.: Herder, 1994], esp. 77–82) overemphasizes the evidence by far (see F. Rohrhirsch, “Die Geltungsbegründungen der Industrie-Rollen-Theorie zu Chirbet Qumran und En Feschcha auf dem methodologischen Prüfstand: Relativierung und Widerlegung,” DSD 6 [1999], 267–281). 63  A complete list of all scribes from whom more than one manuscript survives in the Qumran library is provided by Tov, Scribal Practices (see n. 61), 23–24. See also E. Ulrich, “4QSamc: A Fragmentary Manuscript of 2 Samuel 14–15 from the Scribe of Serek Hay-yaḥad (1QS),” BASOR 235 (1979), 1–25; idem, “The Identification of a Scribe Active at Qumran: 1QPsb – 4QIsac – 1QM,” Meghillot 5–6 (2007), *201–*210; and E. Tigchelaar, “In Search of the Scribe of 1QS,” in Emanuel: Studies in the Hebrew Bible, Septuagint and Dead Sea scrolls in honor of Emanuel Tov (ed. S. M. Paul et al.; VTSup 94,1; Leiden: Brill, 2003), 439–452;

147

From Many to One

ences of his group. The table below lists the nonbiblical and biblical manuscripts produced by five scribes from Qumran and their textual affiliations. Table 6: Scribes Who Copied More Than One Biblical Scroll Nonbiblical manuscripts

Biblical manuscripts

Scribe 1 1QS (incl. 1QSa and 1QSb); 4QNarrative G; 4QTQahat ar; 4QIndividual Thanksgiving A; 4QPersonal Prayer; 4QEschatological Hymn

4QSam

Scribe 2

4QGenf

Corrections in 1QIsa 4QTest

4QGen Scribe 3 Scribe 4 11QM

nonaligned, close to OG a

toward MT64 biblical quotations from the Torah of varying textual affiliations (see below) nonaligned either MT or SP

g k1

nonaligned

4QDeutk2

nonaligned

1QPs

too destroyed

4QDeut b

4QIsa Scribe 5

Textual affiliation

c

c

nonaligned

1QpaleoLev–Numa

too destroyed

6QpaleoGen

too destroyed

Especially interesting are scribe 1 and scribe 2. Scribe 2 copied more than one biblical manuscript. Scribe 1 copied a biblical manuscript, corrected another one, and copied a collection of quotations of varying textual affiliations. 4QSamc, copied by scribe 1, is nonaligned in affiliation, but the corrections that scribe 1 added to 1QIsaa adjust this manuscript at several points toward the text of MT-Isa. Because scribe 1 worked before the canon of the Hebrew Bible existed, he did not necessarily view MT-Isa and MT-Sam as part of the same text. This observation thus implies only that a nonaligned text of Samuel was copied by scribe 1 and that scribe 1 corrected a nonaligned text of Isaiah toward MT-Isaiah. It is all the more interesting that scribe 1 also compiled a list of messianic quotations in the Torah that is called 4QTestimonia today. Each of these quotations has a different textual affiliation. In lines 1–8, 4QTestimonia quotes Exod 21:21b according to its pre-Samaritan text type. Then in lines 9–13, for Num 24:15–17, this collection of messianic references uses a proto-Masoretic see also J. R. Davila, “7. 4QGeng,” in Qumran Cave 4.VII: Genesis to Numbers (ed. E. Ulrich et al.; DJD XII; Oxford: Clarendon Press, 1994), 57–60, here 57. 64  Cf. E. U lrich , “The Developmental Composition of the Book of Isaiah: Light from 1QIsaa on Additions in the MT,” DSD 8 (2001), 288–305.

148

Armin Lange

text. This proto-Masoretic quotation is followed by a quote from Deut 33:8–11 according a nonaligned text type also attested by 4QDeuth.65 But not only scribe 1 regarded biblical texts of different textual affiliations as equally valid. The same is true of scribe 2, who copied the book of Genesis once in a nonaligned text and once in a text that was either close to MT or to SP. Given that neither scribe 1 nor scribe 2 seems to have had preferences with regard to textual affiliation, it is likely that the Qumran community at large was indifferent to the textual affiliation of the biblical scrolls they used. The quotations of and allusions to biblical books in nonbiblical literature from the Second Temple period demonstrate the same phenomenon for other religious groups as well.66 Two examples illustrate this phenomenon. Nehemiah 10:35 relates that priests, Levites, and the people should use the oracle of the lot to determine the sequence in which they should bring the wood offering to the house of God (cf. Neh 13:31).67 The precise phrase used to describe the sequence is ‫ית־אב ֵֹתינוּ‬ ֲ ‫ל ֵב‬,ְ “according to the house of our fathers.” Nehemiah 10:35 claims that this is a prescription written in the Torah (‫ַכּ ָכּתוּב‬ ‫תֹּורה‬ ָ ‫בּ‬, ַ “as it is written in the Torah”). Although Neh 10:35 states explicitly that this procedure is commanded in the Torah, no such prescription is attested either in the Masoretic Pentateuch nor in any other of its versions. The lack of such a command in the Pentateuch evoked various explanations. Clines, for example, speaks of the “creation of facilitating law.”68 Blenkinsopp argues that Neh 10:35 “is implicit in the requirement that the priests keep the fire for the morning and evening sacrifice burning continuously, for which fuel was required (Lev. 6:2, 5–6[9, 12–13]).”69

65  See J. A. Duncan, “35. 4QDeuth,” in Qumran Cave 4.IX: Deuteronomy, Joshua, Judges, Kings (ed. E. Ulrich et al.; DJD XIV; Oxford: Clarendon Press, 1995), 61–70, esp. 69; Tov, “Biblical Texts” (see n. 31), 149–150; Crawford, Rewriting Scripture (see n. 56), 35–36. 66  For the quotations of and allusions to the other biblical books, see Lange, Handschriften (see n. 29). For a more detailed discussion of the textual plurality of Jewish scriptures in Hellenistic times based on their use in quotations and allusions, see A. Lange, “From Literature to Scripture: The Unity and Plurality of the Hebrew Scriptures in Light of the Qumran Library,” in One Scripture or Many? Canon from Biblical, Theological, and Philosophical Perspectives (ed. C. Helmer and C. Landmesser; Oxford: Oxford University Press, 2004), 51–107, here 77–81, 98–100. 67  For my ideas concerning Neh 10:35, see also Lange, “From Literature to Scripture” (see n. 66), 51–107, here 77–80; idem, “The Dead Sea Scrolls and the Date of the Final Stage of the Pentateuch” (see headnote), 300–302. 68  D. J. A. Clines, “Nehemiah 10 as an Example of Early Jewish Biblical Exegesis,” JSOT 21 (1981), 111–117, esp. 112; see also H. G. M. W illiamson , Ezra, Nehemiah (WBC 16; Waco, Tex.: Word Books, 1985), 336. 69  J. Blenkinsopp, Ezra-Nehemiah: A Commentary (OTL; Philadelphia, Pa.: Westminster, 1988), 317.

From Many to One

149

Against such exegetical remedies, it needs to be stated that the sequence in which the wood offering is to be brought to the temple is regulated in an expansion of 4QRPc that follows Lev 24:2, namely, 4QRPc 23 9–11.70 9 ] the [fe]stival of fresh oil, they shall bring wood two [ 10 ] the ones who bring on the fir[st] day, Levi [ 11 Reu]ben and Simon [and on t]he forth day [

In contrast to Neh 10:35, 4QRPc (4Q365) 23 9–11 mentions neither the oracle of the lot nor families. Instead, a sequence is given for the tribes of Israel in which they should provide wood for the wood offering. Although this is not an exact parallel, in my opinion, Neh 10:35 attests to an interpretative use of the prescription of 4QRPc.71 A close parallel to 4QRPc 23 9–11 can be found in the Temple Scroll (11QTb VI 11–15 [lines 16–7 overlap with 11QTa XXIII 3–4]):72 (11) [And after the festival of fresh oil, they shall bring,] (12) to the alta[r the woo]d, [namely the twelve tribes of the people of Israel, and they shall offer: on the first day] (13) the tribes[ of Levi ]and Judah, on the [second ]d[ay Benjamin and the sons of Joseph, on the third day, Reuben and Simon] (14) on the fourth day Issachar [and Ze]bulon, [on the fifth day Gad and Asher, on the sixth day Dan] (15) and Naphtali.

It has been argued that in their passages on the feast of the wood offering, both the text of 4QRPc and the Temple Scroll echo Nehemiah 10:35.73 Below I will argue that the Temple Scroll also employs other expansions that are attested by the manuscript 4QRPc. Given this general dependence of the Temple Scroll on the 4QRPc text of the Torah, it seems more likely that the Temple Scroll depends on the text copied in 4QRPc (4Q365). This means that both the Temple Scroll and the Book of Ezra-Nehemiah use the 4QRPc text of the Torah independently of each other. The two uses of the 4QRPc text of the Torah by the book of Nehemiah and the Temple Scroll and the use of three different Torah texts in 4QTestimonia demonstrate that in the Second Temple period Jews were willing to peruse Torah texts of different textual affiliations. That both inside and outside the Qumran community textual plurality was the rule and not the exception in ancient Judaism argues against the idea of group-specific texts. In this context, it should also be noted that no “sectarian” variants, that is, variant readings that advocate 70  Translation according to E. Tov and S. A. White, “Reworked Pentateuch,” in Qumran Cave 4.VIII: Parabiblical Texts, Part 1 (ed. H. Attridge et al.; DJD XIII; Oxford: Clarendon Press, 1994), 187–351, here 293. 71  Against Tov and White, “Reworked Pentateuch” (see n. 70), 295. 72  For the reconstruction and translation of 11QTb VI:11–15, see F. García Martínez et al., “20. 11QTempleb,” in Qumran Cave 11.II: 11Q2–18, 11Q20–30 (ed. F. García Martínez et al.; DJD XXIII; Oxford: Clarendon Press, 1998), 357–409, esp. 381–83. The reconstruction is based on 11QTa and 4QRPc. 73  Y. Yadin, Temple Scroll (Jerusalem: Israel Exploration Soc., 1977–1983), 1:123–24, 128.

150

Armin Lange

the ideology of one of the religious groups of Second Temple Judaism, has been identified to date.74 A few general structures emerged from the above survey for the Second Temple period: Beginning with the third century BCE, there is evidence of a textual plurality of the Jewish scriptures in general and the Torah in particular. The textual plurality of the quotations of and allusions to the Torah as well as the heterogeneous character of the Qumran library, which consists of Essene and non-Essene manuscripts, show that the textual plurality of the Qumran library reflects the overall textual reality of the Torah in ancient Judaism. It seems unlikely that the proto-Masoretic text served as a standard or even dominant text of the Torah for most of the Second Temple period. It is also unlikely that specific groups of Second Temple Jews preferred one Torah text above the others. 3.1.2 The Orthographic Plurality of the Second Temple Period in Light of the Dead Sea Scrolls75 Two orthographic systems dominate the Hebrew biblical scrolls, which survive from the Second Temple period and the Masoretic text: (1) a baroque system, which is characterized by a lavish use of plene spellings and a morphology that prefers elongated suffixes and affixes, and (2) a conservative system, which is known from the consonantal text of MT and uses fewer plene spellings and a shorter morphology. Both systems were employed in various Torah manuscripts among the Dead Sea Scrolls. In my description of the baroque system, I will not distinguish between morphology and orthography, because it remains uncertain whether, for example, ‫ נָ ַת ָּתה‬and ‫ נָ ַת ָּת‬reflect two morphologically different forms that were pronounced differently in antiquity or two orthographic realizations of the same form. In the latter case, as indicated by the masoretic vocalization, the additional he of ‫ נָ ַת ָּתה‬would serve merely as a mater lectionis for two different spellings that were pronounced in the same way.

 E. Ulrich, “The Absence of ‘Sectarian Variants’ in the Jewish Scriptural Scrolls Found at Qumran,” in The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries (ed. E. D. Herbert and E. Tov; London: British Library, 2002), 179–195 (republished in Hebrew translation in Meghillot 2 [2004], 115–135), has shown that with the possible exception of SP, no ideological variants can be identified in the extant textual witnesses of the Jewish scriptures from ancient Judaism: “no variants emerged to indicate that any group – whether Pharisaic, Sadducean, Samaritan, Essene, Christian, or other – had tampered with Scripture in order to bolster their particular beliefs, except for the two Samaritan beliefs that God ‘had chosen’ Mount Gerizim and had commanded the central altar there, and possibly except for the Mount Gerizim/Mount Ebal variant” (191). 75  The following discussion of the orthographic plurality of the Second Temple period is a summary of a more extensive study that I published in A. L ange , “The Question of the So-Called Qumran Orthography, the Severus Scroll, and the Masoretic Text,” in Hebrew Bible and Ancient Israel 3 (2014), 424–475. 74

From Many to One

151

Several detailed descriptions of the baroque orthographyʼs characteristics exist,76 and the reader is referred to these descriptions beyond the following short summary. With few exceptions, baroque orthography indicates ḥôlem, qāmeṣ ḥāṭûp, ḥāṭēp qāmeṣ, and šûreq with the mater lectionis wāw; ḥîreq, ṣērê, and sometimes even sĕgôl are indicated by yôd. Rather common are the digraphs ‫וא‬, ‫או‬, and ‫ יא‬as matres lectionis for these vowels. The verbal affix of the second-person masculine singular is spelled with a hēʾ at the end (e.g., ‫נָ ַת ָּתה‬, “you gave”), and the pronominal suffix of the second person masculine singular is also spelled with a hēʾ at the end (e.g., ‫ּב ִׁש ְמ ָכה‬, ְ “in your name”). Other verbal affixes and pronominal suffixes attest to similarly lengthened forms. Furthermore, lengthened forms occur with various types of nouns, pronouns, conjunctions, adverbs, and other parts of speech (e.g., ‫“ הואה‬he”; ‫“ היאה‬she”; ‫“ כיא‬because”; ‫מואדה‬, ‫מאודה‬, and ‫מודה‬, all meaning “very”). Long verbal forms, known from the MT as pausal forms, occur regularly (e.g., ‫תקטולו‬, “you will kill,” and ‫יקטולו‬, “they will kill”). The history of research on the baroque orthography of many Dead Sea Scrolls is extensive.77 In principal, three positions can be distinguished: (1) Tov regards it as a characteristic of Qumran scribes and as part of the Qumran scribal practice.78 (2) Freedman claims that “the Qumran spelling, while idiosyncratic in certain respects, nevertheless [. . .] reflects accurately the general pattern of Hebrew or Aramaic speaking communities all across the Near East.”79 (3) Cross80  See, e.g., E. Y. Kutscher, The Language and Background of the Isaiah Scroll (1 Q Isaa) (STDJ 6; Leiden: Brill, 1974); E. Q imron , The Hebrew of the Dead Sea Scrolls (HSM 29; Atlanta: Scholars Press, 1986), 17–24; Tov , Scribal Practices (see n. 61), 263–270; idem , “The Qumran Scribal Practice: The Evidence from Orthography and Morphology,” in Verbum et Calamus: Semitic and Related Studies in Honour of the Sixtieth Birthday of Professor Tapani Harviainen (ed. H. Juusola et al.; Helsinki: Finnish Oriental Society, 2004), 353–368, here 356–363; idem, “Dead Sea Scrolls: Orthography and Scribal Practices,” in Encyclopedia of Hebrew Language and Linguistics (ed. G. Khan; Leiden: Brill, 2013), 1:669–673, here 670–671. My summary is based on the characterizations by Tov. 77  For a more detailed survey of the history of research, see Lange, “Question” (see n. 75), 424–428. 78  See especially Tov, Scribal Practices (see n. 61), 261–273; idem, “Qumran Scribal Practice” (see n. 76), 356–363; idem, “Dead Sea Scrolls” (see n. 76), 1:669–673; see also idem, “The Orthography and Language of the Hebrew Scrolls Found at Qumran and the Origin of These Scrolls,” Textus 13 (1986), 31–57; idem, “Orthography of the Hebrew Bible,” in Studien zur Hebräischen Bibel und ihrer Nachgeschichte (ed. J. F. Diehl and M. Witte; KUSATU 12–13; Kamen: Spenner, 2011), 183–209. 79  D. N. F reedman , “The Evolution of Hebrew Orthography,” in Studies in Hebrew and Aramaic Orthography (ed. D. N. Freedman et al.; Biblical and Judaic Studies 2; Winona Lake, IN: Eisenbrauns, 1992), 3–15, 14. 80  F. M. Cross, “Some Notes on a Generation of Qumran Studies,” in The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls; Madrid, 18–21 March, 1991 (ed. J. Trebolle Barrera and L. Vegas Montaner; STDJ 11; Leiden: Brill, 1992), 1–14. 76

152

Armin Lange

finds various types of orthography attested in the Dead Sea Scrolls. Next to the archaic orthography employed by scrolls like 4QSamb, Cross finds a slightly fuller orthography also employed in inscriptions and coin legends of the second century BCE, which he describes as “in the Rabbinic mode.”81 Cross regards the baroque orthography as “an attempt to preserve archaic elevated or poetic speech, lost in vernacular or prosaic Hebrew,”82 while for him “the proto-Rabbinic style is the result of revision, the leveling through of short forms where they were original and where they were not.”83 There can be no doubt that baroque orthography is widespread in copies of sectarian texts from Qumran and that Essene scribes had therefore a preference for a particular orthography and morphology characterized by the features described above. Nevertheless, as Tov acknowledges,84 the baroque orthography was used in non-Essene texts as well. Examples include the Masada inscription 449, line 1 (‫הזות‬, “this”) as well as Mur 42 2 (‫רוש‬, “head”) and the explanation of the name ‫אחשרוש‬, “Ahasuerus,” from the noun ‫ראש‬, “head,” in b. Meg. 11a. A small number of manuscripts employing baroque orthography predate the Qumran settlement (4QQoha, 175–150 BCE; 4QDibHama, ca. 150 BCE; 4QTb, 150–125 BCE), among them 1QIsaa (125–100 BCE). Some features of baroque orthography can already be found in inscriptions from Khirbet Beit Lei, Horvat Uza,85 Arad, and Lachish.86 In addition to baroque orthography, several Essene texts were copied in the conservative orthographic system known from the MT: 4QpIsab (50–25 BCE), 4QpNah (50–25 BCE), 4QCommGen A (30–31 BCE;  Cross, “Some Notes” (see n. 80), 3.  Cross, “Some Notes” (see n. 80), 4. 83  Cross, “Some Notes” (see n. 80), 6. 84  Tov, Scribal Practices (see n. 61), 262, 267, 273. 85  C ross , “Some Notes” (see n. 80), 4. The text of the inscriptions can be found in G. I. Davies, Ancient Hebrew Inscriptions (2 vols.; Cambridge: Cambridge University Press, 1991–2004), 1:89 (15.006; Beit Lei), 2:14–15 (37.006; see also 37.007; Horvat Uzza); J. Renz, Die althebräischen Inschriften: Text und Kommentar, Vol. 1 of Handbuch der althebräischen Epigraphik (Darmstadt: Wissenschaftliche Buchgesellschaft, 1995), 246–247 (Beit Lei); S. Aḥituv, Echoes from the Past: Hebrew and Cognate Inscriptions from the Biblical Period (Jerusalem: Carta, 2008), 173–177 (Horvat Uzza). The transcription of the Beit Lei burial inscription is debated (see Renz, Die althebräischen Inschriften, 247), and only the edition of Cross reflects baroque spellings. 86  Z. Zevit, Matres Lectionis in Ancient Hebrew Epigraphs (Cambridge, Mass.: American Schools of Oriental Research, 1980), 31–32; J. Barr, The Variable Spellings of the Hebrew Bible: The Schweich Lectures of the British Academy 1986 (Oxford: Oxford University Press, 1989), 122. The text of the inscriptions can be found in Davies, Ancient Hebrew Inscriptions (see n. 85), 1–2, 25 (1.002.6; 1.003.8; 2.040.9; ‫ ידעתה‬in Lachish ostraca 2 and 3 as well as in Arad ostracon 40); cf. Aḥituv, Echoes (see n. 85), 60, 63, 142; Renz, Die althebräischen Inschriften (see n. 85), 412, 417, 147; and Davies, Ancient Hebrew Inscriptions (see n. 85), 1:14 (2.007.6; ‫ ו]כתבתה‬in Arad Ostracon 7; see also Aḥituv, Echoes [see n. 85], 108; Renz, Die althebräischen Inschriften [see n. 85], 368). 81 82

153

From Many to One

sporadic spellings of the Qumran orthography), 4QSd (30–31 BCE), 4QSj (50–25 BCE), 4QMMTb (30–31 BCE), 4QBarkhi Nafshia (1–30 CE; sporadic spellings of the Qumran orthography).87 That the baroque orthography was used before and during the existence of the Qumran settlement in non-Essene texts and that MTʼs conservative orthographic system was employed in copies of sectarian texts demonstrates that neither system represents an idiolect of a specific Jewish group. For this reason, I use neutral terms in the designation of the two orthographic systems. With Cross, I describe the so-called Qumran orthography as “baroque.”88 Against Cross, I characterize the orthographic system known from the consonantal text of the MT not as “in a rabbinic mode” but instead as conservative, because the word “rabbinic” evokes group-specific affiliation. The statistic distribution of the two orthographic systems allows for a first appreciation of their importance for the textual history of the Torah. From a group of 400–500 Qumran manuscripts, Tov attributes 167 manuscripts, or ca. 37 percent, to the baroque orthography.89 Given this prominence, it is remarkable that only twenty-four biblical manuscripts – and among them only nine Torah manuscripts – are committed to the baroque orthography.90 Table 7: The Torah Manuscripts among the Dead Sea Scrolls and the Nash Papyrus according to their Orthography conservative

mixed

Ca. 250 BCE

4QExod–Lev

200–150 BCE

5QDeut

175–150 BCE

4QDeuta

Ca. 150 BCE

4QpaleoGenm

150–75 BCE

2QpaleoLev

baroque

f

4QLev–Numa

 See Tov, Scribal Practices (see n. 61), 262; cf. M. G. Abegg, “The Linguistic Analysis of the Dead Sea Scrolls: More than (Initially) Meets the Eyes,” in Rediscovering the Dead Sea Scrolls: An Assessment of Old and New Approaches and Methods (ed. M. L. Grossman; Grand Rapids, MI: Eerdmans, 2010), 48–68, here 56. 88  Cross, “Some Notes” (see n. 80), 3–5. 89  Tov, Scribal Practices (see n. 61), 263. 90  In the following table, only manuscripts that can be classified with certainty as biblical are included. Manuscripts that may attest to quotations are not included. This also applies to Psalms manuscripts of which only one Psalm is preserved (1QPsc, 4QPsp, 4QPsw, 4QPsx, 11QPse?, MS Schøyen 5233/2), because extrabiblical texts sometimes quote complete Psalms, such as Ps 122 in 4Q522 22–25. Furthermore, the table only includes biblical manuscripts whose orthographic system can still be described with some degree of certainty despite textual damages. Also excluded from the table is the semi-Masoretic manuscript 1QDanb, because only passages of the Aramaic parts of Daniel are extant from it. 87

154

Armin Lange

conservative

mixed

150–50 BCE

Nash Papyrus

150–100 BCE

4QExode, 1QpaleoLev–Numa, 4QDeutb,c

150–125 BCE

4QGen-Exoda

125–75 BCE

4QExodd, 4QDeutd

100–25 BCE

4QpaleoExodm, 4QpaleoDeutr, 4QRPe

100–50 BCE

6QpaleoGen, 4QDeuti

75–50 BCE

4QRPd, 4QDeutf,o,p

Ca. 50 BCE

4QGen , 4QLev , 4QDeutl

50–1 BCE

4QLevg, 4QDeuth,q

50–25 BCE

4QGene,j, 8QGen, 4QExodc

30 BCE–20 CE

4QLevd, 1QDeutb

30–1 BCE

1QGen, 4QLeve, MasLeva,b, 4QDeutn, MasDeut

d,g,f

baroque

4QpaleoGen–Exodl

4QRPb,c

b

4QDeutm

4QExodb

4QNumb 1QDeuta, 4QDeutk1,k2

1–68 CE

2QExodb

1–50 CE

1QExod, 11QpaleoLeva, 4QDeutg

1–30 CE

4QGenk

20–68 CE

4QGenc, MurDeut

Ca. 50 CE

2QNuma, 4QDeutj

50–100 CE

SdeirGen, 4QGenb, Mur/ḤevLev, XḤev/SeNumb

50–68 CE

2QExoda, 4QLevc

50–75 CE

XḤevSeDeut

100–135 CE

MurGen-Exod, Numa, ArugLev

11QLevb

The statistical data leave no doubt about the dominance of the conservative orthography throughout the Second Temple period in the transmission of biblical manuscripts. The orthography of 160 biblical manuscripts can still be identified. 130, or ca. 81 percent, of these manuscripts are committed to the conservative orthographic approach; 24 manuscripts, or 15 percent, employ the baroque system; and 6 manuscripts, or 4 percent, are of mixed orthography. The data for

From Many to One

155

the Torah alone compares well. Fifty-nine Torah manuscripts, or 83 percent, are committed to conservative orthography; nine manuscripts, or 13 percent, employ baroque orthography, and only four Torah manuscripts, or 4 percent, attest to a mixed orthography. Given that not all manuscripts found at Qumran were produced there, the statistical evidence from Qumran and the other sites around the Dead Sea allows for conclusions about the spread of the conservative and baroque orthographies in the biblical manuscripts in general and the Torah manuscripts in particular of late Second Temple Judaism. The baroque orthography was hence most likely not characteristic of “the general pattern of Hebrew or Aramaic speaking communities all across the Near East.”91 It might have been the preferred orthography of the Essene community but was tied neither to this nor to any other Jewish group. With regard to group affiliation, the same is true for the conservative orthography. The conservative orthography was furthermore without doubt the dominant orthographic choice for the transmission of biblical texts. It seems probable that scribes had the license to choose the orthographic system in which they copied a biblical text, most likely deciding on the basis of the reading abilities of their customers, as the many plene spellings of the baroque orthography facilitated easier comprehension of the biblical texts. The scribal license for orthographic revision is illustrated by 4QDanb. The scribe of 4QDanb adjusted the conservative orthography of his Vorlage (4QDana) to the baroque orthographic system.92 Beyond these general conclusions, the statistical orthographic evidence allows for observations about the chronological spread of the two orthographic systems. Graph 1: Distribution of Manuscripts with Baroque Orthography

 Freedman, “Evolution” (see n. 79), 14.  Thus E. U lrich , “Orthography and Text in 4QDana and 4QDanb and in the Received Masoretic Text,” in Of Scribes and Scrolls: Studies on the Hebrew Bible, Intertestamental Judaism, and Christian Origins Presented to John Strugnell on the Occasion of His Sixtieth Birthday (ed. H. W. Attridge et al.; Lanham: University Press of America, 1990), 29–42. 91

92

156

Armin Lange

What springs to the eye is that until 175 BCE, no biblical manuscripts were copied in the baroque orthography and that until the first century BCE only four extant copies of biblical books employed the baroque style of writing. The greatest number of baroque manuscripts was produced in the last quarter of the first century BCE, at the time of the highest manuscript production in Judea (see below, p. 161). After this point, the number of baroque biblical manuscripts declines in approximately the same degree as manuscript production decreased in general in Judea. The evidence for the Torah corresponds with this general pattern. Only one Torah manuscript from the middle of the second century BCE employs baroque orthography (4QLev-Numa). Seven baroque Torah manuscripts are from the first century BCE, three of them being penned in the early Herodian bookhand. Only one baroque Torah manuscript was produced in the first century CE. On the whole, scribes wrote biblical manuscripts in baroque orthography only in the last two centuries before the common era, and only even later in this period did baroque orthography become more popular. The reason for this development might be a decreasing command of Hebrew in the late Second Temple period. Particularly the lavish use of matres lectionis in the baroque orthography allows readers to pronounce and thus read unvocalized Hebrew texts with greater ease. Therefore, one of the functions of baroque orthography might have been to increase the factor of intelligibility of unvocalized Hebrew and to thus allow for a better understanding of the biblical text at a time in which the command of the Hebrew language decreased. It is also salient that manuscripts with baroque orthography are especially common among the nonaligned biblical Dead Sea Scrolls. The orthography of thirteen of the forty-seven nonaligned biblical manuscripts from Qumran is baroque, equaling 27.5 percent of all non-aligned manuscripts: 4QRPb,c, 4QDeutk1,k2, 4QSamc, 1QIsaa, 4QIsac, 4QXIIc, 4QPsa, 11QPsa, 4QQoha, 4QLam, 4QDanb. In stark contrast, only one semi-masoretic manuscript (2QJer) and one pre-Samaritan manuscript (4QNumb) display baroque orthography. The remaining baroque manuscripts are too damaged for text-typological classification. Because the lavish use of plene spellings as well as of elongated suffixes and affixes in baroque orthography facilitates the reading of unvocalized Hebrew texts, this prominence could point to a slight preference for nonaligned biblical texts among the readers of biblical scrolls in the Qumran settlement. My presentation of the evidence should not evoke the impression that the baroque and conservative orthographies existed without influencing each other. A survey of the orthography of the consonantal text of MTL demonstrates that this medieval codex includes a significant number of baroque spellings.93 Many of these baroque spellings have to do with the verbal affix ‫תה‬- spelled with a 93  I have conducted this survey with the support of Andrea Colella. It was published in L ange , “Question” (see n. 75). The survey was conducted with the help of the Dead Sea

157

From Many to One

hêʾ for the second person masculine and the suffixes ‫כה‬-, ‫כנה‬-, ‫המה‬-, ‫הנה‬-. One of the most dominant baroque spellings is ‫נָ ַת ָּתה‬, “you gave,” with sixty-four occurrences in MTL. Other examples include the double use of matres lectionis that is typical for the orthography of many Dead Sea Scrolls. Examples are ‫ ֵאֹפוא‬instead of ‫( ֵאֹפו‬Gen 27:33, 37; 43:11; Exod 33:16) and ‫ ְמֹלוא‬instead of ‫ְמלֹא‬ (Lev 5:12). The evidence for the Torah is summarized in the following table and graph. Table 8: Baroque Spellings in MTL-Torah ‫ֲהֹלוא‬

‫ֹלוא‬

Genesis

8×

1×

Exodus

1×

Leviticus

‫ְבֹלוא‬

‫ְכֹלוא‬

‫ְלֹלוא‬

‫ נָ ַת ָּתה‬Other baroque spellings 2×

16×

15×

13×

1×

3×

Numbers

2×

6×

8×

Deuteronomy

1×

7×

10×

Graph 2: Distribution of Baroque Spellings in MTL by Biblical Book

The use of baroque spellings in the consonantal text of MT has various causes. Scribal orthographic indifference, scribal habit, and idiomatic use can all trigger the occasional baroque spelling inside the consonantal text of the MT. In the consonantal text of MT, some biblical books have baroque spellings such as ‫הלוא‬, “Is not?,” and ‫נתתה‬, “you gave,” as distinct and coherent orthographic features, and some biblical books employ baroque spellings for semantic distinction. Scrolls Electronic Library and Accordance. In addition, we have searched MTL manually for baroque spellings.

158

Armin Lange

The distribution of baroque spellings within the consonantal text of MT in individual biblical books is also important. The inconsistency and randomness with which baroque spellings occur is worth noting. Examples include 1 Sam 15:23 and 26 as well as Exod 26:33 and 34. In these cases, the same word is spelled in the same phrase, within a space of three verses, once with the conservative affix ‫ת‬- and once with the baroque affix ‫תה‬-. Ezekiel 40:4 uses both the conservative and the baroque versions of the second-person masculine singular suffix in the same verse. At least for some biblical books, the randomness with which baroque spellings occur in the consonantal text of MT is probably best explained by a process or orthographic revision in which a scribe overlooked individual baroque spellings. If the MT text of some biblical books goes back to a conservative orthographic revision of an earlier baroque manuscript, this would further strengthen the argument that baroque orthography was widespread. Baroque and conservative orthographies likely operated in parallel as the two dominant orthographic approaches of the late Second Temple period. The prominence of the baroque approach probably had to do with greater ease of reading, while the conservative approach seems to have enjoyed the popularity of a respected tradition. With respect to the employment of baroque features, the above evidence shows that the conservative orthography of the Masoretic Text of the Hebrew Bible is not a coherent system but rather varies from book to book. The texts of many books of the Hebrew Bible are conservative, sometimes even very conservative. The book of Job, for example, is more defective than most other biblical books.94 The Torah and 1–2 Kings on the whole reflect a more defective orthography than other books, but they also contain the highest degree of internal inconsistency.95 Other biblical books might have originally been written in baroque orthography or at least transmitted for some time in baroque orthography. Afterward, they underwent orthographic revision toward the conservative orthography. In this process, isolated baroque spellings were overlooked. After baroque features entered the conservative orthography of the (proto)-Masoretic text or remained within it by virtue of being overlooked in revisions, the orthographic and textual standardization of the proto-Masoretic text created a canonical text that is orthographically conservative but that contains irregular occurrences of other surviving orthographies. Once such a normative text of the Hebrew canon came into existence, it exerted an orthographic influence on later  For the discussion of the especially conservative orthography of the book of Job, see, e.g., D. N. Freedman, “Orthographic Peculiarities in the Book of Job,” ErIsr 9 (1969), 35–44; J. Barr, “Hebrew Orthography and the Book of Job,” JSS 30 (1985), 1–33; and C. L. Seow, “Orthography, Textual Criticism, and the Poetry of Job,” JBL 130 (2011), 63–85. 95  Thus F. I. Andersen and A. D. Forbes, Spelling in the Hebrew Bible (BibOr 41; Rome: Biblical Institute Press, 1986), 312–318. A. Murtonen, “The Fixation in Writing of Various Parts of the Pentateuch,” VT 3 (1953), 46–53; and Tov, “2.1 Textual History of the Pentateuch” (see n. 35). 94

From Many to One

159

generations. This normative orthographic influence led to the dominance of the conservative orthography in rabbinic and post-rabbinic Hebrew literature.96 Finally, it must be mentioned that baroque orthography is not the only orthographic system that left traces in the consonantal MT text of the Torah. Another example is the occasional use of the consonant he as a mater lectionis ֲ “Is for an o vowel at the end of a word. Examples include ‫ ֲהֹלה‬instead of ‫הלֹא‬, not?,” in Deut 3:11. Given the orthographic inconsistencies of the consonantal text of MTL, it is all the more surprising that the twenty-two proto-Masoretic Dead Sea Scrolls are all in almost complete orthographic agreement with the consonantal text of MTL, while the other 108 biblical manuscripts with conservative orthography each include several orthographic disagreements toward the consonantal text of MTL. The most likely explanation for this phenomenon is that the textual standardization that created the proto-Masoretic text also included orthographic adjustments (see below, 3.2). The linkage between the proto-Masoretic text and the conservative orthography also explains why no manuscripts with baroque orthography are preserved from after the destruction of the Herodian temple: after this catastrophic event, only proto-Masoretic manuscripts remained. 3.2 Textual Standardization97 Given the textual and orthographic plurality of the Second Temple period, it needs to be asked when and how the proto-Masoretic standard text of the Hebrew Bible developed. Above, I emphasized the fact that all biblical manuscripts from Masada, Naḥal Ḥever, and Wadi Murabbaat are proto-Masoretic in character, while the Qumran manuscripts attest to a rich textual plurality. Graph 3 and graph 4 summarize the evidence for all biblical manuscripts from the second Temple period in general and for the Torah manuscripts in particular.98  Thus also the work of G. Geiger, based on a study of the plene spellings of the active qal participle in the Dead Sea Scrolls (“Plene Writing of the QŌṬĒL Pattern in the Dead Sea Scrolls,” in Hebrew in the Second Temple Period: The Hebrew of the Dead Sea Scrolls and of Other Contemporary Sources: Proceedings of the Twelfth International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature and the Fifth International Symposium on the Hebrew of the Dead Sea Scrolls and Ben Sira, Jointly Sponsored by the Eliezer Ben-Yehuda Center for the Study of the History of the Hebrew Language, 29–31 December, 2008 [ed. S. E. Fassberg et al.; STDJ 108; Leiden: Brill, 2013], 73–81). 97  This section of the present article is a summary of A. L ange , “ ‘They Confirmed the Reading’ (y. Taʿan. 4:68a): The Textual Standardization of Jewish Scriptures in the Second Temple Period,” in From Qumran to Aleppo: A Discussion with Emanuel Tov about the Textual History of Jewish Scriptures in Honor of his 65th Birthday (ed. A. Lange, M. Weigold, and J. Zsengellér; FRLANT 230; Göttingen: Vandenhoeck & Ruprecht, 2009), 29–80. 98  As compared to earlier versions of this graph (Lange , Handschriften [see n. 29], 25; idem , “They Confirmed” [see n. 97], 56), changes are due to the discovery of new manuscripts 96

160

Armin Lange

Graph 3: Text-Typological Chronology of the Hebrew Manuscripts of Biblical Books from the Dead Sea99

Graph 4: Text-Typological Chronology of the Hebrew Torah Manuscripts from the Dead Sea100

Graph 3 and table 5 on pp. 144–145 illustrate a characteristic spread of textual affiliations over the various Hebrew scripts used in Greco-Roman times, both for all biblical Dead Sea Scrolls and for the Torah manuscripts in particular. The correlation between paleography and text typology shows that the other texts in graphs 3 and 4 (i.e., nonaligned manuscripts, pre-Samaritan manuscripts, and manuscripts close to the Hebrew Vorlage of LXX) are the largest group of textual witnesses until the middle of the first century CE. The “other” manuscripts from Qumran therefore represent the textual plurality of Greco-Roman times. The proto-Masoretic manuscripts from the various sites of the Dead Sea are distributed differently over the paleographic epochs from the “other” manuscripts. Although 4QJera and 4QXIIe go back to the years 225–175 and 75–50 BCE respectively, proto-Masoretic manuscripts can be found in larger quantities only since the second half of the first century BCE. In the case of the Torah, no proto-Masoretic manuscripts at all are known from the time before the second half of the first century BCE. Semi-Masoretic manuscripts are attested as early as the and the recognition of new research. 99  Graph 3 does not include manuscripts for which it cannot be decided whether they attest to a semi-Masoretic or a pre-Samaritan text. 100  Graph 4 does not include manuscripts for which it cannot be decided whether they attest to a semi-Masoretic or a pre-Samaritan text.

From Many to One

161

first half of the second century BCE (4QJerc) but occur more regularly only from the end of the second century BCE. For the Torah, only two semi-Masoretic manuscripts from the years 125–25 BCE are preserved. The largest quantities of non-aligned, semi-Masoretic, and proto-Masoretic manuscripts come from the second half of the first century BCE. This spread agrees well with the overall paleographic distribution of the Dead Sea Scrolls. A plurality of manuscripts found in the Qumran library were copied in scripts from the second half of the first century BCE.101 The Masada manuscripts show that this paleographic peak of the Qumran library is not idiosyncratic but is part of an overall phenomenon. On Masada, fifteen manuscripts were found. Nine of these fifteen manuscripts were produced in the second half of the first century BCE.102 The paleographic evidence from Qumran and Masada thus points to an increased manuscript production and an increased amount of scribal activity in the time after the Roman conquest of Judea (63 BCE) and especially during the reign of Herod the Great. It seems that this time marks a zenith of scribal culture. Even more interesting is that after the turn of the eras, the number of nonaligned and semi-Masoretic manuscripts decreases continuously, while the number of proto-Masoretic manuscripts peaks again in the second half of the first century CE. After the year 68 CE, only proto-Masoretic manuscripts can be found. Therefore, the chronological distribution of the biblical manuscripts described above shows that the proto-Masoretic text was created in the second half of the first century BCE. These observations agree well with the appearance of the first recensions of the Old Greek text of the Jewish scriptures toward the proto-Masoretic text. The earliest manuscript evidence for such recensions can be found with P.Fouad Inv. 266b (= LXX948; paleographic date ca. 50 BCE).103 In Judea, the manuscripts 101  See B. Webster, “Chronological Index of the Texts from the Judaean Desert,” in The Texts from the Judaean Desert: Indices and an Introduction to the Discoveries in the Judaean Desert Series (ed. E. Tov et al.; DJD XXXIX; Oxford: Clarendon Press, 2002), 351–446, here 373–374. 102  For the paleographic dates of the nonbiblical manuscripts from Masada, see S. Talmon and Y. Yadin, Masada VI: The Yigael Yadin Excavations 1963–1965, Final Reports, Hebrew Fragments from Masada (Jerusalem: Israel Exploration Society, 1999), 101, 106, 117–118, 120, 133–134, 136, 138–139, 157–158. 103  For the paleographic date of P.Fouad Inv. 266b, see F. Dunand, Papyrus grecs bibliques (Papyrus F. Inv. 266.): Volumina de la Genèse et du Deutéronome: Introduction (Recherches dʼarchéologie, de philologie et dʼhistoire 27; Cairo: Institut français dʼarchéologie orientale, 1966), 12. For the textual character of this manuscript, see Hanhart , “Die Söhne Israels” (see n. 49), 170–178, here 171; idem, review of F. Dunand, Papyrus grecs bibliques (Papyrus F. Inv. 266.), OLZ 73 (1978), 39–45, here 42–45; L. Koenen, “Introduction,” in Three Rolls of the Early Septuagint: Genesis and Deuteronomy; A Photographic Edition Prepared with the International Photographic Archive of the Association Internationale de Papyrologues (ed. Z. Aly; Papyrologische Texte und Abhandlungen 27; Bonn: Habelt, 1980), 1–23, here 1–2 and 9.

162

Armin Lange

8ḤevXII gr104 and 4QLXXNum105 document pre-Christian recensions of Old Greek translations toward a semi- or proto-Masoretic text in the latter part of the first century BCE. 8ḤevXII gr is a representative of the Kaige-recension, which began to emerge at that time.106 Similar recensional activities are furthermore reflected in some of the biblical quotations in the letters of Paul.107 The proto-Lucianic recension108 demonstrates that these early recensional efforts were not always directed toward semi- or proto-Masoretic texts but adjusted the OG to other Hebrew texts as well. The evidence of the early recensions confirms that efforts toward textual standardization started some time in the second half of the first century BCE, that is, after the conquest of Syro-Palestine by Pompey the Great in 63 BCE. 104  For 8ḤevXII gr, see D. Barthélemy, Les devanciers dʼAquila: Première publication intégrale du texte des fragments du Dodécaprophéton trouvés dans le désert de Juda, précédée dʼune étude sur les traductions et recensions grecques de la Bible réalisées au premier siècle de notre ère sous lʼinfluence du rabbinat palestinien (VTSup 10; Leiden: Brill, 1963); E. Tov, The Greek Minor Prophets Scroll from Naḥal Ḥever (8ḤevXII gr) (DJD 8; Oxford: Clarendon Press, 1990). For the paleographic date of 8ḤevXII gr in the latter part of the first century BCE, see P. J. Parsons , “The Scripts and Their Date,” in The Greek Minor Prophets Scroll from Naḥal Ḥever (8ḤevXII gr) (by E. Tov; DJD 8; Oxford: Clarendon Press, 1990), 19–26. For the textual character of 8ḤevXII gr, see, e.g., B arthélemy , Les devanciers, 179–272; Tov, Greek Minor Prophets, 145–158; K. H. Jobes and M. Silva, Invitation to the Septuagint (Grand Rapids, MI: Baker Academic, 2000), 171–173. 105  For the paleographic date of 4QLXXNum in the late first century BCE or the early first century CE, see P. J. Parsons , “The Paleography and Date of the Greek Manuscripts,” in Qumran Cave 4.IV: Palaeo-Hebrew and Greek Biblical Manuscripts (ed. P. W. Skehan et al.; DJD IX; Oxford: Clarendon Press, 1992), 7–13, here 11. For 4QLXXNum as a recension resembling Kaige, see P. W. Skehan, “4QLXXNum: A Pre-Christian Reworking of the Septuagint,” HTR 70 (1977), 39–50, here 39–40; see also J. W. Wevers, “An Early Revision of the Septuagint of Numbers,” ErIsr 16 (1982), 235*–239*; U. Quast, “Der rezensionelle Charakter einiger Wortvarianten im Buche Numeri,” in Studien zur Septuaginta: Robert Hanhart zu Ehren (ed. D. Fraenkel et al.; MSU 20; Göttingen: Vandenhoeck & Ruprecht, 1990), 230–252, here 248–250. 106  See B arthélemy , Les devanciers (see n. 104), 203–265; G. H oward , “The Quinta of the Minor Prophets,” Bib 55 (1974), 15–22; A. van der K ooij , Die alten Textzeugen des Jesaja­buches: Ein Beitrag zur Textgeschichte des Alten Testaments (OBO 35; Freiburg, Schweiz: Uni­ver­sitätsverlag, 1981), 127–150; J obes and S ilva , Invitation (see n. 104), 171–173, 284–287; N. Fernández Marcos, The Septuagint in Context: Introduction to the Greek Version of the Bible (Leiden: Brill, 2001), 142–154. 107  D. A. Koch, Die Schrift als Zeuge des Evangeliums: Untersuchungen zur Verwendung und zum Verständnis der Schrift bei Paulus (BHT 69; Tübingen: Mohr Siebeck, 1986), 56–82. 108  For the Lucianic and proto-Lucianic recensions, see P. G entry , “1.3.1.2 Pre-Hexaplaric Translations, Hexapla, Post-Hexaplaric Translations,” in Textual History of the Bible (ed. A. Lange; Brill Online, 2016), http://referenceworks.brillonline.com/entries/ textual-history-of-the-bible/1312-pre-hexaplaric-translations-hexapla-post-hexaplaric-translations-COM_00000013 (accessed 3/3/16); and, e.g., B. M. Metzger, Chapters in the History of New Testament Textual Criticism (NTTS 4; Leiden: Brill, 1963), 1–41; Fernández Marcos, Septuagint in Context (see n. 106), 223–238.

From Many to One

163

Given that recensions of the Old Greek text toward a proto-Masoretic text are among the first witnesses to the rising dominance of the proto-Masoretic text, it can be ascertained that Greco-Roman culture contributed to the development of the proto-Masoretic standard text. This is all the more the case given that a considerable command of the Greek language was necessary to revise the OG text toward the consonantal text of MT. To achieve such an advanced command of Greek, the revisers of the Old Greek text toward MT must have been exposed to advanced training in Greek language and culture. Awareness of the Greek character of the idea of standardizing the Jewish scriptures is still perceptible in Josephusʼs apology Against Apion. When he emphasizes, “With us are not myriads of books which are at variance and contradict each other” (Ag. Ap. 1.38) and when he stresses, “For during so many ages as have already passed, nobody has dared either to add to them, to take from them, or to make changes” (Ag. Ap. 1.42),109 Josephus employs the typical argumentative strategy of an apology. He claims that Judaism better fulfills a cultural value of the Greco-Roman culture against whose anti-Semitism he argues in his apology Against Apion. In my article “They Confirmed the Reading,”110 I have argued that the increased acculturation to which Roman rule exposed Judea was the catalyst that led to the development of the proto-Masoretic standard text of the Hebrew Bible. The first steps toward this process were taken in the first half of the first century BCE. The development of the proto-Masoretic standard text thus coincided with the first efforts to close the Jewish canon.111 When such a standard text was needed after the destruction of the Herodian temple in the year 70 CE, it was readily available. It remains to be asked who was responsible for the textual standardization of the Hebrew Bible since the second half of the first century BCE. As argued above, the recension of the OG toward MT – being a part of the early standardization effort – required a command of the Greek language, which could only be found among Jews exposed to Greek paideia. The example of Josephus Flavius demonstrates that the priestly elites at the Jerusalem temple benefitted from an education in both Jewish and Greek traditions. The Jerusalem Temple thus emerges as a likely milieu for the development of the proto-Masoretic standard text of the Jewish scriptures. The writings of Josephus and rabbinic literature corroborate the involvement of the Jerusalem temple and its priestly elites in the development of the pro My translation.  See n. 97. 111  For my dating of the closing of the Jewish canon, see A. Lange, “The Qumran Library in Context: The Canonical History and Textual Standardization of the Hebrew Bible in Light of the Qumran Library,” in The Dead Sea Scrolls at Qumran and the Concept of a Library (ed. S. W. Crawford and C. Wassen; STDJ 116; Leiden: Brill, 2016), 261–279, and idem, “1.1.2.1 The History of the Jewish Canon,” in The Hebrew Bible, Vol. 1 of Textual History of the Bible (ed. A. Lange and E. Tov; Leiden: Brill, 2016), 35–48. 109 110

164

Armin Lange

to-Masoretic standard text. In Ant. 3.38; 5.61; and 10.57 (cf. 12.56), Josephus mentions that books or the holy books of Moses were laid up in the Jerusalem temple. In Ant. 4.303–04, Josephus even claims that Moses would have given autographs of his writings to the priests for safekeeping. Especially when read in light of Ag. Ap. 1.38–42, these remarks could refer to master copies of Jewish scriptures in the Jerusalem Temple in the first century CE.112 Further confirmation that various Torah scrolls and other sacred books were kept in the Jerusalem temple can be found in Josephusʼs remarks that he took sacred books from the spoils that the Romans looted in Jerusalem (Life 418). That Vespasian and Titus presented the law of the Jews during their triumphal march in Rome together with other items from the Jerusalem temple (J.W. 7.150) and put them on display in the Roman temple of peace afterward (J.W. 7.162) could also point to master copies at the Jerusalem temple. Further evidence of a maintained proto-Masoretic text at the Jerusalem temple may be preserved in rabbinic literature.113 It stands to reason that the cultural memory of the rabbis as expressed in the Mishnah, the Talmudim, and other rabbinic literature reflects the last phase of the Second Temple period – the time of the Herodian temple – rather than earlier circumstances. Like the works of Josephus, rabbinic literature thus helps to illuminate how the biblical text was treated at the Jerusalem temple in the late first century BCE and the first century CE. The scholarly interpretation of rabbinic references to biblical scrolls at the Jerusalem temple is influenced by the work Blau. Blau understood the Scroll of the Temple Courtyard, discussed below, as a liturgical scroll used by the high priest that also served as a model text against which all other scrolls were compared.114 Blauʼs idea of a master copy influenced the work of Lieberman,115 Kutscher,116 Gordis,117 and Tov118 on this issue. 112  For master copies of Jewish scriptures in the Jerusalem temple, see also Tov, “Text of the Hebrew/Aramaic and Greek Bible” (see n. 31), 246–248. 113  I am much obliged to my colleague and friend Günter Stemberger, who helped with my analysis of the rabbinic evidence and guided my thoughts with these important but difficult sources. All remaining errors are, of course, my own. 114  L. B lau , Studien zum althebräischen Buchwesen und zur biblischen Literatur- und Textgeschichte (Strasbourg i. E.: Adolf Alkalay, 1902), 107–111. 115  S. L ieberman , Hellenism in Jewish Palestine: Studies in the Literary Transmission, Beliefs and Manners of Palestine in the I Century B.C.E.–IV Century C.E. (Text and Studies of the Jewish Theological Seminary of America 18; New York: The Jewish Theological Seminary, 1962), 22. 116  Kutscher, Language (see n. 76), 82–83. 117  R. G ordis, The Biblical Text in the Making: A Study of the Kethib–Qere (New York: Ktav, 1971), xi–lvi, here xxvi–xxviii. 118  E. Tov, Hebrew Bible, Greek Bible, and Qumran: Collected Essays (TSAJ 121; Tübingen: Mohr Siebeck, 2008), 178–180; idem, Textual Criticism (see n. 31), 30–31; idem, “The Dead Sea Scrolls and the Textual History of the Masoretic Bible,” in The Hebrew Bible in

From Many to One

165

Several times, rabbinic texts mention special scrolls kept in the Jerusalem temple. While these scrolls were certainly revered, and their text was most likely written very carefully, nothing is known about their textual character and textual affiliation. The rabbinic references to these scrolls provide no information on whether they were used as master copies or served other purposes. A liturgical use of a Torah scroll by the high priest, for example, is mentioned in m. Yoma 7:1 (cf. b. Yoma 68b–70a). B. Soṭah 40b discusses where the High Priest should read the Torah scroll mentioned in m. Yoma 7:1: in the womenʼs courtyard or elsewhere. M. Soṭah 7:8 describes how the king reads from a Torah scroll in a temple courtyard, emphasizing king Agrippa as a positive example. In contrast, according to m. Yoma 1:6, sages read before the high priest from the books of Job, Ezra, Chronicles, or Daniel during the seven days before Yom Kippur. Among the scrolls of Jerusalem Temple that are mentioned in rabbinic literature, the Scroll of the Temple Courtyard119 (‫)ספר עזרה‬120 is probably the most important one. One of the earliest statements about the Scroll of the Temple Courtyard is m. Kelim 15:6: “All scrolls render the hands unclean, excepting the Scroll of the Temple Courtyard.”121 The phrase “render the hands unclean” is much debated in the context of the canonical history of the Hebrew Bible.122 Applied to a specific copy of a biblical book, it should be understood as a reference to a liturgical use of such a Light of the Dead Sea Scrolls (ed. N. David et al.; FRLANT 239; Göttingen: Vandenhoeck & Ruprecht, 2012), 41–53, here 47–48; idem, “Scribal and Textual Transmission of the Torah” (see n. 31), 59–60. 119  M. Kelim 15:6; m. Moʿed Qaṭ 3:4; Sipre Devarim 160; y. Šeqal. 4:2; y. Moʿed Qaṭ 3:4; y. Sanh. 2:6, 20c; b. B. Bat. 14b. 120  For m. Moʿed Qaṭ 3:4, C. Albeck notes in the apparatus of his edition (The Six Orders of the Mishnah: Interpreted and Vocalized; Tractate Moʿed [Jerusalem: Bialik Institute, 1958], ad loc. [Hebrew]) the reading “scroll of Ezra” (‫)ספר עזרא‬. The reading ‫ ספר עזרא‬is also attested in t. Kelim B. Meṣiʿa 5:8. 121  Translations of the Mishnah are guided by H. Danby, The Mishnah: Translated from the Hebrew with Introduction and Brief Explanatory Notes (Peabody: Hendrickson, 2011 [reprint of the 1933 edition]) but are adapted to my line of argument on several occasions. 122  See, e.g., S. Z. L eiman , The Canonization of Hebrew Scripture: The Talmudic and Midrashic Evidence (Hamden: Archon Books, 1976), esp. 102–120; J. B arton , Oracles of God: Perceptions of Ancient Prophecy in Israel after the Exile (London: Darton, Longman and Todd, 1986), 68–71; J. Barton, The Spirit and the Letter: Studies in the Biblical Canon (London: SPCK, 1997), 108–121; idem, “The Canonicity of the Song of Songs,” in Perspektiven der Hoheliedauslegung/Perspectives on the Song of Songs (ed. A. Hagedorn; Berlin: de Gruyter, 2005), 1–7; M. G oodman , “Sacred Scripture and ‘Defiling the Hands’,” JTS 41 (1990), 99–107; S. Friedman, “The Holy Scriptures Defile the Hands: The Transformation of a Biblical Concept in Rabbinic Theology,” in Minḥah le-Naḥum: Biblical and Other Studies Presented to Nahum M. Sarna in Honour of his 70th Birthday (JSOTSup 154; Sheffield: Sheffield Academic Press, 1993), 117–132. T. H. L im , “The Defilement of the Hands as a Principle Determining the Holiness of Scriptures,” JTS 61 (2010), 501–515.

166

Armin Lange

scroll.123 At first glance, m. Kelim 15:6 would thus describe the Scroll of the Temple Courtyard as a nonliturgical scroll. But very little is known about the use of scripture in the temple liturgy. Furthermore, no prerabbinic evidence exists for the phrase “render the hands unclean.” That the Scroll of the Temple Courtyard does not render the hands unclean therefore at best reflects the understanding of the Mishnah but says little about the use of this scroll in the Jerusalem temple. Despite the opinion of the Mishnah, the very name of this scroll points to a liturgical context. Why should a specific copy of a biblical book be named Scroll of the Temple Courtyard if it is not read publicly in this courtyard? Furthermore, t. Kelim B. Meṣiʿa 5:8 contradicts m. Kelim 15:6 in its claim that the Scroll of the Temple Courtyard would not defile the hands: t. Kelim Baba Meṣiʿa 5:8 ‫ספר עזרא שיצא לחוץ מטמא את הידים ולא ספר עזרה בלבד אמרו אלא אפלו נביאים וחומשים‬ ‫וספר אחר שנכנס לשם מטמא את הידים‬ The Scroll of Ezra that went forth outside [the court] renders the hands unclean. And not only of the Scroll of the Temple Courtyard alone did they speak, but even the Prophets and the individual books of the Torah. And (every) other scroll that entered there renders the hands (un)clean.124

In contrast to all other references to the Scroll of the Temple Courtyard, t. Kelim B. Meṣiʿa 5:8 speaks about the ‫“( ספר עזרא‬scroll of Ezra”?) and not about the ‫ספר‬ ‫עזרה‬, Scroll of the Temple Courtyard. Because both phrases are distinguished only by one character, and because both t. Kelim B. Meṣiʿa 5:8 and t. Kelim B. Meṣiʿa 5:7 discuss matters of purity and impurity connected with the Temple courtyard, it is likely that ‫ עזרא‬is just a different orthographic realization of ‫עזרה‬ in t. Kelim B. Meṣiʿa 5:8. That t. Kelim B. Meṣiʿa 5:8 uses ‫ספר עזרה‬, Scroll of the Temple Courtyard, to refer to the whole Torah becomes evident when it is seen that it distinguishes it from the ‫חומשים‬, the individual books of the Pentateuch. T. Kelim B. Meṣiʿa 5:8 thus demonstrates that in a rabbinic tradition other than m. Kelim 15:6, the Scroll of the Temple Courtyard was regarded as a liturgical Torah scroll rendering the hands unclean.

123  Cf. G. S temberger , “Jabne und der Kanon,” Jahrbuch für Biblische Theologie 3 (1988), 163–174; idem, “Entstehung und Auffassung des Kanons im rabbinischen Denken,” in Biblische Traditionen im rabbinischen Judentum, Vol. 1 of Judaica Minora (G. Stemberger; TSAJ 133; Tübingen: Mohr Siebeck, 2010), 68–87, esp. 72–75. 124  The text from t. Kelim B. Meṣiʿa 5:7 is quoted according to manuscript Vienna 46. The translation is guided by J. N eusner , The Tosefta: Translated from the Hebrew; Sixth Division; Tohorot (The Order of Purities) (New York: Ktav Publishing House, 1977), 43. Neusnerʼs translation contains a typo. The last word of the translation quoted above should be “unclean” and not “clean,” as the Tosefta reads here ‫מטמא‬. Neusner renders ‫ ספר עזרה‬as the Scroll of Ezra, which I have changed above to Scroll of the Temple Courtyard, given my own understanding of the passage. That manuscript Vienna 46 reads ‫אחד‬, “one,” instead of ‫אחר‬, “other,” goes back to scribal error.

From Many to One

167

The special awe in which the Scroll of the Temple Courtyard was held is evident in a list of texts for which scribal activity is prohibited on the days between the beginning and the end of Passover and Sukkot. These days are referred to in the Mishnah as “little festival” (‫)מועד קטן‬, or “mid-festival days” in Danbyʼs translation. M. Moʿed Qaṭ 3:4 and its parallel b. Moʿed Qaṭ 18b put the Scroll of the Temple Courtyard at the peak of a list of revered texts. m. Moʿed Qaṭan 3:4 Bonds of indebtedness may not be written out during mid-festival, but one may be written out if any man was not deemed trustworthy, or if he had naught to eat. They may not write out (Torah) scrolls or phylacteries or mezuzahs during mid-festival or ‫ מגיהין‬a single letter even in the Scroll of the Temple Courtyard (‫ואין מגיהין אות אחת‬ ‫)אפלו בספר העזרה‬.

In m. Moʿed Qaṭ 3:4, the word ‫ מגיהין‬cannot refer to the correction of a new scroll in comparison with a master copy, because the Scroll of the Temple Courtyard is an already existing and highly revered biblical manuscript. Nevertheless, this revered manuscript is the object of the verb ‫נגה‬. The hiphil of the verb ‫ נגה‬in m. Moʿed Qaṭ. 3:4 therefore most likely refers to the repair of this scroll and does not identify it as a master copy.125 The prohibition of incorrect scrolls, that is, scrolls with mistakes and/or corrections, can only be found in much later texts than m. Moʿed Qaṭ 3:4 (b. ʿErub. 13a; Sepher Torah 3:4–5; cf. y. Šabb. 16:1) and thus does not undermine my interpretation. For the Second Temple period, various kinds of repairs are evident among the Dead Sea Scrolls. They are described carefully by Tov.126 The most likely type of repair that the ‫מגיהין‬ could have applied to the Scroll of the Temple Courtyard is the re-inking of characters that had faded.127 Such faded characters are also mentioned in b. Meg. 18b with regard to an Esther scroll. In the passages discussed thus far, the scrolls mentioned in connection with the Jerusalem temple are not described explicitly as master copies. But beyond the above references to the use of scrolls in liturgical contexts, “correctors” or “proofreaders” of (biblical) scrolls (‫ )מגיהי ספרים‬are mentioned in rabbinic literature. B. Ketub. 106a states that correctors of (Torah) scrolls in Jerusalem received their wages from the temple funds, and y. Šeqal. 4:2 includes the technical detail that these correctors of the Scroll of the Temple Courtyard were paid out of temple funds. A group of specialists responsible for the repair and correction of manuscripts thus seems to have existed at the Herodian temple. 125  For this meaning of ‫נגה‬, see M. J astrow, A Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature (2nd ed.; New York: Pardes, 1903), 872 and below, 3.3. 126  Tov, Scribal Practices (see n. 61), 122–125. 127  See already Blau, Studien (see n. 114), 109. For examples of re-inked characters in the Dead Sea Scrolls, see M. Martin , The Scribal Character of the Dead Sea Scrolls (2 vols.; Bibliothèque du Muséon 44–45; Leuven: Publications Universitaires, 1958), 2:424.

168

Armin Lange

A further reference to these correctors may be included in t. Sanh. 4.7, which states, according to manuscript Erfurt, that they corrected (‫ )ומגיהין‬the scroll that the king should write for himself (see Deut 17:18). But manuscript Vienna reads here ‫מניחין‬, “they deposited,” instead of “they corrected.”128 Two passages in Talmud Yerushalmi could create the impression that the Scroll of the Temple Courtyard was a master copy used by these correctors. Gordis therefore views y. Šeqal. 4:2 as a “reference to an ongoing process of collation”129 for which the scroll of the temple courtyard would have served as a model text. y. Šeqalim 4:2 R. Aha, R. Tanhum bar Hiyya in the name of R. Simlai: “Those who examine and correct the scroll [of the Torah belonging to the Temple] court receive their salary from the heave offering of the chamber (‫)מגיהי ספר העזרה היו נוטלין שכרן מתרומת הלשכה‬.”130

Against Gordis, the text is actually similar to the statement of m. Moʿed Qaṭ 3:4. In y. Šeqal. 4:2, the scroll of the temple courtyard is in not a master copy, a model text against which all other texts are corrected. To achieve his interpretation, Gordis actually needed to insert several elements into his translation of y. Šeqal. 4:2: “The passage almost surely means: ‘The revisers [of Bible manuscripts] by means of the ʿAzarah scroll.’ ”131 In fact, the reverse is true: the Scroll of the Temple Courtyard in y. Šeqal. 4:2 is the object of the work of the ‫מגיהין‬, “correctors,” and not the manuscript they cross-reference. Just like m. Moʿed Qaṭ 3:4, y. Šeqal. 4:2 says nothing more than that the “correctors” work on the ‫ספר העזרה‬, the Scroll of the Temple Courtyard. Thus, just as in m. Moʿed Qaṭ 3:4, in y. Šeqal. 4:2 the verb ‫ נגה‬does not describe a correction of an already existing scroll but the repair of damages to this scroll. The only text that explicitly describes the use of the Scroll of the Temple Courtyard as a master copy for the correction of other manuscripts is y. Sanh. 2:6. It reports that the kingʼs Torah scroll was corrected in comparison to it. y. Sanhedrin 2:6 ‫ומגיהין אותו מספר עזרה על פי בית דין של שבעים ואחד‬ And they correct his scroll by comparing it to the scroll of the Temple courtyard, on the authority of the Sanhedrin of the seventy-one members.132

 ‫ומגיהין אותו בבית דין של כהנים ובבית דין של לוים ובבית דין של ישראל המשיאין לכהונה‬. I am obliged to Günter Stemberger, who guided my attention to the two different readings. 129  Gordis, Biblical Text (see n. 117), xxvii. 130  Translation according to J. Neusner, Sheqalim, Vol. 15 of Talmud of the Land of Israel (Chicago: University of Chicago Press, 1982–1994), 77. 131  Gordis, Biblical Text (see n. 117), xxvii. 132  Translation according to J. Neusner, Sanhedrin and Makkot, Vol. 31 of Talmud of the Land of Israel (Chicago: University of Chicago Press, 1982–1994), 91. 128

From Many to One

169

Among all references to the Scroll of the Temple Courtyard, it is only in y. Sanh. 2:6 that it becomes not the object of the work of the “correctors” but the master copy in comparison with which they work. Given this irregularity, it seems likely to me that the Talmud Yerushalmi here reflects later rabbinic regulations that prohibit the use of uncorrected scrolls (see below, 3.3). Y. Sanh. 2:6 debates at length the regulations about the king and his Torah scroll (Deut 17:18) from m. Sanh. 2:6. But m. Sanh. 2:6 does not mention any correcting of this important scroll. In light of the rabbinic regulations of its own time on how a Torah scroll has to be produced, y. Sanh. 2:6 thus introduces a correction of the kingʼs Torah scroll against another famous scroll from the Jerusalem temple with which it was familiar, the Scroll of the Temple Courtyard. Such an educated speculation about the Scroll of the Temple Courtyard should not surprise, as it became the object of educated rabbinic speculation in another Talmudic text as well. B. B. Bat. 14b remarks that the Scroll of the Temple Courtyard was rolled around one cylinder (‫ )ספר עזרה לתחלתו הוא נגלל‬so that it could fit into the ark. To summarize: Early rabbinic texts, and even most references in the Talmudim, do not mention master copies kept at the Jerusalem temple but document only the existence of biblical scrolls. It is nevertheless likely that correctors worked at the Jerusalem temple,133 given the details about their payments specified in y. Šeqal. 4:2 and b. Ketub. 106a. That master copies of biblical texts were kept at the Herodian temple can only be inferred from the existence of such correctors. Without such master copies, the work of the correctors would have been impossible.134 Whether the scrolls from the Herodian temple that are mentioned in rabbinic literature served as master copies or whether the “correctors” used different scrolls remains unknown. The role of the “correctors” was most likely at least twofold. On the one hand, they had to repair liturgical scrolls, which were damaged by their use. On the other hand, they corrected newly produced scrolls based on existing scrolls. Speculation might be allowed if the “correctors” referenced for their repair of liturgical scrolls the same master copies that they used to correct newly produced manuscripts. Talmud Yerushalmi, tractate Taʿanit 4:4 could point to the involvement of the “correctors” in the standardization of the proto-Masoretic text.135 133  Cf. L ieberman , Hellenism (see n. 115), 22; Tov , Hebrew Bible, Greek Bible (see n. 118), 179; idem, “Masoretic Bible” (see n. 118), 48. 134  See, e.g., S. Krauss, Talmudische Archäologie (3 vols.; Leipzig: Buchhandlung Gustav Fock, 1910–1912), 3:170–171. 135  See also Sipre Deut. 356; y. Šeqal. 4:3; ʾAbot R. Nat. B 46. For the historicity of this tradition, see G. Stemberger and C. Dohmen, Hermeneutik der Jüdischen Bibel und des Alten Testaments (Kohlhammer Studienbücher Theologie 1/2; Stuttgart: Kohlhammer, 1996), 76. For further study, see A. Geiger, Urschrift und Übersetzungen der Bibel in ihrer Abhängigkeit von der innern Entwickelung des Judentums (Breslau: Heinauer, 21857; repr. Frankfurt a.M.: Madda, 1928), 232–251: Blau, Studien (see n. 114), 100–106; J. Z. Lauterbach, “The Three

170

Armin Lange

Three scrolls did they find in the Temple courtyard. These were the Maon scroll [“Dwelling”], the Zaatuti scroll [“Little ones”], and the He scroll. In one of these scrolls they found it written “The eternal God is your dwelling place (maon)” (Deut. 33:27: “The eternal God is your dwelling place, and underneath are the everlasting arms. And he thrust out the enemy before you, and said, ‘Destroy’ ”). And in two of the scrolls it was written, “The eternal God is your dwelling place” (meonah). They confirmed the reading found in the two and abrogated the other. In one of them they found written “They sent the little ones of the people of Israel” (Ex. 24:5: “And he sent young men of the people of Israel, who offered burnt offerings and sacrificed peace offerings of oxen to the Lord.”). And in the two it was written, “They send young men. . . .” They confirmed the two and abrogated the other. In one of them they found written, “He [he written in the feminine spelling] nine times, and in two, they found it written that way eleven times.” They confirmed the reading found in the two and abrogated the other.136

Y. Taʿan. 4:4 shows that variant readings of three Torah scrolls were compared with each other and evaluated by way of majority decision. It is not unlikely that the three manuscripts used for this purpose were liturgical scrolls kept at the Jerusalem temple and/or master copies, which formed the basis of the work of the “correctors.”137 Next to the evaluation of variant readings based on statistical decisions, another approach – probably restricted to books that were not extant in numerous copies – was to condemn a whole version of a biblical book. Thus t. Šabb. 13:2; y. Šabb. 16.1; and b. Šabb. 115a report how Rabban Gamaliel the Elder ordered an Aramaic translation of Job to be immured below a stairway of the Temple. Given the free Aramaic rendering of the book of Job in 11QtgJob,138 it seems likely to me that Gamaliel the Elder condemned his Aramaic translation for a textual reason. Only the three rabbinic texts linked the story about Gamaliel the Elder with the question of the legitimacy of later Targumim. It therefore seems that two procedures existed that were employed by sages and priests at the Jerusalem temple to achieve textual standardization. In the case of some biblical books and/or biblical texts, whole textual versions and texts were excluded from use. For each biblical book, the temple authorities decided on either a specific manuscript or a specific text tradition. In the case of the Books Found in the Temple at Jerusalem,” JQR 8 (1917–1918), 385–423; S. Zeitlin, “Were There Three Torah-Scrolls in the Azarah?” JQR 56 (1966), 269–272; Kutscher, Language (see n. 76), 80–81; and esp. S. Talmon, “The Three Scrolls of the Law That Were Found in the Temple Court,” Textus 2 (1962), 14–27. 136  Translation according to J. N eusner , Besah and Tanit, Vol. 18 of The Talmud of the Land of Israel (Chicago: University of Chicago Press, 1982–1994), 253–254. 137  See already Blau, Studien (see n. 114), 100–104, who, however, doubts the historicity of most of this passages. 138  D. Shepherd, “11.3.3.2 Job: Primary Translations; Targum and Qumran Aramaic Versions; Qumran Aramaic Versions,” in Textual History of the Bible (ed. A. Lange; Brill Online, 2016), http://referenceworks.brillonline.com/entries/textual-history-of-the-bible/11332-qumran-aramaic-versions-COM_00000240 (accessed 3/3/16).

From Many to One

171

Torah, it remains unclear whether the temple authorities made this decision for the Torah as a whole or individually for each of its books. That the text of MTL varies in its use of baroque spellings between Genesis, Exodus, Leviticus, Numbers, and Deuteronomy could argue for the latter. In addition, the text of some, if not all, biblical books underwent text-critical work to achieve a high-quality standard text. Y. Taʿan. 4:4 suggests, for example, that this text-critical work compared various semi-Masoretic manuscripts of the Torah to make majority decisions about orthographic disagreements and smaller textual differences. Some time before, or more likely, during, this process, the manuscripts that were to become the proto-Masoretic text underwent orthographic revision that eliminated most nonconservative spellings from their texts (see below). Although textual standardization was originally due to Hellenistic influence, the editorial schemes of the Alexandrian library do not seem to have influenced the text-critical methods employed at the Jerusalem temple as compared to the approach described above. The majority decisions of the correctors, described in y. Taʿan. 4:4, have nothing in common with the methodology of the Alexandrian scholars.139 The result of the textual standardization of the Hebrew Bible was the consonantal text of MT, the proto-Masoretic text. How close the ancient proto-Masoretic standard text was to the consonantal text of MTL is illustrated for the Torah in the following table: Table 9: Variants toward MTL in proto-Masoretic Dead Sea Scrolls Paleographic Date

proto-Masoretic manuscript

50–25 BCE

4QDeute

126 words; 1 time against MT; one orthographic variant

30–31 BCE

MasLeva MasDeut

63 words; one orthographic variant 68 words; one orthographic variant

10 BCE–30 CE MasLevb

457 words; no textual and orthographic variants (Lev 11:28 ‫ הםה‬instead of ‫)המה‬

1–50 CE

4QDeutg

151 words; 1 possible variant against MT; no orthographic variants

50–100 CE

4QGenb Mur/ḤevLev XHev/SeNumb SdeirGen

358 words; no textual variants; one orthographic variant 107 words; no textual and orthographic variants 95 words; no textual and orthographic variants 117 words; no textual and orthographic variants

100–135 CE

MurGen-Exod, 260 words; no textual and orthographic variants Numa

139  For the Alexandrian editorial strategies, see R. P feiffer , History of Classical Scholarship: From the Beginnings to the End of the Hellenistic Age (Oxford: Clarendon Press, 1968), 105–122, 171–181; and the study of H. van T hiel , “Der Homertext in Alexandria,” ZPE 115 (1997), 13–36.

172

Armin Lange

The table points to the surprising lack of textual and even orthographic variation in the proto-Masoretic Torah manuscripts toward the consonantal text of MTL. In a total of 1802 preserved words of Torah text, only one certain and one possible textual variant toward MTL occur, and for the same 1,802 words, only five orthographic variants can be detected. Although 1,802 words is a relatively small amount of text compared to the 79,978 words of the Torah, these statistics argue strongly that the proto-Masoretic standard text created at the Herodian temple was virtually identical to the consonantal text of MTL in both its text and its orthography. An orthographic standardization was therefore most likely a part of the textual standardization of the proto-Masoretic text. It is highly probable that this orthographic standardization was achieved through an orthographic recension for some of the biblical books. But the absence of baroque spellings in the book of Canticles makes it likely that in other cases the “correctors” agreed on one manuscript that already followed the conservative orthographic approach as their standard text. Rabbinic evidence makes it likely that once this standard text existed, it was carefully maintained at the Jerusalem temple by way of master copies against which all newly created copies of a biblical book were corrected. The extent to which biblical manuscripts in synagogues and other places outside the temple benefitted from the textual maintenance of correction in reference to master copies remains unclear due to insufficient evidence. If the proto-Masoretic standard text was created in the second half of the first century BCE, the existence of non-Masoretic manuscripts with conservative or baroque orthography in the first century CE (see above table 4 and table 7) leaves little doubt that the proto-Masoretic standard text needed some time after its creation before it became accepted. For this brief period, the textual situation of the first century CE can be compared with the transmission of the Homeric text. A maintained standardized text of the Hebrew Bible coexisted with manuscripts that were even more popular than the standard text in certain circles. These popular texts can be described as a vulgar texts. Before the creation of the proto-Masoretic standard text, manuscripts that became vulgar texts in the first century CE were regarded as regular manuscripts of the Jewish scriptures without distinction from any other biblical text. To summarize: In the second half of the first century BCE, the textual standardization of the Hebrew Bible responded to the increased influence of Greco-Roman culture after the Roman conquest of Judah. Those responsible for the first efforts toward a standard text of Jewish scriptures were Jewish (priestly) elites who were highly educated both in Jewish and in Greco-Roman culture. Whether these priests were called “correctors” (‫ )מגיהין‬remains unknown. In the case of some biblical books (e.g., Job and Canticles), they chose a specific manuscript, but for others (e.g., Genesis–Deuteronomy), they created a standard text by way of statistical comparison of several manuscripts of a specific textual

From Many to One

173

version. Because all proto-Masoretic manuscripts among the Dead Sea Scrolls agree with the consonantal text of MTL even in small orthographic idiosyncrasies, the textual standardization of the Hebrew Bible included an orthographic standardization as well. The text-critical work to create this standard text concerned mostly, if not exclusively, small textual variants and orthographic differences. 3.3 The Late Ancient and Early Medieval Transmission of proto-MT Very little manuscript evidence survives for the textual transmission of the Hebrew text of the Bible from the period after the end of the Second Jewish War (135 CE). The first extant Masoretic manuscripts are from the ninth century. The five manuscripts that might precede the ninth century (see above, 1.2) are all proto-Masoretic as far as their textual character can still be determined. Masoretic annotations point to the existence of earlier western and eastern master copies, which are now lost (see above 1.2), and whose preserved variants require further investigation. However, these lost master copies in all probability attested to proto-Masoretic texts as well. Further back in time reaches the Severus Scroll variant list and the variant readings attributed to Rabbi Meir.140 The Severus Scroll variant list not only details the variant readings of a Torah scroll of the late Second Temple period but also quotes a proto-Masoretic text to compare these readings with (SevComp). Although SevComp differs slightly from the consonantal text of MTL, it still provides confirmation of the authority of the proto-Masoretic text in late antiquity. The same is true of the five variant readings attributed to a Torah scroll and to an Isaiah scroll of Rabbi Meir. I have argued elsewhere that these readings were transmitted in the form of a variant list.141 Although both the readings of the Severus Scroll and the readings of Rabbi Meir disagree with MTL either in orthography or in text (or both), the very fact that they were transmitted as oddities illuminates the dominant role of the proto-Masoretic text. Rabbinic and postrabbinic literature also records textual changes in the Septuagint intended to please the Ptolemaic king (y. Meg. 1:9 and parallels).142  See 1.2 and the literature quoted there.  Lange, “Rabbi Meir” (see n. 2). 142  Z. F rankel , Vorstudien zu der Septuaginta (Leipzig: Vogel, 1841; repr. Westmead: Gregg, 1972), 25–32; V. A ptowitzer , “Die rabbinischen Berichte über die Entstehung der Septuaginta,” Haḳedem 2 (1908), 11–27, 102–133; 3 (1900), 4–17, esp. part 1:11–27; McCarthy, The Tiqqune Sopherim (see n. 51), 131–137; E. Tov, The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999), 1–20 (slightly revised version of JSJ 15 [1984], 65–89); G. Veltri, Eine Tora für König Talmai: Untersuchungen zum Übersetzungsverständnis in der jüdisch-hellenistischen Literatur (TSAJ 41; Tübingen: Mohr Siebeck, 1994); E. Tov, Greek and Hebrew Bible, 75–82 (slightly revised version of the authorʼs review of Veltriʼs book in Scripta Classica Israelica 14 [1995], 178–183). 140 141

174

Armin Lange

The lists of ʿIṭṭur Sopherim (‫עיטור סופרים‬, “omission of the scribes”; b. Ned. 37b)143 and the Tiqqune Sopherim (‫תיקוני סופרים‬, “corrections of the scribes”)144 are further examples of rabbinic lists of biblical readings. Except for the Severus Scroll variant list and the readings attributed to Rabbi Meir, these rabbinic lists go back to exegetical interests of the rabbis. However, despite their exegetical motivation, these rabbinic lists still illustrate some of the roots of the work of the Masoretes. 143  B. Ned. 37b lists five places (Gen 18:5; 24:55; Num 31:2; Ps 36:7; 68:26) where a conjunctive waw is expected but is missing. For the corrections of the scribes, see G eiger , Urschrift (see n. 135), 251–257; G insburg , Introduction (see n. 6), 307–309; A. W eiser , “Corrections of the Scribes,” in The Book of Auerbach: Essays in Biblical Research; Offered in Honor of Eliyahu Auerbach on the Occasion of his Seventieth Birthday (ed. A. Biram; Jerusalem: Kiryat Sepher, 1955), 216–221 (Hebrew); Yeivin, Introduction (see n. 4), 51–52; P. H. K elley et al., The Masorah of the Biblia Hebraica Stuttgartensia: Introduction and Annotated Glossary (Grand Rapids, MI: William B. Eerdmans, 1998), 40; R. E. Fuller, “The Text of the Tanakh,” in The Medieval through the Reformation Periods, Vol. 2 of A History of Biblical Interpretation (ed. A. J. Hauser and D. E. Watson; Grand Rapids, MI: William B. Eerdmans, 2009), 213–214; Tov , Textual Criticism (see n. 31), 61; G. K han , A Short Introduction to the Tiberian Masoretic Bible and its Reading Tradition (2d ed.; Piscataway: Gorgias Press, 2013), 29. 144  The earliest lists are attested by the Mekhilta of Rabbi Ishmael Shira 6 (at Exod 15:7) and Siphre on Numbers § 84 (at Numbers 10:35). For the Tiqqune Sopherim, see, e.g., Geiger, Urschrift (see n. 135), 308–345; Ginsburg, Introduction (see n. 6), 347–363; W. E. Barnes, “Ancient Corrections in the Text of the Old Testament (Tiḳḳun Sopherim),” JTS 1 (1899– 1900), 387–414; Lieberman, Hellenism (see n. 115), 28–37; Y. Levinson, “Eighteen Scribal Emendations,” in Jubilee Volume: Presented in Honor of Rabbi Dr. Israel Elfenbein by His Friends and Admirers; In Honor of Thirty Years of Creativity in Torah and Scientific Inquiry (ed. J. Rosenthal; Jerusalem: Mossad ha-Rav Kook, 1962), 138–139 (Hebrew); W. McKane, “Observations on the Tiḳḳûnê Sôperim,” in On Language, Culture and Religion: In Honor of Eugene A. Nida (ed. M. Black; The Hague: Mouton, 1974), 53–77; D. Barthélemy, Études dʼhistoire du texte de lʼAncien Testament (OBO 21; Fribourg: Éditions Universitaires, 1978), 91–110; Yeivin, Introduction (see n. 4), 49–51; McCarthy, The Tiqqune Sopherim (see n. 51); R. E. Fuller, “Early Emendations of the Scribes: The Tiqqune Sopherim in Zechariah 2:12,” in Of Scribes and Scrolls: Studies on the Hebrew Bible, Intertestamental Judaism, and Christian Origins (ed. H. W. Attridge et al.; Lanham: University Press of America, 1990), 21–28; idem, “Text” (see n. 143), 213; L. Díez Merino, “Los Tiqqune Soferim en la Tradición Targúmica,” in Tradition of the Text: Studies Offered to Dominique Barthélemy in Celebration of his 70th Birthday (OBO 109; Göttingen: Vandenhoeck & Ruprecht, 1991), 18–44; M. A. Zipor, “Some Notes on the Origin of the Tradition of the Eighteen Tiḳḳûnê Sôperim,” VT 44 (1994), 77–102; Kelley et al., Masorah (see n. 143), 37–40; S. Schorch, Euphemismen in der Hebräischen Bibel (Orientalia et Christiana 12; Wiesbaden: Harrassowitz, 2000), 22–24; Tov , Textual Criticism (see n. 31), 59–61; Khan, Short Introduction (see n. 143), 29; A. Schenker, “Die Tiqqune Sopherim im Horizont der biblischen Textgeschichte: Theologische Korrekturen, lite­ rarische Varianten im alttestamentlichen Text und Textvielfalt: wie gehen sie zusammen?” in Making the Biblical Text: Textual Studies in the Hebrew and the Greek Bible (ed. I. Himbaza; OBO 275; Göttingen: Vandenhoeck & Ruprecht, 2015), 33–47.

From Many to One

175

The stability of the proto-Masoretic text in rabbinic times is also emphasized by the lists of words that “are read but not written” (‫ )קריין ולא כתבין‬and words that “are written but not read” (‫ ;כתבין ולא קריין‬b. Ned. 37b–38a; cf. also the later tractate Sop. 6:8–9).145 The rabbis thus made special regulations for words they deemed either lacking or superfluous in the biblical text, which they therefore wanted added or ignored when it was read to an audience. That the rabbis made these halakic efforts instead of simply changing the biblical text demonstrates how unwilling they were to alter the text of the Bible even in cases where they disagreed with it. Further evidence of how untouchable the biblical text was for the rabbis can be found in ʾAbot de Rabbi Nathan A 34:6, which describes how doubtful words are dotted above the line but remain in the biblical text.146 The same is true for references to biblical words that rabbinic texts introduce with the formula ʾal tiqre (‫אל תקרי‬, “do not read”).147 Examples can be found in Lev. Rab. 18:3; b. ʿErub. 54a; and b. Ber. 64a, although these references concern only questions of vocalization (e.g., the recommendation of b. ʿErub. 54a to ֵ “freedom,” instead of ‫חרּות‬, ָ “engraved,” in Exod. 32:16). Instead of read ‫חרּות‬, introducing clarifying matres lectiones (e.g., ‫ ֵחירּות‬in Exod. 32:16), the rabbis restricted themselves to a reading instruction. The above examples document an approach of the rabbis to the biblical text that is quite different from the license that scribes of the Second Temple period took to change the text of biblical books at will while they copied it. The general dominance of the proto-Masoretic text is also confirmed by various non-Hebrew sources. The first column of Origenʼs Hexapla contained a proto-Masoretic Hebrew text and the second column a Greek transcription of that same proto-Masoretic Hebrew text.148 The little text that survived of the Hexaplaʼs second column is complemented by isolated Latin transcriptions and the remarks about Hebrew readings in the œuvre of Jerome. Jewish translations of the Hebrew Bible from rabbinic times such as the Targumim, Aquila, and  For the Qere we-la‍ʾ Ketiv and Ketiv we-la‍ʾ Qere lists, see Yeivin , Introduction (see n. 4), 58; Kelley et al., Masorah (see n. 143), 40–41; Tov, Textual Criticism (see n. 31), 56. 146  See Lieberman, Hellenism (see n. 115), 43–46; Kutscher, Language (see n. 76), 81. 147  For the ʾal tiqre readings, see A. Rosenzweig, “Die Al-tikri-Deutungen,” in Festschrift zu Israel Lewyʼs siebzigsten Geburtstag (ed. M. Brann and I. Elbogen; Breslau: Marcus, 1911; reprint Jerusalem: Makor, 1972), 204–253; G ordis , Biblical Text (see n. 117), 78; Y eivin , Introduction (see n. 4), 61–62; M c C arthy , The Tiqqune Sopherim (see n. 51), 139–166; Fuller, “Text” (see n. 143), 214. 148  For the discussion about the first and second column of Origenʼs Hexapla, see R. G. Jenkins, “The First Column of the Hexapla: The Evidence of the Milan Codex (Rahlfs 1098) and the Cairo Genizah Fragment (Rahlfs 2005),” in Origenʼs Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th–3rd August 1994 (TSAJ 58; Tübingen: Mohr Siebeck, 1998), 73–87; G. J. Norton, “Observations on the First Two Columns of the Hexapla,” in op. cit., 103–24; P. W. Flint, “Columns I and II of the Hexapla: The Evidence of the Milan Palimpsest (Rahlfs 1098),” in op. cit. 125–132. 145

176

Armin Lange

Symmachus are all based on a proto-Masoretic text. And Jerome rendered the Vulgate from proto-Masoretic scrolls that Jews provided to him. At least to a certain extent, the Peshitta relies on MT as its base text, too. This wealth of translations does not indicate a wide dispersion of biblical manuscripts. On the whole, Torah scrolls were rare outside the rabbinic movement. Thus, t. B. Meṣiʿa 11:23 rules that the inhabitants of a village can force each other to buy a Torah scroll and scrolls of the Prophets. That y. Meg. 4:1 prohibits reading the Torah from memory shows that such a recitation by heart was practiced even in synagogues due to lack of Torah scrolls. Y. Meg. 4:1 reasons as follows: “Even one who is experienced in the reading the Torah as was Ezra should not recite it from memory”149 (Gen. Rab. 36:8). Various remarks in rabbinic literature make it nevertheless likely that the proto-Masoretic text was maintained in rabbinic times.150 Corrected and/or uncorrected scrolls and the correction of scrolls are mentioned in rabbinic literature not only in connection with the Jerusalem temple (see above, 3.2) but for the time of the rabbis as well. This process of correction is described by the root ‫ נגה‬in the hiphʿil, which has various meanings in this stem: “to look over; to revise a manuscript, to correct, restore (when faded).”151 It is not always clear whether the meaning is the repair of a scroll whose ink has faded or the correction of scribal errors. Given the importance of this issue for the transmission of the proto-Masoretic text in rabbinic times, I will quote at least a part of the evidence to make my point. The Babylonian Talmud warns against the retention of “uncorrected books” in b. Ket. 19b:152 It has been stated: A book that is not corrected (‫ – )ספר שאינו מוגה‬R. Ami said: Until thirty days one is allowed to keep it, from then and further on, it is forbidden to keep it, because it is said: Let not unrighteousness dwell in thy tents.

B. Ber. 13a deliberates about a statement of the Mishnah (m. Ber. 2:1–2) regarding the recitation of the Shemaʿ. The Mishnah rules that if a person reads the Shemaʿ at the time of its recitation instead of actually reciting it, the intention would be recognized. The Gemara specifies that in the Mishnah the word “to read” must refer to revising a Deuteronomy scroll: This proves that precepts must be performed with intent. [No, perhaps] what “if he had the intention” means is, if it was his intention to read the Scripture? “To read”? But surely he is reading! – [The Mishnah may refer] to one who is reading [a scroll] in order to revise it (‫)בקורא להגיה‬.  Translation according to J. Neusner, Megillah, Vol. 19 of Talmud of the Land of Israel (Chicago: University of Chicago Press, 1982–1994), 145. 150  Again, I am obliged to Günter Stemberger for his support in my analysis of the rabbinic evidence (see above n. 114). 151  See Jastrow, Dictionary (see n. 125), 872. 152  Translations of the Babylonian Talmud are taken from the Soncino translation if not noted otherwise: I. E pstein , The Babylonian Talmud (18 vols.; London: Soncino Press, 1938–1952). 149

From Many to One

177

Quite similar is a statement of the Mishnah itself (m. Meg. 2:2) regarding the obligation to read the book of Esther. M. Meg. 2:2 regards the correction of an Esther scroll as a fulfillment of this obligation: M. Megillah 2:2 If a man read it piecemeal or drowsily, he fulfilled his obligation; if he was copying it, expounding it, or correcting a copy of it (‫)ומגיהה‬, and he directed his heart [to the reading of the Scroll], he fulfilled his obligation; otherwise he has not fulfilled his obligation.

That these passages do not formulate a rule to correct a biblical scroll but presuppose the existence of such a practice to make another argument confirms how widespread and accepted the existence of corrected scrolls must have been. The widespread use of corrected scrolls is also reflected in a statement attributed to Rabbi Akiba: b. Pesaḥim 112a If you wish to be strangled, be hanged on a large tree, and when you teach your son, teach him from a corrected scroll (‫)בספר מוגה‬.

Akibaʼs statement sounds proverbial and presupposes thus a general agreement that a father should teach his son from a corrected scroll. Regardless of whether this statement actually goes back to the sage or – more likely – was only attributed to him, b. Pes. 112a points to a widespread agreement about this use of corrected scrolls in teaching. Two passages from tractate y. Šabbat may illustrate how the rabbis envisioned the procedure of reviewing and/or correcting a biblical scroll. Both statements are part of an interpretation of m. Šabb. 16:1 ruling that on the Sabbath scriptures may be saved from burning, even including those scrolls that are not read on the Sabbath. The rabbis discuss what the phrase “whether they read in them or do not read in them” (‫ )בין שקורין בהן ובין שאין קורין בהן‬means. For the rabbis of Talmud Yerushalmi, one way to explain the difference between scrolls that are read and those that are not read is to refer to the number of corrections they contain. y. Šabbat 16:1 And lo, it has been taught: As to a scroll which contains two or three errors (‫ )טעיות‬on every column, one corrects (‫ )מתקנו‬that scroll and reads in it. If there are four, he does not read in it.153

The number of scribal mistakes a scroll contained per page determined whether it could be read for liturgical purposes or not. Although y. Šabb. 16:1 employs a form of the root ‫( תקן‬pael “to repair/correct”) instead of the root ‫( נגה‬hiphil “to correct”), it nevertheless indicates the same act of scribal correction, because the 153  Translation according to J. Neusner, Shabbat, Vol. 1 of Talmud of the Land of Israel (Chicago: University of Chicago Press, 1982–1994), 408.

178

Armin Lange

noun ‫ טעיות‬clearly refers to “mistakes.” As an a propos, it should be noted that y. Šabb. also establishes an upper limit for how many mistakes per page scrolls can have if they are to be read. As the correction of a scroll presupposes the use of a reference text, the evidence quoted above implies that such master copies were accessible for rabbinic scribes despite the lack of biblical scrolls in some parts of the diaspora. A late reflection of such a corrected master copy might be preserved in Masoretic annotations, which report the readings of the so-called Codex Mugah.154 It is referred to as “in the scroll Mugah” (‫)בספר מוגה‬, “in the Mugah” (‫)במוגה‬, or simply “Mugah” (‫ )מוגה‬in the Masoretic annotations of MTB and MTP and of various other Masoretic codices. The technical term “corrected scroll” thus became the name of an important master copy. The Codex Mugah was accessible through copies in different parts of the Jewish world and seems to have been read mostly with the proto-Masoretic textus receptus.155 The degree to which the Severus Scroll and the famous Torah and Isaiah scrolls of Rabbi Meir, discussed above, were used in late antiquity as master copies is difficult to decide. However, because variant lists existed that documented readings of these biblical scrolls, it seems not unlikely that they once functioned as master copies. That rabbinic literature emphasizes the use of corrected scrolls strongly most likely also reflects the existence, and perhaps even widespread use, of uncorrected scrolls. Alongside the maintained proto-Masoretic text, other biblical texts might have existed in rabbinic times as well. The dominance of the proto-Masoretic standard text in rabbinic times should thus not hide the fact that even in this period a limited amount of textual plurality existed as well. This is confirmed by the biblical text in rabbinic quotations of the Jewish scriptures. It must be admitted that many of the non-Masoretic readings that are attested in rabbinic quotations of the Bible are either due to scribal errors in the transmission of rabbinic literature or go back to exegetical changes inserted by the quoting rabbinic text into its biblical quotations. Nevertheless, some non-Masoretic variant readings survive in rabbinic quotations of the Jewish scriptures.156 Some of these non-Masoretic readings may go back to the Second Temple period, while others were caused by scribal errors of biblical copyists in the rabbinic period. On the whole, the textual situation in rabbinic times might be comparable to the limited textual variety of medieval Masoretic manuscripts. Although the Tiberian (Ben Asher) text was highly regarded, medieval Masoretic manuscripts  For the Codex Mugah, see Ginsburg, Introduction (see n. 6), 429–431, and the literature quoted in n. 27. Yeivin, Introduction (see n. 4), 138, doubts that the name “Mugah” refers to a specific scroll and regards it as a more general term for a corrected scroll. 155  See Ginsburg, Introduction (see n. 6), 430–431. 156  See Lange, “1.7.1 Jewish Quotations and Allusions” (see n. 3) and Rosen-Zvi, “Rabbinic Literature” (see n. 3) and the literature quoted in these articles. 154

From Many to One

179

attest to a certain amount of textual variation, as already documented by the work of Kennicott and De Rossi.157

4 Higher Criticism of the Torah in Light of Its Textual History The above survey of the textual history of the Torah from the third century BCE through late antiquity and the early Middle Ages has many implications for the historical-critical study of the Torah, which I will address here in place of a conclusion. As a specific example, at the end of this article I will discuss the importance of the textual history of the Torah for determining the date of its final redactional stage. 4.1 General Implications On the whole, the consonantal text of MT remains the best and most important textual witness to the Hebrew biblical text. It was created as a text of the whole Hebrew Bible when the latter reached canonical status and has existed as a textual version of the Jewish Bible only since it achieved canonical status. In this sense, MT can be considered the Jewish text of the biblical canon. Nevertheless, the textual history of the MT text of the Torah does not begin with the Jewish canon. The Torahʼs proto- and/or semi-Masoretic text existed already before the canon was established. Often, but not always, the proto- and/or semi-Masoretic manuscripts attest to the best available text of the Torah. It is therefore important to emphasize that the textual character of the proto-Masoretic text of the Torah changes to a certain extent from book to book and even chapter to chapter. In some cases, it is rather close to its final redaction (e.g., MT-Lev), while in other cases scholars debate whether it attests to a variant literary edition. For some parts of the Torah, MT might thus not represent the best available text. The most important example of this is the Old Greek translation of Exodus (see below 4.2.3.2). Vervenne has claimed that the proto-Masoretic text could be a reworking of the Hebrew Vorlage of OG-Exodus: “It is quite possible that the literary formation process of LXX-Vorlage was completed prior to that of MT, while MT in its pre-Masoretic consonantal form continued to be retouched”.158 Against such claims, it must be emphasized that in general the OG text of Exodus harmonizes its Hebrew Vorlage.159 “But the dominant characteristic of 157  Kennicott, Vetus Testamentum hebraicum (see n. 39); J. B. De Rossi, Variae lectiones Veteris Testamenti (5 vols.; Parma: Regio, 1784–1788; repr. Amsterdam: Philo, 1969). 158  M. Vervenne, Exodus 14,20 MT – LXX: Textual or Literary Variation? in Lectures et relectures de la Bible, Festschrift P.-M. Bogaert (ed. J. M. Auwers and A. Wénin; BEThL 44, Leuven: Peeters, 1999), 3–25, here 25. 159  See J. W. Wevers, Translation and Canonicity: A Study in the Narrative Portions of the Greek Exodus, in Scripta signa vocis, Festschrift J. H. Hospers (ed. H. L. J. Vanstiphout et al.;

180

Armin Lange

Exod as a translation document is its expansionist character. On the whole Exod expands far more than contracts. Where the Hebrew is abrupt, the Greek tends to smoothen out the text.”160 It is furthermore important to recognize that over time the consonantal text of (proto-)MT accumulated individual secondary readings in the Torah and elsewhere from the earliest stages of its textual history until the production of MTL as well as in later medieval Masoretic manuscripts. These individual secondary readings were inserted either unintentionally, by way of scribal error, or intentionally, by way of harmonizing readings, editorial readings, linguistic actualizations, and so forth. An example of a late secondary reading even in MTL itself can be found in Jonah 1:3: In this verse, MTL originally read for the second occurrence of ‫ּת ְר ִשׁישׁ‬, ַ “Tarshish,” ‫תרשישה‬, “to Tarshish.” But a later scribe erased the he locale to create the text of Jonah 1:3 also known from MTA and MTC.161 In addition to such late scribal adjustments, the MT text of the Torah also includes earlier secondary readings. I have already discussed the example of the monotheistic reworking of MT-Deut 32:8, 43 as compared to 4QDeutq (see above 2.2 and below 4.2.4). A comparison of 4QGen–Exoda with MT provides more examples of such early secondary readings in the MT text of the Torah. 4QGen–Exoda dates paleographically to the years 125–100 BCE,162 before the creation of the proto-Masoretic standard text of the Torah, as argued above (see 3.2). It is impossible to discuss all twenty variant readings of 4QGen-Exoda toward MTL. It needs to be emphasized that both MTL and 4QGen-Exoda contain original as well as secondary readings. Secondary readings in 4QGen-Exoda can be found in Gen 22:14; Exod 4:6; and 6:8. Secondary readings in MTL can be found at least in Gen 35:19, 23, 25, 26; Exod 1:5, 14; 3:13, 15. Given how close MTL and 4QGen-Exoda are textually, it seems likely that both texts share a common textual ancestor, which was reworked in both the proto-Masoretic text and in 4QGen-Exoda. Due to space constraints, in this article I can engage only with two examples from MTLʼs secondary readings. In Exod 1:5, 4QGen-Exoda reads ‫שבעים] וחמש‬, “seventy]-five,” instead of the simple ‫שבעים‬, “seventy” in MT and SP. Both MT and SP corrected the number of seventy-five descendants of Jacob to harmonize the text with Gen 46:27 and Deut 10:22 to the number seventy. The original reading for Exod 1:5 is preserved in 4QGen-Exoda. Exod 1:5 speaks of seventy-five descendants of Groningen 1986), 295–303. 160  J. W. W evers , Text History of Greek Exodus (MSU 21; Göttingen: Vandenhoeck & Ruprecht, 1992), 148. 161  Cf. A. G elston , The Twelve Minor Prophets (Biblia Hebraica Quinta 13; Stuttgart: Deutsche Bibelgesellschaft, 2010), 92*. 162  For the paleographic date of 4QGen–Exoda, see J. R. D avila , “1. 4QGen–Exoda,” in Qumran Cave 4.VII: Genesis to Numbers (ed. E. Ulrich et al.; DJD XII; Oxford: Clarendon Press, 1994), 7–30, here 8.

From Many to One

181

Jacob because it includes the children and grandchildren of Joseph, unlike Gen 46:27 and Deut 10:22. In Exod 1:14, 4QGen-Exoda reads the singular ‫וימרר‬, “and it made bitter,” instead of the plural ‫וַ יְ ָמ ְררּו‬, “and they made bitter,” in MT, SP, and LXX. The subject of the verb is the singular ‫מצרים‬, “Egypt,” in Exod 1:13. Exodus 1:14 emphasizes that Egypt made the lives of the Israelites bitter. MT, SP, and LXX use the plural form ‫ וימררו‬instead, in keeping with the plural verbal forms of Exod 1:12–13. These few examples demonstrate that the consonantal text of MT can never be regarded as representing the final redactional stage of a biblical book word for word. Even in the many parts of the Torah where MT might be close to it, individual secondary readings can and do occur. This situation requires great caution in applying source and redaction criticism to the biblical text based exclusively on the consonantal text of MTL. Only when MTʼs secondary readings are identified and carefully studied by way of variant analysis can further exegetical approaches be applied. 4.2 The Stemma of the Torah Texts and the Date of the Final Redactional Stage of the Torah163 The Torah manuscripts from the Qumran library allow us to establish a relative chronology of Torah texts and text types. Because the paleographic classification of these early Torah manuscripts allows for their dating, the relative chronology of the Qumran Torah manuscripts and their texts provides a terminus ante quem for the final redactional stage of the Pentateuch. To achieve this goal, it must be asked how the early versions of the Pentateuch preserved by manuscripts from the Qumran library, in the Septuagint, and in the Samaritan Pentateuch relate to the final stage of the Torah. Do the quotations of Torah texts and their adaptions in ancient Jewish literature provide evidence of when the various texts of the Pentateuch developed? 4.2.1 The Temple Scroll and the Book of Ezra–Nehemiah Of key importance for the question of the date of the final stage of the Pentateuch is how the Book of Ezra–Nehemiah, the Temple Scroll from Qumran, and 4QRPc (4Q365) are related with each other. Nehemiah 10:35 relates that priests, Levites, and the people should use the oracle of the lot to determine the sequence in which they should bring the wood offering to the house of God (cf. Neh 13:31), and the Temple Scroll (11QTb VI 11–15) commands that, beginning with the tribe 163  This part of my article is a summary of a more detailed argument that I made in Lange, “Dead Sea Scrolls and the Date of the Final Stage of the Pentateuch” (see headnote). For a more detailed discussion of the stemma of the textual witnesses of the Torah, see now also Tov. “2.1 Textual History of the Pentateuch” (see n. 35).

182

Armin Lange

of Levi, the twelve tribes of Israel should bring wood to the altar. Both texts depend on a fragmentary regulation about the feast of wood in 4QRPc (4Q365) 23 9–11. That both the Temple Scroll and the Book of Ezra–Nehemiah use the 4QRPc text of the Pentateuch provides a fixed point in the relative chronology of the Pentateuchʼs various texts and recensions. Both the Temple Scroll and the Book of Ezra–Nehemiah are therefore qualified to provide a terminus ante quem for one of the later recensions of the Pentateuchʼs text, the 4QRPc text of the Pentateuch. The question of when the Temple Scroll and the book of Ezra–Nehemiah were written therefore becomes crucial for establishing a terminus ante quem for the final stage of the Pentateuch. The Temple Scroll (4QTb [4Q524]; 11QTa-c [11Q19–21]) combines parts of the Pentateuch with formerly unknown passages and claims to have been revealed to Moses on Mount Sinai. For the Temple Scroll, the terminus ante quem is set by its oldest preserved manuscript, 4QTb, which is dated paleographically around 150 BCE.164 The terminus post quem can be derived from relative chronology and the history of architecture. Swanson argues that the Temple Scroll depends on 1–2 Chronicles, which were written in late Persian or early Hellenistic times.165 As Broshi has shown, in Palestine, the spiral staircase described by the Temple Scroll for the northwestern tower of the sanctuary (11QTa XXX:3–XXXI:9) and the peristyles situated in the middle and the outer courtyards are attested only from Hellenistic times onward.166 Both observations argue for a terminus post quem in the third century BCE. In my opinion, the extremely free approach to the Temple Scrollʼs pentateuchal base text and the Temple Scrollʼs own claim to be a legal text that supersedes the Torah in authority argue for a date in the third century BCE for the Temple Scroll. As for the book of Ezra–Nehemiah, due to lack of space, I cannot provide a detailed reconstruction of its redaction history in the present paper. With Japhet and others,167 I understand Ezra–Nehemiah and 1–2 Chronicles as two separate  See É. Puech, Qumrân grotte 4 XVIII: Textes hébreux (4Q521–4Q528, 4Q576–4Q579) (DJD XXV; Oxford: Clarendon, 1998), 87–88. 165  D. D. Swanson, The Temple Scroll and the Bible: The Methodology of 11QT (STDJ 14; Leiden: Brill, 1995), 237–239. 166  M. B roshi , “Visionary Architecture and Town Planning in the Dead Sea Scrolls,” in Time to Prepare the Way in the Wilderness: Papers on the Qumran Scrolls (ed. D. Dimant and L. H. Schiffman; STDJ 16; Leiden, 1995), 9–22, here 19. 167  See, e.g., S. Japhet, “The Supposed Common Authorship of Chronicles and Ezra-Nehemiah Investigated Anew,” VT 18 (1968), 332–372; idem, I & II Chronicles: A Commentary (OTL; Louisville, KY: Westminster John Knox Press, 1993), esp. 3–7; H. G. M. Williamson, Israel in the Book of Chronicles (Cambridge: Cambridge University Press, 1977), 5–82, esp. 70; idem , 1 and 2 Chronicles: Based on the Revised Standard Version (NCBC; Grand Rapids, MI: Eerdmans, 1982), 5–11; G. N. Knoppers, I Chronicles 1–9: A New Translation with Introduction and Commentary (AB 12; New York, N.Y.: Doubleday, 2003), 96–100; see 164

From Many to One

183

books that were written subsequent to one another. The date of Ezra–Nehemiah thus must be determined separately from the date of 1–2 Chronicles. For this purpose, it is important to recognize that the last high priest mentioned in the lists of Nehemiah 12 is Jaddua (Neh 12:11, 22).168 Because the evidence of Josephus shows that this Jaddua was still in office during the reign of Alexander the Great (A.J. 11.302),169 a date for the final stage of the book of Ezra–Nehemiah in early Hellenistic times is likely.170 Because both the Temple Scroll and Ezra/Nehemiah employ the 4QRPc text of the Torah, this text of the Torah needs to be dated at the latest to early Hellenistic times. That both books employ the 4QRPc text of the Torah independent of each other recommends an even earlier date. 4.2.2 4QExod-Levf, the 4QRPc Text of the Torah and the Pre-Samaritan Pentateuch 4.2.2.1 4QRPc (4Q365) 4Q364–67 constitute a special group among the nonaligned Pentateuch manuscripts from Qumran. Tov and White Crawford regarded the manuscripts 4Q158 and 4Q364–67 as five copies of a reworked version of the Pentateuch,171

also G. Steins, Die Chronik als kanonisches Abschlußphänomen (BBB 93; Weinheim: Beltz Athenäum, 1995), 49–82. 168  For the high priest Jaddua, see J. C. Vander K am , From Joshua to Caiaphas: High Priests after the Exile (Minneapolis, Minn., and Assen: Fortress and Van Gorcum, 2004), 63–85. 169  F. M. Cross doubted the historicity of the lists in Neh 12:11, 22 and wanted to add at least two more names, which in his opinion would have been lost due to haplography (see, e.g., F. M. Cross, “A Reconstruction of the Judean Restoration,” JBL 94 [1975], 4–18). But based on a reevaluation of the evidence from the Wadi ed-Daliyeh papyri and Samarian coins, J. Dušek was able to prove the historicity of the lists in Neh 12:11, 22 (see Les manuscrits araméens du Wadi Daliyeh et la Samarie vers 450–332 av. J.-C. [Culture and History of the Ancient Near East 25; Leiden: Brill, 2007], 549–598; Dušek also provides a survey of the history of research). For criticism of Crossʼs theory, see also J. C. VanderKam, “Jewish High Priests of the Persian Period: Is the List Complete?,” in Priesthood and Cult in Ancient Israel (ed. G. A. Anderson and S. M. Olyan; JSOTSup 125; Sheffield: Sheffield Academic Press, 1991), 67–91 and idem, Joshua (see n. 168), 85–99. 170  For a good summary of the arguments for this date, see O. K aiser , Die erzählenden Werke, Vol. 1 of Grundriß der Einleitung in die kanonischen und deuterokanonischen Schriften des Alten Testaments (Gütersloh: Gütersloher Verlagshaus, 1992), 147–148; see also Williamson, Israel (see n. 167), 83–86; idem, 1 and 2 Chronicles (see n. 167), 15–17; idem, Ezra, Nehemiah (see n. 68), XXXV–XXXVI; Japhet, I & II Chronicles (see n. 167), 23–28; Knoppers, I Chronicles (see n. 167), 101–117. 171  For a synopsis of 4Q158 and 4Q364–67, see E. Tov , “4QReworked Pentateuch: A Synopsis of Its Content,” RevQ 16 (1993–1995), 647–653, esp. 650–653.

184

Armin Lange

which they attributed to the literary genre of Rewritten Bible.172 This supposedly extended version of the Pentateuch was originally named “Pentateuchal Paraphrase” (PP),173 but in their editio princeps, Tov and White Crawford used the name “Reworked Pentateuch” (RP).174 Against Tov and White Crawford, Segal has demonstrated that the manuscripts 4Q158 and 4Q364–67 display different characteristics and that they need to be classified differently. He showed that 4Q364–67 attest to different texts of the Torah, while 4Q158 preserves remnants

 Tov and White, “Reworked Pentateuch” (see n. 70), 187–196; see also E. Tov, “The Textual Status of 4Q364–367 (4QRP),” in The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid, 18–21 March, 1991 (2 vols.; ed. J. Trebolle Barrera and L. Vegas Montaner; STDJ 11/1; Leiden and Madrid: Brill and Universidad Complutense, 1992), 1:43–82, here 45–48; idem, “Biblical Texts as Reworked in Some Qumran Manuscripts with Special Attention to 4QRP and 4QParaGen–Exod,” in The Community of the Renewed Covenant: The Notre Dame Symposium on the Dead Sea Scrolls (ed. E. Ulrich and J. C. VanderKam; Christianity and Judaism in Antiquity 10; Notre Dame: University of Notre Dame Press, 1994), 111–134, here 123–133; S. A. White, “4Q364 & 365: A Preliminary Report,” in The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid, 18–21 March, 1991 (2 vols.; ed. J. Trebolle Barrera and L. Vegas Montaner; STDJ 11/1; Leiden and Madrid: Brill and Universidad Complutense, 1992), 1:217–228, here 218, 228; S. W. C rawford , “The Rewritten Bible at Qumran,” in The Bible and the Dead Sea Scrolls: The Princeton Symposium on the Dead Sea Scrolls (ed. J. H. Charlesworth; 3 vols.; Waco: Baylor University Press, 2006), 1:131–147, here 140–145; Crawford, Rewriting Scripture (see n. 56), 39–59; idem, “Genesis in the Dead Sea Scrolls,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; VTSup 152; Leiden: Brill, 2012), 353–373, here 354–357; G. J. Brooke, “Torah in the Qumran Scrolls,” in Bibel in jüdischer und christlicher Tradition: Festschrift für Johann Maier zum 60. Geburtstag (ed. H. Merklein et al.; BBB 88; Frankfurt am Main: Anton Hain, 1993), 97–120, here 113–114; H.-J. Fabry, “Das Buch Levitikus in den Qumrantexten,” in Levitikus als Buch: Tagung der Arbeitsgemeinschaft der Deutschsprachigen Alttestamentlerinnen und Alttestamentler vom 1. bis 5. September 1997 in Trier (ed. H.-J. Fabry and H. W. Jüngling; BBB 119; Bodenheim: Philo, 1999), 309–341, here 318; D. K. Falk, The Parabiblical Texts: Strategies for Extending the Scriptures in the Dead Sea Scrolls (Library of Second Temple Studies 63: London: T&T Clark, 2007), 107–119; I. Knohl, “The Bible Reworked at Qumran: The Temple Scroll and 4QReworked Pentateuch,” in The Qumran Scrolls and their World (ed. M. Kister; Jerusalem: Yad Izhak Ben-Zvi, 2009), 157–168 (Hebrew); A. B. Perrin, “The Variants of 4Q (Reworked) Pentateuch: A Comprehensive List of the Textual Variants in 4Q158, 4Q365–7 in Biblical Sequence,” JJS 63 (2012), 127–157. Unlike in his earlier publication (“Torah,” 113–114), Brooke suggested in response to Segalʼs ideas (see n. 175 below) that 4Q158 and 4Q364–67 attest to five different texts, which share a literary genre (“4Q158: Reworked Pentateuch or Reworked Pentateuch A?” DSD 8 [2001], 219–241). 173  See, e.g., Tov , “Textual Status” (see n. 172), 43; W hite , “Preliminary Report” (see n. 172), 217. The name “Pentateuchal Paraphrase” was suggested by John Strugnell (see E. Tov, “From 4QReworked Pentateuch to 4QPentateuch (?),” in Authoritative Scriptures in Ancient Judaism [ed. M. Popović; JSJSup 141; Leiden: Brill, 2010], 73–91, here 74). 174  Tov and White, “Reworked Pentateuch” (see n. 70). 172

From Many to One

185

of a Rewritten Bible composition.175 Therefore, 4QRPc (4Q365) needs to be treated separately from the other manuscripts that were designated “Reworked Pentateuch” in the DJD series. An added obstacle to the text-critical study of 4QRPc are the five fragments that Tov and White designated 4Q365a (4QT?).176 These five fragments were copied by the same scribe as the rest of 4QRPc and employ the same orthography,177 but they include only nonbiblical text. “Frg. 1 contains material concerning the Festival of Unleavened Bread. Frg. 2 I discusses the Day of Firstfruits and other sacrificial ordinances, and then contains the beginning of the description of the middle court of the temple. Col. II give the names of the gates of the outer court. Finally, frgs. 3, 4 and 5 I contain the specifications of certain buildings and objects probably associated with the temple.”178 Furthermore, 4Q365a 2 175  M. S egal , “4QReworked Pentateuch or 4QPentateuch?,” in The Dead Sea Scrolls Fifty Years after their Discovery: Proceedings of the Jerusalem Congress, July 10–25, 1997 (ed. L. H. Schiffman et al.; Jerusalem: Israel Exploration Society, 2000), 391–399; see also M. S egal , “Biblical Exegesis in 4Q158: Techniques and Genre,” Textus 19 (1998), 45–62. Like Segal, E. Ulrich (“The Qumran Scrolls and the Biblical Text,” in The Dead Sea Scrolls Fifty Years after their Discovery: Proceedings of the Jerusalem Congress, July 10–25, 1997 [ed. L. H. Schiffman et al.; Jerusalem: Israel Exploration Society, 2000], 51–59, 57; idem, “The Dead Sea Scrolls and the Hebrew Scriptural Texts,” in The Bible and the Dead Sea Scrolls: The Princeton Symposium on the Dead Sea Scrolls [3 vols.; ed. J. H. Charlesworth; Waco: Baylor University Press, 2006], 1:77–99, here 88) understands the manuscripts 4Q158 and 4Q364–67 as textual witnesses to the Pentateuch. He describes RP as a “variant literary edition” of the Pentateuch. But unlike Segal, Ulrich understands 4Q158 and 4Q364–67 as several copies of the same literary edition of the Pentateuch. J. C. Vander K am ʼs position (“Questions of Canon Viewed through the Dead Sea Scrolls,” in The Canon Debate [ed. L. M. McDonald and J. A. Sanders; Peabody: Hendrickson Publishers, 2002], 91–109, here 96–100) marks a middle ground between the ideas of Tov and White on the one hand and Segal and Ulrich on the other hand. VanderKam describes the RP-manuscripts as belonging to a twilight zone between biblical manuscripts and Rewritten Bible. The diverse nature of 4Q158 and 4Q364–67 is also emphasized by M. M. Z ahn (“The Problem of Characterizing the 4QReworked Pentateuch Manuscripts: Bible, Rewritten Bible, on None of the Above?” DSD 15 [2008], 315–339; idem, Rethinking Rewritten Scripture: Composition and Exegesis in the 4QReworked Pentateuch Manuscripts [STDJ 95; Leiden: Brill, 2011]). Like VanderKam, she treats the texts of 4Q158 and 4Q364–67 as something between pentateuchal manuscripts and Rewritten Scripture (Zahn, Rethinking Rewritten Scripture). Zahn thinks that without further study it cannot be decided whether 4Q158 and 4Q364–67 are pentateuchal or represent various Rewritten Bible compositions (Z ahn , “Problem”). Influenced by Segal, Tov has revised his understanding of 4Q158 and 4Q364–67. He now regards these manuscripts as five different copies of the Pentateuch, each of which attests to a different pentateuchal text (Tov, “4QPentateuch” [see n. 174], 78–82). Despite her recent publications to the contrary (see above n. 172), Crawford also describes 4Q364 and 4Q365 in her book Rewriting Scripture in Second Temple Times (see n. 56) as “regular pentateuchal texts.” 176  Tov and White, “Reworked Pentateuch” (see n. 70). 177  Tov and White, “Reworked Pentateuch” (see n. 70), 319–320. 178  Tov and White, “Reworked Pentateuch” (see n. 70), 319.

186

Armin Lange

resembles the text of 11QTa XXXVIII and XLI–XLII. The five fragments of 4Q365a could therefore be 1. a further copy of the Temple Scroll,179 2. a copy of a literary source of the Temple Scroll,180 3. a modified version of the Temple Scroll,181 or 4. five further fragments of 4QRPc.182 In the latter case, the text attested by 4QRPc would be a source of the Temple Scroll. An observation by Hartmut Stegemann is decisive.183 Stegemann argued that the damage patterns of fragments 4Q365a 2 and 4QRPc 6 resemble each 179  Thus Y. Yadin , Temple Scroll (see n. 73), 1:8–9; see also É. P uech , “Fragments du plus ancient exemplaire du Rouleau du Temple (4Q524),” in Legal Texts and Legal Issues: Proceedings of the Second Meeting of the International Organization for Qumran Studies, Cambridge 1995: Published in Honour of Joseph M. Baumgarten (ed. M. Bernstein et al.; STDJ 23; Leiden: Brill, 1997), 18–64, here 49–50, 55–57. 180  Cf. M. O. Wise, A Critical Study of the Temple Scroll from Qumran Cave 11 (SAOC 49; Chicago: University of Chicago Press, 1990), 44–59, esp. 58–59; similarly Tov , “Textual Status” (see n. 172), 48–49; Tov and W hite , “Reworked Pentateuch” (see n. 70), 320; S. W. Crawford, “Three Fragments from Qumran Cave 4 and Their Relationship to the Temple Scroll,” JQR 85 (1994), 259–273, here 260–261, 271; idem, The Temple Scroll and Related Texts (Companion to the Qumran Scrolls 2; Sheffield: Sheffield Academic Press, 2000), 14–15; B. A. Levine, “The Temple Scroll,” BASOR 232 (1978), 5–23, here 5–6. 181  Thus, e.g., B. Z. Wacholder , “The Fragmentary Remains of the 11QTorah (Temple Scroll): 11QTorahb and 11QTorahc plus 4QparaTorah Integrated with 11QToraha,” HUCA 62 (1991), 1–116, here 4 (either a reworked version of the text of 11QTa or a divergent source). 182  See J. Strugnellʼs letter to B. Z. Wacholder, quoted in B. Z. Wacholder , The Dawn of Qumran: The Sectarian Torah and the Teacher of Righteousness (HUCM 8; Cincinnati: Hebrew Union College Press, 1983), 205–206; H. Stegemann, “The Origins of the Temple Scroll,” in Congress Volume: Jerusalem 1986 (ed. J. A. Emerton; VTSup 40; Leiden: Brill, 1988), 235–256, here 237; idem , “The Literary Composition of the Temple Scroll and Its Status at Qumran,” in Temple Scroll Studies: Papers Presented at the International Symposium on the Temple Scroll, Manchester, December 1987 (ed. G. J. Brooke; JSPSup 7; Sheffield: JSOT Press, 1989), 123–148, here 125–126, 135; F. García Martínez, “New Perspectives on the Study of the Dead Sea Scrolls,” in Perspectives on the Study of the Old Testament and Early Judaism: A Symposium in Honour of Adam S. van der Woude on the Occasion of his 70th Birthday (ed. F. García Martínez and E. Noort; VTSup 73; Leiden: Brill, 1998), 230–248, here 237–239; idem, “The Temple Scroll and the New Jerusalem,” in The Dead Sea Scrolls after Fifty Years: A Comprehensive Assessment (2 vols.; ed. P. W. Flint and J. C. VanderKam; Leiden: Brill, 1998–1999), 2:431–460, here 435; E. Qimron, The Temple Scroll: A Critical Edition with Extensive Reconstructions (Judean Desert Studies; Jerusalem: Israel Exploration Society, 1996), 4; S. W. Crawford, “4QTemple? (4Q365a) Revisited,” in Prayer and Poetry in the Dead Sea Scrolls and Related Literature: Essays in Honor of Eileen Schuller on the Occasion of Her 65th Birthday (ed. J. Penner et al.; STDJ 98; Leiden: Brill, 2012), 87–95; M. M. Zahn, “4QReworked Pentateuch C and the Literary Sources of the Temple Scroll: A New (Old) Proposal,” DSD 19 (2012), 133–158. 183  Oral communication to the author; see Lange, Handschriften (see n. 29), 40.

From Many to One

187

other so much that both fragments must have been part of the same scroll. If 4Q365a 2 and 4QRPc 6 come from the same scroll, then all other fragments of 4Q365a need to belong to 4QRPc as well. This means in turn that the text of 4QRPc preserves one of the sources of the Temple Scroll. Including the fragments wrongly designated as 4Q365a, sixty-seven fragments survive of 4QRPc. Fragments 1–38 attest to 1204 words and partial words (x indicates unidentified text included in the fragments) out of Gen 21:9–10; Exod 8:13–19; 9:9–12; 10:19?, 20; 14:10, 12–17, 19–21; 15:16–20, x, 22–26; 17:3–5; 18:13–15; 26:34–36; 28:16–17, 19–20; 29:20–22; 30:37–38; 31:1–3; 35:3–5; 36:32–38; 37:29; 38:1–7; 39:1–5, 8–19; Lev 11:1–2, 17, 21–24, 32, 40–42, 44–45; 13:6–8, 15, 17–18, 51–52; 16:11–12?; 18:26–28; 23:42–44; 24:1–2, x; 25:7–9; 26:17–28, 30–32; 27:34?; Num 1:1–5; 3:26–30; 4:47–49; 7:1, 78–80; 8:11–12; 9:15, 17–19, 22–23; 10:2–3; 13:12, 14–15, 13, 16–25, 29–30; 15:26–28; 17:20–24; 27:11; 36:1–2; Deut 19:20–21; 20:1. Fragment 37 attests to remnants of either Deut 2:24 or 2:36.184 Because passages of all five books of the Torah are extant from 4QRPc, the scroll originally included the whole Pentateuch. This hypothesis is confirmed by the large column height of 4QRPc (43–47 lines or 30.1–32.25 cm).185 Stegemann estimated the original length of 4QRPc to ca. 25m.186 Tov and White date 4QRPc paleographically to the years 75–50 BCE and describe its script as “transitional between the late Hasmonean and early Herodian periods.”187 The orthography of 4QRPc is baroque and somewhat fuller than the one of 4QRPb.188 4QRPc reads fifty-eight times with and 146 times against MT, sixty-nine times with and 135 times against SP, thirty-six times with and 146 times against LXX, as

184  E. J. C. Tigchelaar, “Gleanings from the Plates of Unidentified Fragments: Two PAM 43.674 Identifications (4Q365 and 4Q416),” in ‘Go Out and Study the Land’ (Judges 18:2): Archaeological, Historical and Textual Studies in Honor of Hanan Eshel (ed. A. M. Maeir et al.; JSJSup 148; Leiden: Brill, 2011), 317–322, here 318–319, identifies PAM 43.691 frg. 11 as a further fragment of 4Q365, which would attest to four words and partial words out of Exod 22:5–6. Although Tigchelaarʼs identification seems plausible, it must remain speculative given that only eleven characters are preserved of this fragment. 185  For the measures of 4Q365, see Tov and White, “Reworked Pentateuch” (see n. 70), 256–257. 186  Stegemannʼs work is unpublished but summarized in Tov, “Textual Status” (see n. 172), 53, and in Tov and White, “Reworked Pentateuch” (see n. 70), 192. 187  White, “Preliminary Report” (see n. 172), 271; see also Tov and White, “Reworked Pentateuch” (see n. 70), 260–261. The paleographic date into the years 125–75 BCE (C rawford , “Temple Scroll” [see n. 180], 15) is due to a typo (email correspondence of Crawford with the author on February 20th, 2006). 188  See Brooke, “Torah” (see n. 172), 113; Tov, “Textual Status” (see n. 172), 45–46; Tov and White, “Reworked Pentateuch” (see n. 71), 261.

188

Armin Lange

well as ninety-eight times nonaligned.189 Among these variant readings are seven macro-variants in which 4QRPc reads once with and six times against MT, never with and seven times against SP, and once with and six times against LXX. The seven macro-variants include, on the one hand, additional text before Exod 15:22 (remnants of an extensive Song of Miriam)190 and after Lev 24:2 (detailed regulations for a wood sacrifice reminiscent of 11QTa XXII–XXIV par 11QTb VI).191 On the other hand, 4QRPc never included the text of Exod 39:6–7 and attests to a divergent textual sequence: Num 4:49 → 7:1; Num 13:15 → 13:13; Num 27:11 → 36:1–2. To these macro-variants the nonbiblical passages of the five fragments of 4Q365a need to be added. Given the agreements with SP, there can be no doubt that 4QRPc attests to a nonaligned text that developed out of a pre-Samaritan text.192 Tov describes the relationship between 4QRPc and SP as follows: 1. In minutiae, 4Q365 is closer to ⅏ than to the other sources. 2. No evidence, positive or negative, has been preserved with regard to harmonizing additions. 3. In the layout of the Song at the Sea, the spacing resembles that of ⅏, and not that of 𝔐. 4. 4Q365 does not reflect two cases of editorial manipulation of ⅏ (sequence: Exod 26:35; 30:1–10; 26, 36 ff. [frg. 8a–b]; transposition of Exod 29:21 after v 28 [frg. 9b II]). 5. 4Q365 often differs from ⅏ in “non-aligned readings” which should be ascribed to 4QRPʼs rewriting of the biblical text. Among these are omissions of 4Q365 of the following verses: Exod 14:18; 39:6–7; Lev 11:19; and the different internal sequence of Num 13:12–16.193

That the text of 4QRPc grew out of a pre-Samaritan text is for Tov also underlined by its agreements with 4QNumb. “4QNumb is closely related to 4Q365, even though no major variants are involved.”194 To conclude: 4QRPc attests to a text that included the whole Torah. This Torah text depends on the pre-Samaritan text of the Torah and develops it further. That the 4QRPc Torah text is a further development of the pre-Samaritan text means that the 4QRPc text of the Torah is not identical to the final stage of the Torah but is rather a relatively late branch in the textual tree of the Torah. That the 4QRPc Torah text reworks a pre-Samaritan text thus pushes the terminus ante quem of the final stage of the Torah back to late Persian times. 189  The above statistics include neither the text of 4QRPc 37 nor the five fragments of 4Q365a. 190  For the Song of Miriam in 4Q365, see A. Feldman, “The Song of Miriam (4Q365 6a ii + 6c 1–7) Revisited,” JBL 132 (2013), 905–911; and H. Tervanotko, Denying Her Voice: The Figure of Miriam in Ancient Jewish Literature (Jyväskyla: Bookwell Oy, 2013), 140–157. 191  See Tov and White, “Reworked Pentateuch” (see n. 70), 293–295. 192  See Tov, “Textual Status” (see n. 172), 65–73; Tov and White, “Reworked Pentateuch” (see n. 70), 194–195. Against Fabry, “Levitikus” (see n. 172), 317 (4QRPc is pre-Samaritan) and A. Y. Kim, “The Textual Alignment of the Tabernacle Section of 4Q365 (Fragments 8a–b, 9a–b i, 9b ii, 12a i, 12b iii),” Textus 21 (2002), 45–69, here 53–68. 193  Tov and White, “Reworked Pentateuch” (see n. 70), 194. 194  Tov and White, “Reworked Pentateuch” (see n. 70), 194.

From Many to One

189

4.2.2.2 The Torah Text of 4QExod-Levf (4Q17) 4QExod-Levf survives in five fragments. Fragments 1–4 preserve 259 words and partial words from Exod 38:18–22; 39:4–7, 9–11, 13–24; 40:8–27; Lev 1:13–15, 17; 2:1. The unusual column height of sixty lines195 makes it likely that 4QExod-Levf once included the whole Pentateuch. The paleographic date is set in the middle of the third century BCE. 4QExod-Levf is thus one of the earliest manuscripts from the Qumran library: “The script is not the formal script of the third century represented by 4QSamb and 4QJera, but belongs to a less formal script tradition which I have called ‘protocursive.’ ”196 The orthography of 4QExod-Levf is conservative but fuller than MT.197 The manuscript reads two times with and twenty-seven times against MT, six times with and twenty-three times against SP, five times with and thirteen times against LXX, as well as twenty-one times non-aligned. “The order of the text of 4QExod-Levf is significant in determining its textual affiliations. Chap. 39:3–24 preserved in this manuscript immediately precedes chap. 40 in agreement with the order of the Massoretic and the Samaritan traditions. On the contrary, in the Septuagint, chap. 39:10–32 is found at 36:10–32 in a radically different ordering of the tradition.”198 Despite these structural agreements with MT and SP and six further readings with SP, 4QExod-Levf cannot be viewed as a pre-Samaritan text.199 The disproportionally high number of non-aligned readings characterizes the text of 4QExod-Levf as non-aligned instead.200 In Exod 39:21, 4QExod-Levf (4Q17) reads: And [they bound the breastplate by its rings to the rings of] the ephod, j[ust] as the Lor[d] commanded Mos[es, with a blue lace, that might not be on the woven band]  See C ross , “17. 4QExod–Levf,” in Qumran Cave 4.VII: Genesis to Numbers (ed. E. Ulrich et al.; DJD XII; Oxford: Clarendon Press, 1994), 134. 196  C ross , “17. 4QExod–Levf” (see n. 195), 134; see also idem , “The Development of Jewish Scripts,” in The Bible and the Ancient Near East: Essays in Honor of William Foxwell Albright (ed. G. E. Wright; Garden City: Doubleday, 1961), 133–202, here 153–158. 197  See Cross, “17. 4QExod–Levf” (see n. 195), 135–136. For a list of orthographic variants, see op.cit. 135. For the orthography of 4QExod–Levf, see also D. N. Freedman, “The Massoretic Text and the Qumran Scrolls: A Study in Orthography,” Textus 2 (1962), 87–102, here 96, 98–100. Against D. N. F reedman and K. M athews , The Paleo-Hebrew Leviticus Scroll: 11 Qpaleo Lev (Winona Lake, IN: Eisenbrauns, 1985), 59–60, who list 4QExod–Levf under the category of “Proto-Samaritan Spelling”; see also Brooke, “Torah” (see n. 172), 103. 198  Cross, “17. 4QExod–Levf” (see n. 195), 135–136. 199  Against C ross , “17. 4QExod–Levf” (see n. 195), 136; Tov , Textual Criticism (see n. 31), 108; idem , “Handschriften” (see n. 31), 1–34, here 18; idem , “Biblical Texts” (see n. 31), 139–166, here 155; Fabry , “Levitikus” (see n. 172), 315; P. F lint , “The Book of Leviticus in the Dead Sea Scrolls,” in The Book of Leviticus: Composition and Reception (ed. R. Rendtorff and R. A. Kugler; VTSup 93; Leiden: Brill, 2003), 323–341, here 325. 200  Cf. Tov, “Synthese” (see n. 199), 20; idem, “Biblical Texts” (see n. 199), 156; Lange, Handschriften (see n. 29), 62. 195

190

Armin Lange

on the inside of the e[pho]d, and that the breastplate might not come loose [from the ephod, as the Lord commanded Moses.] And he made the Urim and [the Thummim as the Lord commanded] Moses.201

This long text shows that 4QExod-Levf attests to an expansionistic reworking of an earlier Pentateuch text. That 4QExod-Levf is not the autograph of this recension is demonstrated, for example, by two haplographies in Exod 40:18.202 To conclude: 4QExod-Levf expands a text that knows already some of the pre-Samaritan variant readings. This means that the base text of 4QExod-Levf is not semi-Masoretic in character but is rather related to the pre-Samaritan text. Because 4QExod-Levf was copied in the middle of the third century BCE, this pre-Samaritan base text must go back at least to the early third century BCE, most likely to late Persian times. This means that both the 4QRPc text and the 4QExod-Levf text suggest a likely date in the late Persian period for the pre-Samaritan Pentateuch. 4.2.3 The Pre-Samaritan Pentateuch, the Old Greek Text and the Semi-/ProtoMasoretic Text of the Torah 4.2.3.1 The pre-Samaritan Text The study of the Samaritan Pentateuch and its relation to LXX and MT has a long history that cannot be discussed here in detail.203 The work of Gesenius was decisive.204 He demonstrated convincingly that, as compared to SP, MT preserves the more original text of the Pentateuch. Careful study has shown that most of the 6000 variants of SP (964 of which read with the LXX against MT) go back to a pre-Samaritan text that is known today from manuscripts found in the Qumran library (4QpaleoExodm [4Q22] and 4QNumb [4Q27]). Only the variant readings of the SP that focus on Mount Gerizim and the Samaritans were inserted into this pre-Samaritan text by a Samaritan reworking. A paleographic comparison between the paleo-Hebrew script of the Samaritans and ancient Jewish pa Translation according to M. A begg et al., The Dead Sea Scrolls Bible (see n. 57), 74–75. The long texts of 4QExod–Levf (4Q17) are printed in italics; the long texts of SP are underlined. For the expansionistic character of the variants in question, see Cross, “17. 4QExod–Levf” (see n. 195), 139. 202  For the two haplographies, see Cross, “17. 4QExod–Levf” (see n. 195), 142. 203  For instructive surveys, see R. T. Anderson and T. Giles, Tradition Kept: The Literature of the Samaritans (Peabody: Hendrickson, 2005), 3–47; and S. S chorch , “Der Pentateuch der Samaritaner: Seine Erforschung und seine Bedeutung für das Verständnis des alttestamentlichen Bibeltextes,” in Die Samaritaner und die Bibel/The Samaritans and the Bible: Historische und literarische Wechselwirkungen zwischen biblischen und samaritanischen Traditionen/Historical and Literary Interactions between Biblical and Samaritan Traditions (ed. J. Frey, U. Schattner-Rieser, and K. Schmid; SJ 70; Berlin: de Gruyter, 2013), 6–29. 204  W. G esenius , De pentateuchi samaritani origine, indole et auctoritate commentatio philologico-critica (Halle: Renger, 1815). 201

From Many to One

191

leo-Hebrew scripts points to the reign of John Hyrcanus I as the time when this Samaritan reworking, that is, the Samaritan Pentateuch, was produced – most probably after John Hyrcanus I destroyed Samaria in 128 BCE.205 A date for SP some time after 128 BCE would in turn imply that the final (semi-Masoretic) stage of the Pentateuch predates the reign of John Hyrcanus I, because SPʼs underlying pre-Samaritan basetext in turn reworks a semi-Masoretic text. Kartveit thinks that the large-scale editorial changes of pre-SP document a “Moses layer” that was added to an earlier version of the Torah, which would predate the mid-third century BCE, as some of these macro-variants would also be attested in 4QExod-Levf.206 The 964 shared readings between LXX and (pre-)SP show that this pre-Samaritan text either developed out of the Hebrew Vorlagen of the LXX or was influenced by them.207 4.2.3.2 The Old Greek Translation of the Torah Not only (pre-)SP but also the Hebrew Vorlagen of the Pentateuch-LXX have been understood as better texts and hence earlier versions of the Pentateuch than MT. Since the work of Frankel,208 it has been a common view in LXX research that each book of the Pentateuch was translated separately into Greek. These translations were produced subsequently by different translators, most probably in the third century BCE.209

205  See W. F. Albright, From the Stone Age to Christianity: Monotheism and the Historical Process (Baltimore: Johns Hopkins University Press, 1946), 336 n. 12; F. M. C ross , “A Report on the Biblical Fragments of Cave Four in Wâdī Qumrân,” BASOR 141 (1956), 9–13, 12 note 5a; idem , Ancient Library (see n. 53), 127–128; idem , “Jewish Scripts” (see n. 196), 133–202, here 189 n. 4; J.-D. P urvis , The Samaritan Pentateuch and the Origin of the Samaritan Sect (HSM 2; Cambridge, MA: Harvard University Press, 1968), 98–118; B. K. Waltke, Prolegomena to the Samaritan Pentateuch (Ph.D. diss., Harvard University, 1965), 225–226. 206  M. Karveit, “The Major Expansions in the Samaritan Pentateuch: The Evidence from the 4Q Texts,” in Proceedings of the Fifth International Congress of the Société dʼÉtudes Samaritaines: Helsinki, August 1–4, 2000: Studies in Memory of Ferdinand Dexinger (ed. H. Shehadeh et al.; Paris: Geuthner, 2005), 117–124; idem , The Origins of the Samaritans (VTSup 128; Leiden: Brill, 2009), 259–312. 207  For the relationship of SP and LXX, see K.-R. Kim, “Studies in the Relationship between the Samaritan Pentateuch and the Septuagint” (Ph.D. diss., Hebrew University of Jerusalem, 1994). Kim records 964 cases in which SP agreed with LXX against MT. Forty-seven of these agreements may have arisen independently. Four hundred ninety-three point to a common ancestor. 208  Frankel, Vorstudien zu der Septuaginta (see n. 142); Z. Frankel, Über den Einfluss der palästinischen Exegese auf die alexandrinische Hermeneutik (Leipzig: Barth, 1851). 209  For the dates of the Old Greek versions of Genesis, Exodus, Leviticus, Numbers, and Deuteronomy, see, e.g., G. Dorival, “Les origins de la Septante: La traduction en grec des cinq livres de la Torah,” in La Bible grecque des Septante: Du judaïsme hellénistique au

192

Armin Lange

The differences from MT in the OG texts of Genesis–Deuteronomy cannot be attributed exclusively to their translators. The Qumran manuscripts 4QLevd (4Q26) and 4QDeutq (4Q44) are not identical to the Hebrew Vorlagen of these Old Greek translations but are so close to them that, at least to a significant extent, their disagreements with MT go back to one or more variant texts of the Pentateuch. The Qumran evidence is confirmed by the studies of Wevers on the OG translations of Genesis–Deuteronomy.210 Lack of space does not allow for a comprehensive discussion of how far the Old Greek translations of Genesis, Exodus, Leviticus, Numbers, and Deuteronomy reflect earlier versions than the Pentateuchʼs semi-Masoretic text. But OG-Exodus provides a good example illustrating the problem. In general, the OG text of Exodus harmonizes its Hebrew Vorlage.211 “But the dominant characteristic of Exod as a translation document is its expansionist character. On the whole Exod expands far more than contracts. Where the Hebrew is abrupt, the Greek tends to smoothen out the text.”212 The expansionist character of OG-Exodus is just one side of the picture, however. The significant differences between MT and LXX in Exodus 35–40 go back to the Hebrew Vorlage of OG-Exodus. In MT, the description of how the tabernacle was built (Exodus 35–40) corresponds with the ordinances for the tabernacle in Exodus 25–31. In the OG text, Exodus 35–40 differs significantly from the ordinances in Exodus 25–31. Compared with the report of the building of the tabernacle in MT-Exodus 35–40, OG-Exodus 35–40 attests to a shorter text with a different sequence. OG-Exodus 35–40 summarizes the building report and attests to a text sequence that is technically more plausible. MT-Exodus 35–40 harmonized the shorter version with the ordinances of Exodus 25–31.213 That the christianisme ancient (ed. M. Harl et al.; Initiations au christianisme ancien; Paris: Cerf and CNRS, 1994), 39–82. 210  J. W. W evers , Text History of the Greek Genesis (MSU 11; Göttingen: Vandenhoeck & Ruprecht, 1974); idem , Text History of the Greek Exodus (MSU 21; Göttingen: Vandenhoeck & Ruprecht, 1992); idem , Text History of the Greek Leviticus (MSU 19; Göttingen: Vandenhoeck & Ruprecht, 1986); idem , Text History of the Greek Numbers (MSU 16; Göttingen: Vandenhoeck & Ruprecht, 1982); idem , Text History of the Greek Deuteronomy (MSU 13; Göttingen: Vandenhoeck & Ruprecht, 1978); idem , Notes on the Greek Text of Genesis (SBLSCS 44; Atlanta, Ga.: Scholars Press, 1990); idem, Notes on the Greek Text of Exodus (SBLSCS 44; Atlanta, Ga.: Scholars Press, 1993); idem, Notes on the Greek Text of Leviticus (SBLSCS 44; Atlanta, Ga.: Scholars Press, 1997); idem, Notes on the Greek Text of Numbers (SBLSCS 44; Atlanta, Ga.: Scholars Press, 1998); idem, Notes on the Greek Text of Deuteronomy (SBLSCS 44; Atlanta, Ga.: Scholars Press, 1995). 211  See J. W. Wevers, “Translation and Canonicity” (see n. 159), 295–303. 212  Wevers, Text History of Greek Exodus (see n. 160), 148. 213  See A. Aejmelaeus, “Septuagintal Translation Techniques: A Solution to the Problem of the Tabernacle Account,” in Septuagint, Scrolls and Cognate Writings: Papers Presented to the International Symposium on the Septuagint and Its Relations to the Dead Sea Scrolls and Other Writings (ed. G. J. Brooke and B. Lindars; SBLSCS 33; Atlanta, Ga.: Scholars Press, 1992), 381–402; Kim, “Textual Alignment” (see n. 192), 45–69.

From Many to One

193

text of OG-Exodus 35–40 goes back to a Hebrew Vorlage is also demonstrated by 4QExod-Levf (4Q17) because this manuscript reads with LXX, against MT, in Exod 40:17, 20, 22.214 To summarize so far: SP and pre-SP are both reworkings of a Pentateuch text that is similar to MTʼs but shares significant readings with its Old Greek translations as well. The Old Greek translations of the books of Genesis–Deuteronomy go back to Hebrew Vorlagen that sometimes differ in small details from MT and in rare cases attest to a significantly different text that may have preceded the semi-Masoretic text of the Pentateuch. 4.2.4 The Text of 4QDeutq and the Consonantal Text of MT Of 4QDeutq, five fragments survive, which preserve fifty-six words and partial words from Deut 32:9–10, x, 37–43. Frgs. 2–5 derive from two columns of the original scroll. Fragment 2 is followed by a small handlesheet. 4QDeutq therefore ended with Deut 32:43. That 4QDeutq ended before the final chapter of Deuteronomy as well as the small height of eleven lines per column, or 11.4 cm, identifies 4QDeutq as an excerpted manuscript that probably included only the Song of Moses from Deut 32:1–43.215 The manuscript was copied in an early Herodian bookhand from the second half of the first century BCE.216 The orthography of 4QDeutq is conservative. The manuscript reads once with and twelve times against MT, two times with and eleven times against SP, seven times with and six times against LXX, as well as two times nonaligned. Skehan and Ulrich describe the textual affiliation of 4QDeutq as follows (for the text of 4QDeutq, see also above 2.2): 4QDeutq and the Massoretic textus receptus display distinctly variant forms of the text – more than one variant for every pair of the scrollʼs short lines. 4QDeutq, or its Vorlage, however, should not be naïvely dismissed as a so-called “vulgar text” for a number of reasons. Virtually all of its readings are documented in other biblical manuscripts; some readings (‫ שמים‬32:43) are more ancient than those preserved in 𝔐, which revised polytheistic terms secondarily for theological purposes; and other readings (‫ אדמת‬32:43) appear superior to unusual forms in 𝔐. Though not identical to 𝔊, 4QDeutq shares several unique readings with the Septuagint version of Deuteronomy and bears witness to the existence of the variant Hebrew Vorlage used by the Septuagint

214  I did not have access to the important PhD thesis of Brandon Bruning: B. Bruning, “The Making of the Mishkan” (Ph.D. diss., University of Notre Dame, 2014). 215  See H. Stegemann, “Weitere Stücke von 4Qp Psalm 37, von 4QPatriarchal Blessings und Hinweis auf eine unedierte Handschrift von Höhle 4Q mit Exzerpten aus dem Deuteronomium,” RevQ 6 (1967–1969), 193–227, here 220; Brooke, “Torah” (see n. 172), 108–109; Skehan and Ulrich, “44. 4QDeutq” (see n. 53), 137–138; Duncan, “Excerpted Texts” (see n. 56), 44–45. 216  Cf. S kehan and U lrich , “44. 4QDeutq” (see n. 53), 138. Against W. F. A lbright , “Some Remarks” (see n. 53), 338–339 (“from about the Christian era or a little earlier”).

194

Armin Lange

translator, at least for this section of Deuteronomy [. . .] 4QDeutq and 𝔊 agree in seven readings against 𝔐, including all of the significant readings.217

Although little text is preserved of 4QDeutq, the unequivocal evidence described by Skehan and Ulrich argues that the text attested by 4QDeutq was close to the Hebrew Vorlage of LXX-Deut but had nonaligned tendencies as well. 4QDeutq either attests to the Hebrew Vorlage of OG-Deut or, more likely, to a shared ancestor of both the Hebrew Vorlage of OG-Deut and a semi-Masoretic text. In either case, 4QDeutq attests to a text of the Pentateuch that was later reworked by a semi- or proto-Masoretic text. Above, I have discussed the secondary interpretative readings of MT-Deut 32:8, 43 in detail (2.2), which underline this point. It is therefore likely that the excerpted text of 4QDeutq preserves all that is left of a text of Deuteronomy, and most probably the whole Torah, that precedes the consonantal text of MT. The 4QDeutq text of the Torah moves the terminus ante quem of the final stage of the Torah even further into Persian times. 4.2.5 The Date of the Final Redactional Stage of the Torah To come back to the question of the terminus ante quem of the Pentateuchʼs final stage: the third century BCE dates of the Temple Scroll and the Book of Ezra– Nehemiah show that the 4QRPc text of the Torah, as one of its latest recensions, is already employed by two literary works from the third century BCE. That both the Temple Scroll and the Book of Ezra–Nehemiah use the 4QRPc text of the Torah independently of each other points to the high degree of authority that the 4QRPc text of the Torah must have acquired in the third century BCE. Because such an acquisition of authority normally requires a few decades, the 4QRPc text of the Pentateuch should be dated no later than the early third century BCE, most likely to late Persian times. But the 4QRPc text of the Pentateuch is only a reworking of a pre-Samaritan text, which in one or more stages218 rewrites and expands an early proto-Masoretic text, which is based on a text with close ties to the Hebrew Vorlagen of the Old Greek translations. 4QDeutq is a Qumran manuscript that is at least relatively close, if not identical, to such a Vorlage. 4QDeutq, furthermore, points to the existence of a Pentateuch text that precedes both the Vorlagen of the OG translations of the books of the Pentateuch and its semi-Masoretic text and was theologically reworked by the latter text. All of the above means that at least two or three reworkings took place between the final stage of the Torah and its 4QRPc text. These two or three 217  Skehan and Ulrich, “44. 4QDeutq” (see n. 53), 138; see also Skehan, “Fragment” (see n. 53), 12, 14–15; idem, “Qumran Manuscripts” (see n. 53), 149–150 and 150 n. 1; Cross, The Ancient Library of Qumrân (see n. 53), 135–136; Lange, Handschriften (see n. 29), 99–100. 218  That the pre-Samaritan text of the Pentateuch evolved in several stages may be indicated by the few pre-Samaritan readings of 4QExod–Levf (4Q17).

From Many to One

195

reworkings exclude a date for the final stage of the Torah in the third century BCE and point to the fourth century BCE instead. It is even likely that the final redactional stage of the Torah should be dated closer to the beginning of the fourth century BCE than its end, if not earlier. Relative chronology thus shows that the text of the Torah as we presuppose it for its final stage goes back to late Persian times, if not earlier. A date of the final stage of the Torah in the fourth century BCE or earlier is corroborated by 4QExod-Levf. 4QExod-Levf expands a Pentateuch text that includes some of the pre-Samaritan variant readings. The base text of 4QExod-Levf is hence not semi-Masoretic in character but is rather a reworked semi-Masoretic text. Because 4QExod-Levf was copied in the middle of the third century BCE, the semi-Masoretic base text of its recension must go back at least to the early third century BCE but is most probably significantly earlier. Relative chronology thus proves that the terminus ante quem for the final stage of the Pentateuch is in the fourth century BCE under Persian rule and even raises the possibility that it might go back to the early fourth or even late fifth century BCE. The final redactional stage of the Torah as determined by relative chronology thus must be set relatively close to the missions of Nehemiah and Ezra. It is therefore tempting to speculate whether the final redaction of the Torah was connected to either one of these missions. However, the answer to this question goes beyond the present study.

Part Two Can the Pentateuch Be Read in Its Present Form? Narrative Continuity in the Pentateuch in Comparative Perspective

Introduction Jeffrey Stackert The question of narrative continuity in the Pentateuch – namely, the sequence of events narrated and the consistency (or inconsistency) observable in this narration – already occupied readers in antiquity and continued to do so throughout the premodern period. The same concern has animated modern critical study of the Hebrew Bible, from Spinoza to the present. Though containing a large number of subgenres, the Pentateuch is readily characterized as a narrative history featuring a chronological plot progression. Yet its narrative discontinuities and other internal inconsistencies are also immediately evident. These issues, as well as their many implications, are the concern of the essays in this part. Analysis of narrative continuity, particularly in relation to ancient texts like the Pentateuch, requires a number of specific considerations. At the most basic level, it must be asked, what counts as continuity in pentateuchal narrative? And even more basically, what do we mean by continuity? The locus of inquiry may likewise be directed toward the textʼs reader: How much continuity is necessary for the reader to achieve coherence? And what other factors, either within the text or outside of it, contribute to the construal of satisfactory meaning in the reading process? The essays in this part address these questions both directly and indirectly. In so doing, they interrogate the pentateuchal text as well as the assumptions, methods, and arguments regularly employed in the field of biblical studies. A number of considerations inform their conclusions. Repeatedly cited are ancient literary conventions, observable within the Hebrew Bible and in other texts and collections. Such evidence promises insight into both the pentateuchal texts as narratives and the attendant issues of their compositional and redactional strategies, scribalism, and materiality. Also referenced are culturally and temporally disparate evidence and theory, regarding both literary genres and texts and reading strategies. These considerations help to clarify the pentateuchal narratives themselves and shed light on how they have been and continue to be interpreted. There are a number of challenges that attend discussions of narrative continuity in the Pentateuch. One such challenge, hinted at already, is the varied nomenclature that scholars employ and the claims their categories imply. For example, is narrative continuity to be equated with literary unity? Is the former the basis for the latter? Can a narrative text that does not exhibit robust continuity still be considered unified, cohesive, coherent, fluent, or readable? The

200

Jeffrey Stackert

essays in this part employ such diverse terminology, which both enriches and complicates the discussion. A second and related challenge is delimiting the text to be analyzed, which in the case of the Pentateuch oftentimes depends on the relative importance that the interpreter affords the textʼs various narrative features, including plot, characterization, thematics, theological perspective, and style. Traditional textual demarcations, including the individual pentateuchal scrolls or “books” (Genesis, Exodus, Leviticus, Numbers, Deuteronomy) and the Pentateuch as a whole, likewise play a role, at least for some. Thus, even though all interpreters begin with the canonical text, they quickly part company after beginning their analyses. For example, for those for whom narrative continuity is the defining characteristic of literary unity (or coherence or readability), there is little reason to attempt to make sense of even an individual pentateuchal scroll, to say nothing of the Pentateuch as a whole; such a synchronic reading, in the face of narrative discontinuities, is deemed impossible. Yet others, while acknowledging the plot discrepancies and inconsistencies in the Pentateuch, seek an alternative unity (or coherence or readability) for it, either in part or in whole. Such an approach makes possible analyses of individual story cycles (e.g., the Jacob cycle or the Joseph cycle), pentateuchal scrolls, and the entire Pentateuch. As this partʼs essays make clear, these divergent approaches significantly impact reconstructions of the Pentateuchʼs compositional history. They also have potential implications for the ongoing reception of the Pentateuch, not least because such literary theorizations may readily serve as frameworks for future interpretations.

What Do We Mean by Plot and by Narrative Continuity? Jean Louis Ska 1 The Problem of Narrative Continuity according to Joel Baden Such is only to be expected: the identifying feature of a narrative is not its theme, but its plot, and it is from the plot and the way in which the coherent story is told that the theme emerges, secondarily. As with any literary work, the message and structure of each document are apparent only once the document has been read in its entirety. And in order to understand the themes of any of the sources, the sources must first be separated on the more objective grounds of narrative continuity.1

This quotation from Joel Badenʼs The Composition of the Pentateuch contains a certain number of challenging statements that must be taken seriously. The main problem is that of narrative continuity, more technically the problem of “plot.” As we will see, the word plot must be defined in more precise way. Badenʼs affirmation is based on another observation, namely, that the Pentateuch as such is not “readable.” It becomes understandable only when the sources are separated: “The analysis of the Pentateuch is grounded in the basic inability to read the text as a whole, and that inability is not manifested in the variety of themes or style.”2 There is coherency only in the different sources, once they are separated, but not in the whole. Hence, Baden argues for the necessity of distinguishing and reconstructing the different sources of the Pentateuch, since only the individual sources make sense. To put it in other words, there is no clear “narrative continuity” in the Pentateuch as a whole, and it is not possible to identify a clear plot in the first five books of the Bible. Plot is a characteristic of the separate sources, not of their compilation. This position raises a certain number of questions. First of all, the Pentateuch has been read in its present form over many centuries by rabbis, fathers of the church, and hundreds of medieval and Renaissance  J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven, CT: Yale University Press, 2012), 28–29. The title of his book surely recalls the well-known work of J. W ellhausen , Die Composition des Hexateuchs und der historischen Bücher des Alten Testament (Berlin: Reimer, 1866) and the two major works of E. Blum, Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), and idem, Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990). 2  Baden, Composition (see n. 1), 30. 1

202

Jean Louis Ska

scholars who did not notice this difficulty. For this reason, we must put Badenʼs thesis to the test in detail. This means that we have in the first place to define more precisely what plot and narrative continuity are. Second, we have to submit the Pentateuch as a whole to closer scrutiny to verify whether this definition applies to it and in what way. Accordingly, my first contention in the present article is that the sources of the Pentateuch have no real narrative continuity, but the so-called Priestly writer does, if we stick to a clear and workable definition of narrative plot. My second contention is that a literary study of the Pentateuch leads to the identification of important breaks between narratives complexes, especially between Genesis and Exodus but also between the different parts of Genesis (the story of origins, the Abraham cycle, the Jacob cycle, and the Joseph story).

2 A Definition of Plot Let us start with a basic question: What is narrative and what is plot? According to a study on the nature of narrative published some years ago by R. Scholes and R. Kellogg, “Plot can be defined as the dynamic, sequential element in narrative literature.”3 Elements in a narrative are disposed, first of all, along chronological lines. This element is essential and is mentioned in all definitions of the plot, but this is not sufficient to create a real plot. What is required for a real narrative is a logical sequence, i.e., a chain of causes and effects that is present in the succession of events recounted. The logical connection between the events is sometimes implied or simply presupposed; it is indispensable in the creation of a real plot. This difference between chronological and logical sequence in narratives is expressed by the distinction between story and plot proposed by some modern theories of narratology. This distinction, however, goes back to Aristotle, who in his Poetics points out the difference between events happening in real life (πρᾶξις) and events selected and arranged in a certain order in what he calls μῦθος, the equivalent of plot. In a similar way, the English novelist and literary critic E. M. Forster defined plot as a cause-and-effect relationship between events recounted in a story.4 A story, therefore, is a mere sequence of events, such as, to take Forsterʼs example, “The king died, and then the queen died.” A plot introduces an explicit and logical connection between the two events, as, for instance, “The king died, and then the queen died of grief.”5 3  R. Scholes and R. Kellogg, The Nature of Narrative (Oxford: Oxford University Press, 1966), 207. 4  E. M. Forster, Aspects of the Novel (London: Edward Arnold, 1927), 93–94. 5  G. Prince, A Dictionary of Narratology (2nd ed.; Lincoln: University of Nebraska Press, 2003), 73.

What Do We Mean by Plot and by Narrative Continuity?

203

Or, to say it with Forster, “[A story] is a narrative of events arranged in their time sequence – dinner coming after breakfast, Tuesday after Monday, decay after death, and so on.”6 As for the plot, Forster proposes this definition: “A plot is also a narrative of events, the emphasis falling on causality.”7 In a story, the readerʼs question is “And then?” whereas in a plot, the question is “Why?”8 We find an example of the difference between story and plot when comparing the sequence in Gen 22 and 23 in the Bible and in Genesis Rabbah. In the book of Genesis we have a simple sequence, a story in Forsterʼs vocabulary, since the death of Sarah (Gen 23) follows immediately after the testing of Abraham, the Akedah (Gen 22). In the midrash, this chronological succession becomes a plot because it is said that Abraham, when coming back home after his testing on Mount Moriah, had to tell Sarah about what happened, and the latter, hearing that she had nearly lost her son, cried six times and died on the spot (Gen. Rab. 58:5). In the midrashic retelling, Sarah dies in Gen 23 because Abraham was asked to sacrifice Isaac in Gen 22. Chronological and logical elements are combined in innumerable ways, obviously, and the logical link may remain implicit in many cases. This is well summarized by the old Latin dictum post hoc, ergo propter hoc, “after this, therefore because of this.” We may find, in a narrative tradition, simple forms of chronological sequences, i.e., stories, and more sophisticated forms in which the logical connection is made explicit, i.e., plots. It is surely possible to classify different forms of narrative on the basis of this simple distinction. This is what we will endeavor to do now, before applying these categories to the Pentateuch. Our purpose is to show that there are many types of narrative and that it would be opportune to introduce some distinctions into the discussion. As we will see, it will not be possible to apply one and the same concept of plot or narrative continuity to the entire Pentateuch. This means that it will be indispensible to take some basic differences into consideration before speaking of continuous sources in the first five books of the Bible.

3 Simple Forms of Narrative in Ancient Literature Simpler forms of narrative are built on a mere chronological sequence where the logical element is absent or almost completely absent. This type of narrative generally finds its unity in an external element, for instance the setting, one or more characters, or a chronological framework for such events that took place in a given period of time. In these cases, we have stories but no plots.  Forster, Aspects of the Novel (see n. 4), 27.  Forster, Aspects of the Novel (see n. 4), 27. 8  Forster, Aspects of the Novel (see n. 4), 27; see also H. Abrams, A Glossary of Literary Terms (5th ed.; Fort Worth, TX: Holt, Rinehart & Winston, 1988), 139. 6 7

204

Jean Louis Ska

3.1 Chronicles, Genealogies, and Itineraries9 The simplest forms of narrative are chronicles of events.10 Thus the word chronicle contains the same Greek root, chronos (χρόνος), “time,” as the word chronology. Chronicles, as we know, are series of reports or short reports of events classified in chronological order and covering a given, often considerable, period of time. The same holds true for the annals, records of events classified by years. A chronicle of this type is most probably to be found in the ‫ספר דברי‬ ‫הימים‬, “the book of the days [of King so-and-so]” (1 Kgs 14:29, etc.). Let us note, in this biblical expression, the presence of the chronological term day (‫)יום‬. Chronicles are well-known in the ancient Near East, as for instance in royal inscriptions and reports of military campaigns. In the same category where chronology is omnipresent and logic almost completely absent, we can classify some elementary forms of narrative such as genealogies and itineraries. In genealogies, such as the Sumerian King List or Hesiodʼs Theogony and Catalogue of Women, the main principle of structuring is a chronological succession. We also have several examples of genealogies in the Bible, as in the book of Genesis and in the books of Chronicles.11 Genealogies as such, however, are not yet narratives and belong rather to the literary genre of lists. They endeavor to structure time according to ideological principles and often have little to do with what we usually call historical facts. Ideas and meanings, however, do not appear (yet) in logical links connecting the different elements listed.12 As one might expect, genealogies can be fleshed out with anecdotes or longer narratives about gods and goddesses, kings and queens, and heroes and heroines who appear in the list. Hesiodʼs Theogony and his Catalogue of Women use a similar technique in certain cases, using some names as catchwords to add a few anecdotes or even longer narratives, as in the case of the flood.13 This is also the  About genealogies in general, see, among others, M. D. J ohnson , The Purpose of the Biblical Genealogies with Special Reference to the Setting of the Genealogies of Jesus (SNTSMS 8; 2nd ed.; Cambridge: Cambridge University Press, 1988), 3–36; R. R. Wilson, Genealogy and History in the Biblical World (YNER 7; New Haven, CT: Yale University Press, 1977); R. B. Robinson, “Literary Functions of the Genealogies of Genesis,” CBQ 48 (1986), 595–608. 10  See Abrams, Glossary (see n. 8), 26. 11  For several authors, narratives grew out of genealogies. See W ellhausen , Die Composition (see n. 1), 8–9; K. B udde , Die biblische Urgeschichte: Gen. 1–12,5 (Gießen: Ricker, 1883), 184–186; C. Westermann, Genesis 1–11 (BK I/1; 3rd ed.; Neukirchen-Vluyn: Neukirchener Verlag, 1983), 388–389. The same Westermann, however, also states the opposite, namely, that genealogies are summaries of more developed narratives (Genesis 1–11, 14). 12  See, for instance, P. M ichalowski , “History as Charter: Some Observations on the Sumerian King List,” JAOS 103 (1983), 237–248. 13  Hesiod, Catalogue of Women, frags. 2–7 and 234. On this topic, see, for instance, R. S. Hess, “The Genealogies of Genesis 1–11 and Comparative Literature,” Bib 70 (1989), 9

What Do We Mean by Plot and by Narrative Continuity?

205

case in Gen 10:8–10, with Nimrod, or in Gen 6–9, the story of the flood attached to the genealogy of Noah. In his analysis of Genesis, Claus Westermann distinguishes two types of genealogy, namely, aufzählende and erzählende Genealogien.14 The former refers to simple genealogies that “list” names, whereas the latter refers to “narrating genealogies” entailing a series of narratives about one person. The latter type is present above all in the narratives about Abraham, Isaac, Jacob, and Joseph and his brothers (Gen 12–50). A second category of simple narrative is the journey, with the record of different places visited during the journey.15 As with the genealogy, a simple list of stages can be more fully developed. Incidents can also be related more than once; there can be different versions of the same incident, and there can be several competing versions of the itinerary compiled in the same record. In the Bible, we have the wanderings of the people of Israel in the wilderness.16 A summary of this itinerary is to be found in Num 33, a kind of skeleton, since we have only a list of stages, with very few exceptions (see Num 33:3, 8, 38–39, 40). In itineraries, the most important element is still chronology. There is sometimes, but not necessarily, a logical link between the different episodes. The main difference between genealogies and chronicles on the one hand and itineraries on the other is the presence, in itineraries, of a spatial element besides the chronological one. The unity of the narratives comes from a chronological framework (the succession of time) and a spatial framework (the journey by stages, from place to place, from departure until arrival). In the Bible, the best example is found, as already mentioned, in the wanderings of the people of Israel in the wilderness. This long itinerary or series of wanderings begins in Exod 12:37–38 and finishes in front of Jericho, on the banks of the Jordan, in the land of Moab (Num 22:1; 33:50; 36:13; Deut 1:5; 34:1). The whole book of Deuteronomy is integrated into this framework, since Deut 1:1–5 situates Mosesʼs farewell discourses in the plains of Moab, on the banks of the Jordan, in front of Jericho, at the end of the forty years of Israelʼs wanderings in the wilderness. 241–254, esp. 251–254; J. Van S eters , Prologue to History: The Yahwist as Historian in Genesis (Louisville, KY: Westminster John Knox, 1992), 89–90, 176–177. 14  Westermann, Genesis 1–11 (see n. 11), 8 and 24; idem, Theologie des Alten Testament in Grundzügen (GAT 6; Göttingen: Vandenhoeck & Ruprecht, 2011), 75. 15  On this topic, see further, within this volume, Y. A mit , “Travel Narratives and the Message of Genesis.” 16  On the wanderings of Israel in the desert, see G. I. Davies, The Way of the Wilderness: A Geographical Study of the Wilderness Itineraries in the Old Testament (SOTSMS 5; Cambridge: Cambridge University Press, 1979); idem , “The Wilderness Itineraries and Recent Archaeological Research,” in Studies in the Pentateuch (ed. J. A. Emerton; VTSup 41; Leiden: Brill, 1990), 161–175; A. R. Roskop, The Wilderness Itineraries: Genre, Geography, and the Growth of Torah (HACL 3; Winona Lake, IN: Eisenbrauns, 2011).

206

Jean Louis Ska

In ancient Egypt, we can mention the travels of Sinuhe and those of WenAmon; in Mesopotamia, the journey of Gilgamesh in quest of immortality.17 In classical literature, we have the Odyssey in Greece and the Aeneid in Rome. Chronicles of military campaigns can be classified in this category as well.18 In this kind of narrative, it is possible to skip some stages because not all of them are interesting or necessary to the understanding of the whole. Not all incidents in the wilderness are as decisive for the fate of Israel and for the journey itself as are Num 13–14, for instance. The long sojourn of Israel at Mount Sinai/Horeb is definitely the longest and the most important stage in this journey through the wilderness (Exod 19–Num 10). 3.2 Thematic Unity: The Deeds of a Hero, the Collection of Tales We have a more unified type of narrative when a series of stories has at least one common element, for instance, the person of a hero. Sometimes, and this is obvious, the series begins with the birth of the hero and finishes with his or her death. The best example in the Bible is Samson, with his birth in Judg 13 and his death in Judg 16.19 In other cases, we have a collection of stories about an individual or a group, but the order of stories is indifferent. There is no strict chronology and there is no visible development from one story to the other. Stories may be read in a different order without difficulty. The only real link between the narratives is the people involved.20 Examples in the Bible include Daniel and his companions or the miracle stories about Elisha.21 A similar case is found in the English epic Beowulf, which tells of the three main battles fought by the hero, Beowulf, who dies in the third one. 22 The

17  See, for instance, M. B árta , Sinuhe, the Bible, and the Patriarchs (Prague: Set Out, 2003); W. Sallaberger, Das Gilgamesch-Epos: Mythos, Werk und Tradition (Munich: Beck, 2008). 18  J. Van Seters, “Joshuaʼs Campaign of Canaan and Near Eastern Historiography,” SJOT 4 (1990), 1–12, reprinted in idem, Changing Perspectives 1: Studies in the History, Literature, and Religion of Biblical Israel (Copenhagen International Seminar; London: Equinox, 2011), 73–83. 19  See, on Samson, the collective work E. Eynikel and T. Nicklas (eds.), Samson, Hero or Fool? The Many Faces of Samson (Themes in Biblical Narrative: Jewish and Christian Traditions 17; Leiden: Brill, 2014). See also the still very valuable work by H. G unkel , “Simson,” in Reden und Aufsätze (Göttingen: Vandenhoeck & Ruprecht, 1913), 38–64. 20  See T. L. H olm , Of Courtiers and Kings: The Biblical Daniel Narratives and Ancient Story-Collections (Explorations in Ancient Near Eastern Civilizations 1; Winona Lake, IN: Eisenbrauns, 2013). 21  For a recent study on Elisha, see K. Bodner, Elishaʼs Profile in the Book of Kings: The Double Agent (Oxford: Oxford University Press, 2013). 22  Scholes and Kellogg, Nature (see n. 3), 34–37.

What Do We Mean by Plot and by Narrative Continuity?

207

German Nibelungenlied, about the Burgundian dynasty, can be classified in this category as well.23 The Gilgamesh Epic is more unified, since the deeds of the hero are about his friendship with Enkidu and, after Enkiduʼs death, the quest for immortality. I am speaking, of course, of the full-fledged version of the epic. This means that the Gilgamesh Epic represents a more sophisticated type of narrative. We do not find a mere collection of great deeds, as was probably the case in earlier versions of the epic. In its Babylonian version, the complete narrative develops around one thread, namely, the friendship between Gilgamesh and Enkidu, the end of this friendship, caused by Enkiduʼs death, and the attempt to restore it when Gilgamesh endeavors to find the secret of immortality.24 3.3 Narratives within a Frame Narrative Another type of collection of tales, which are less useful for our understanding of the Pentateuch, are those collections of tales attached to a narrator or a series of narrators and not to a character in the story. The most famous examples are The Tales of a Thousand and One Nights or The Arabian Nights, The Decameron by Giovanni Boccaccio, and The Canterbury Tales by Geoffrey Chaucer. In the case of Boccaccio and Chaucer (who most probably depends on Boccaccio), we know the author. In Boccaccioʼs Decameron, all the stories are told by ten young people, who decide that each should tell a story during ten days spent outside Florence during the Black Death in 1348–1349. Chaucerʼs work perhaps remains unfinished but is in a way similar to The Decameron: a certain number of pilgrims decide to tell two stories on their way toward and two more on their way back from Canterbury. We do not know the author of The Arabian Nights, but the frame narrative is well-known (Sheherazade tells stories to Shahryār “in order to survive”), and the same cultural milieu gives unity to this anthology. The stories themselves, as in the case of Boccaccio and Chaucer, may have very different origins. In all these cases, we have a series of stories that are only chronologically connected. Every story generally has its own action, its own setting, and its own set of characters. There are only a few exceptions to this rule. We have many stories but no plot unifying them, just an external narrative framework. To say it again with Forster, “[Sheherazade] only survived because 23  Scholes and Kellogg, Nature (see n. 3), 45, 58. For a modern edition, see A. Behmel, Das Nibelungenlied (2nd ed.; Stuttgart: Ibidem Verlag, 2001). 24  See J. Tigay, The Evolution of the Gilgamesh Epic (Philadelphia: University of Pennsylvania Press, 1982), 39–54; for a more recent treatment of the question, see A. R. George, The Babylonian Gilgamesh Epic: Introduction, Critical Edition and Cuneiform Texts (2 vols.; Oxford: Oxford University Press, 2003), 1: 17–24, esp. 20: “The fashioning around the character of Gilgameš of a majestic epic poem, with its great, uniting themes of kingship and power, wilderness and civilization, friendship and love, victory and arrogance, death and life, man and god – this was a Babylonian achievement.”

208

Jean Louis Ska

she managed to keep the king wondering what would happen next.”25 Neither the king Shahryār nor the reader asks why each story is told after the previous one. The same holds true for Boccaccio and Chaucer. We do actually know of some collections of tales and legends compiled by the same person(s), such as Les Contes de ma mère lʼOye or Histoires et contes du temps passé (1697) by Charles Perrault (1628–1703) in France,26 the famous collections of fairy tales by the brothers Grimm in Germany,27 or those collected by Hans Christian Andersen (1805–1875) in Denmark.28 Among more recent publications of the same kind, let us mention the many collections of tales and legends from different parts of the world.29 Several exegetes have noticed the similarity between Gunkelʼs analysis of the book of Genesis and this kind of research.30 Obviously, the aforementioned compilers did not create or compose the tales that they gathered, although it is evident that they reworked and rewrote the materials collected. Charles Perrault and the brothers Grimm are not the authors of the stories they published. But without them, these stories would not exist in the forms in which we know them, since the compilers played an active role and are responsible for the talesʼ style, organization, and literary shape.31 In Gunkelʼs eyes, the ancient writers who compiled the stories now present in the book of Genesis did a work similar to that of the brothers Grimm. Hence the famous and often-quoted sentence, “Genesis  Forster, Aspects of the Novel (see n. 4), 27.  C. P errault , Histoires et contes du temps passé (Paris: Barbin, 1697). For a modern edition, see idem, Contes (ed. Jean-Pierre Collinet; Paris: Gallimard, 1981). 27  J. G rimm and W. G rimm , Kinder- und Hausmärchen (2 vols.; Berlin: Realschulbuchhandlung, 1812–1815). 28  For a modern edition, see H. C. A ndersen , The Complete Illustrated Works of Hans Christian Andersen (ed. L. Owens; Vacaville, CA: Bounty Books, 1994). 29  See, for instance, L. G inzberg , The Legends of the Jews (7 vols.; Philadelphia, PA: Jewish Publication Society of America, 1909–1938); new edition by D. M. Stern (Philadelphia, PA: Jewish Publication Society of America, 2003). Ginzberg, as many other modern compilers, collected mostly written sources. This is not always the case, however, as for instance with I. C alvino , Fiabe italiane (2 vols.; Turin: Einaudi, 1956 [2nd ed., 1971; 3rd ed., Milan: Mondadori, 1996]). 30  See, for different opinions on the topic, W. K latt , H. Gunkel: Zu seiner Theologie der Religionsgeschichte und zur Entstehung der formgeschichtliche Methode (FRLANT 100; Göttingen: Vandenhoeck & Ruprecht, 1969), 106–110; P. G ibert , Une théorie de la légende: Hermann Gunkel (1862–1932) et les légendes de la Bible (Paris: Flammarion, 1979), 43–55; P. G. K irkpatrick , The Old Testament and Folklore Study (JSOTSup 62; Sheffield: Sheffield Academic, 1988), 23–38, 73–79; S. Niditch, Oral World and Written Word: Ancient Israelite Literature (Library of Ancient Israel; Louisville, KY: Westminster John Knox, 1996), 2; D. Dieckmann, Segen für Isaak: Eine rezeptionsästhetische Auslegung von Gen 26 und Kotexten (BZAW 329; Berlin: de Gruyter, 2003), 37–38; K. Hammann, Hermann Gunkel: Eine Biographie (Tübingen: Mohr Siebeck, 2014), 230. 31  See J. Zipes, The Brothers Grimm: From Enchanted Forests to the Modern World (2nd ed.; New York: Palgrave Macmillan, 2002). 25

26

What Do We Mean by Plot and by Narrative Continuity?

209

ist eine Sammlung von Sagen.”32 The authors of Genesis put into writing hoary legends about the origin of their nation.33 The analogy can be queried, evidently, but we cannot deny that Gunkel was influenced by the Germanistik of his time.

4 Cycles of Stories in the Book of Genesis In the Pentateuch, we may find some elements of this latter kind of narrative, namely, cycles of stories around one protagonist, in the patriarchal narrative, especially in the Abraham and the Jacob cycles, as also in the beginning of Mosesʼs life in Exod 2. In the Abraham cycle, one thread unifies a good number of narratives, namely, the hope for an heir. The single narratives are all well rounded, but each one describes either an attempt to find an heir or an episode in which an heir is promised.34 A few narratives are not integrated into this framework, for instance, the story of the war between kings in Gen 14, the end of Sodom and Gomorrah in Gen 19, or Gen 21:22–24, the treaty with Abimelek. After the birth of Isaac in Gen 21, the plot is much less unified. We have disconnected incidents such as the testing of Abraham (Gen 22), Sarahʼs death and burial (Gen 23), and Isaacʼs betrothal and marriage to Rebekah (Gen 24). The Jacob cycle is much more unified, since most of its components fall into one of two categories, namely, episodes of a conflict with Esau (Gen 25, 27, and 33) and episodes of a conflict with Laban (Gen 28–32).35 The first conflict is also the cause of the second, since Jacob must leave his country because of Esauʼs enmity (see Gen 27:41–44). Again, as in the Abraham cycle, the thread of the story becomes much looser after Jacobʼs reconciliation with Esau in Gen 33:1–20. Genesis 34 and 35 are a mosaic of disconnected pieces of narrative: the rape of Dinah and the revenge taken by her brothers (Gen 34), Jacobʼs return to Bethel (Gen 35:1–15), the birth of Benjamin and Rachelʼs death (35:16–20), some anecdotes about Jacob, his sons, and Isaacʼs death (35:21–29). As in many biblical and nonbiblical narrative cycles, we usually find later additions loosely  H. Gunkel, Genesis (3rd ed.; Göttingen: Vandenhoeck & Ruprecht, 1910), vii.  Niditch, Oral World (see n. 30), 2. 34  See D. Sutherland, “The Organization of the Abraham Promise Narratives,” ZAW 95 (1983), 337–384; L. R. Helyer, “The Separation of Abram and Lot,” JSOT 26 (1983), 77–88; J. P. Fokkelman, “Time and Structure of the Abraham Cycle,” in New Avenues in the Study of the Old Testament: Festschrift M. J. Mulder (ed. A. S. van der Woude; OTS 25; Leiden: Brill, 1989), 96–109. This sentence from the latterʼs article is worth quoting: “Indeed, the life of Abraham (as rendered in Gen. xii-xx) is dominated by a faithful, trusting and despairing wait, for how and when God will keep his promise that Abraham shall have progeny, counter to the obstacle of Sarahʼs infertility” (96). 35  See J. P. F okkelman , Narrative Art in Genesis: Specimens of Stylistic and Structural Analysis (SSN 17; Assen: Van Gorcum, 1975); M. Fishbane, Text and Texture: Close Readings of Selected Biblical Texts (New York: Schocken, 1979). 32 33

210

Jean Louis Ska

associated with the mainstream narrative at the end. The principle is, “there is still something to tell about X or Y.”36 In other words, stories are appended to the plot. The Joseph story is the most unified narrative in the book of Genesis.37 In this case we have the different stages of only one conflict, caused by Jacobʼs favoritism, and its final resolution, when Joseph and his brothers are reconciled (Gen 45:1–20; cf. 50:14–21). Genesis 38, the story of Judah and Tamar, is clearly a digression inserted into this narrative. As in the Abraham and Jacob cycles, the conclusion of the Joseph story is much less neatly tied up. Among the loose ends of this part of the story, we may count the information about Josephʼs administration in Egypt (Gen 47:13–26) and the last moments of Jacobʼs life in Egypt (Gen 47:27–48:22); Jacobʼs blessing (49:1–27); Jacobʼs death and burial (Gen 49:28–50:13); and the last part of Josephʼs life (50:22–26). The first part of Mosesʼs life, namely, the narrative of his birth (Exod 2:1–10), the first incidents of his adulthood (2:11–15a), and his sojourn in Midian (2:15b–22) can be classified in the same category of stories about the same personage. The sequence is merely chronological at the beginning, between the birth narrative and the first events in Mosesʼs public life (2:1–10 and 2:11–14; see 2:11: “and it happened in those days”), whereas Mosesʼs journey in Midian is the direct consequence of what happens in 2:11–14. The narrator also provides a logical link between the second and the third parts of the story: (2:15: “When/ because Pharaoh heard of it [. . .]”). We will come back to Moses later on.38

5 The Unified Plots A unified plot is a narrative where all the elements are necessary and where every episode is the consequence of what precedes and prepares for what follows. A good example from outside the Bible is the Iliad. The epic does not begin with the birth of the hero, Achilles, and does not finish with his death. It does not even start with the beginning of the Trojan War and does not finish with the conquest 36  F okkelman , Narrative Art (see n. 35): “Because the conflicts have disappeared, the ‘Deutungsbedürftigkeit’ immediately diminishes to a great extent, and in fact Gen. 35 can be left to tell its own story” (232); “the Story of Jacob shows a somewhat fragmented ending” (235). See, for a summary, Blum, Vätergeschichte (see n. 1), 149–151. 37  R. A lter , The Art of Biblical Narrative (New York: Basic Books, 1981), 155–177; M. Sternberg, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading (ISBL; Bloomington: Indiana University Press, 1985), 285–308; A. Wénin, Joseph ou lʼinvention de la fraternité (Le livre et le rouleau 21; Brussels: Lessius, 2005). 38  J. S. A ckermann , “The Literary Context of the Moses Birth Story (Exodus 1–2),” in Literary Interpretations of Biblical Narratives (ed. K. R. R. Gros Louis; Nashville, TN: Abingdon Press, 1974), 74–119; G. F. Davies, Israel in Egypt: Reading Exodus 1–2 (JSOTSup 135; Sheffield: Sheffield Academic, 1992); Fishbane, Text and Texture (see n. 35), 63–70.

What Do We Mean by Plot and by Narrative Continuity?

211

of Troy. The epic is only about Achillesʼs anger. It starts with the anger and finishes with the appeasement of this anger, after Achillesʼs victory over Hector.39 In the Bible, we find unified plots in the book of Ruth and in the book of Jonah. In the Pentateuch, the most unified plot is the Joseph story, as we have just seen. In all these cases, we have real plots, to use Forsterʼs expression. In the book of Jonah, for instance, the narrative begins with the prophetʼs flight to Tarshish (Jon 1:2), and the ideal reader immediately asks the question, “Why does Jonah fly to Tarshish instead of going to Nineveh?” The question is not “What next?” but “Why?” In Jon 4:2, Jonah explains the reason why (‫)על כן‬ he fled to Tarshish. This element is the logical linchpin of the whole narrative, which makes a plot for the book of Jonah.40 In the book of Ruth, we have more than a simple story or chronicle of Elimelechʼs family, because the narrative explains the reason why Naomi goes back to Bethlehem (Ruth 1:6: “For [‫ ]כי‬she had heard in the fields of Moab that the L ord had visited his people and given them food”). Ruth 2 describes Boazʼs generosity toward Ruth, and the latter asks in 2:10, “Why [‫ ]מדוע‬have I found favor in your eyes, that you should take notice of me, since I am a foreigner?” In chapter 3, Ruth proposes to Boaz to marry her for a very specific reason: “I am Ruth, your servant. Spread your wings over your servant, for you are a redeemer [‫]כי גאל אתה‬.” These elements give the narrative a logical backbone, and we therefore have a real plot in the book of Ruth.41 In the Joseph story, the cause of Josephʼs reconciliation with his brothers is given twice (45:5; 50:19–21). In Gen 45:5, Joseph says: “And now do not be distressed or angry with yourselves because [‫ ]כי‬you sold me here, for [‫]כי‬ God sent me before you to preserve life.” In 50:19–21, Joseph repeats the same explanation, introducing it with these words: “Do not fear, for [‫ ]כי‬can I take up the role of God?” The Joseph story does not only register events as in a simple chronicle. An effort is made to explain the reason why the protagonist made the decisions that gave this narrative its unique profile.42

6 Continuity or Discontinuity within the Book of Genesis? Since we have five books and not four separate sources, J, E, D, and P, we are entitled, I think, to inquire about the nature of the narratives contained in  Scholes and Kellogg, Nature (see n. 3), 209–210.  For more details, see, among others, Sternberg, Poetics (see n. 37), 318–320. 41  See, for instance, B. Green, “The Plot of the Biblical Story of Ruth,” JSOT 23 (1982), 55–68, reprinted in Beyond Form Criticism: Essays in Old Testament Form Criticism (ed. P. R. House; Sources for Biblical and Theological Study 2; Winona Lake, IN: Eisenbrauns, 1992), 206–218. 42  On this point, see Sternberg, Poetics (see n. 37), 308. 39 40

212

Jean Louis Ska

these five books. Arranging these narratives in five books may have a meaning and may have looked more meaningful to the compilers of the Pentateuch than the preservation of separate sources. In the New Testament, after all, the four Gospels are kept separate and Tatianʼs Diatessaron was never really accepted within Christian communities as a viable solution. A first observation is made by many commentators. The book of Genesis is organized according to the toledot formula, which means that the principle of organization is that of the genealogies.43 We have different types of genealogies; some are linear and contain only or almost only lists of names (Gen 5, 10, 11; 25:12–18; 36). Others are fleshed out with narratives, as in Hesiodʼs Theogony, for instance, Gen 12–25 (the Abraham cycle), Gen 25–35 (the Jacob cycle), and Gen 37–50 (the Joseph story). In the Abraham and especially in the Jacob cycles, we also have details of itineraries. Jacob travels a lot, from Canaan to Haran and back and then from Canaan to Egypt. Genesis 1–11, the origins of the world, is a special case, wherein we have an alternation of genealogies and longer narratives (Gen 1:1–2:4; 2:5–3:23; 4:1–26; 6:5–9, 17; 9:18–29; 11:1–9). Let me add a short note to conclude this very superficial analysis of the book of Genesis. There are several tensions, contradictions, and doublets within the book of Genesis. No doubt about that. But we have only one storyline, which starts with the creation of the world and stops with the twelve sons of Jacob/ Israel in Egypt. We find competing versions of individual episodes; we find competing descriptions of individual characters; but we do not find competing genealogies. The ancestors of Israel are Abraham, Isaac, and Jacob, and the whole family of Jacob goes down to Egypt. We do not find a genealogy in which Jacob is the sole ancestor of Israel or a genealogy where Ishmael is the son of the promise. There is one unifying element, namely, the genealogy of Israelʼs ancestors. Tensions, variants, doublets, and contradictions occur within this framework; they are not a feature of the basic framework itself, at least in its final version. A second note is also in order. There are discontinuities in the episodic plot of Genesis. The major discontinuities, however, are discontinuities between the different sections of Genesis, namely, Gen 1–11, the story of the origins of the universe; Gen 12–25, the Abraham cycle; Gen 25–35, the Jacob cycle; and Gen 37–50, the Joseph story. According to Robert L. Cohn, God is less and less present, the plot is more and more unified, and the characters have more and  On this point see, among others, B lum , Vätergeschichte (see n. 1), 432–446, with bibliography; D. M. C arr , “Βίβλος γενέσεως Revisited: A Synchronic Analysis of Patterns in Genesis as Part of the Torah,” ZAW 110 (1998), 159–172, 327–347; M. A. Thomas, These Are the Generations: Identity, Covenant, and the ‘toledot’ Formula (LHBOTS 551; London: T&T Clark, 2011). On the genealogies themselves, see T. Hieke, Die Genealogien der Genesis (Herders biblische Studien 39; Freiburg: Herder, 2003). 43

What Do We Mean by Plot and by Narrative Continuity?

213

more depth and personality.44 Speaking of plot and characters, we may notice that in Gen 1–11 every episode has a new set of characters, a different kind of plot answering a different question and coming to a different solution (Adam and Eve, Cain and Abel, Noah and his family, Noah and his three sons, the generation of Babel, etc.).45 Things change in the so-called patriarchal narratives (Gen 12–50), where we have the same set of characters in each series of episodes: Abraham and his family, Jacob and his family, Joseph and his brothers (Gen 12–25; 25–36; 37–50). As we will see, in Gen 12–25 the expectation of a son unifies most of the narratives; in Gen 25–36 Jacobʼs journey toward his uncle Labanʼs home and his return to Canaan is the main storyline; and in Gen 37–50 we attend the different stages of the brothersʼ reconciliation with Joseph after the initial conflict. There are thus four parts and four different kinds of plots in the book of Genesis. When there are tensions or contradictions within these different parts of Genesis, they are more or less of the same kind. They do not affect the general plot, however. More important, these four parts are connected on a very superficial level. We have a sequence of four distinct blocks of narrative, first about the origins of the universe and then about Israelʼs ancestors. We have stories; we do not have a real plot in the book of Genesis, and surely no unified plot. I am not speaking of the question of authors and sources. These are simple observations on the surface of the texts in their final, canonical, shape. When we pass from Gen 1–11 to Gen 12–25, from the origins of the universe to the Abraham cycle, we find hardly any real literary connection. The setting, the atmosphere, the problems, and the characters are different. The Abraham cycle can be read without knowing anything about Adam and Eve, Cain and Abel, Noah and the flood, and the tower of Babel.46 We can also say that all expectations are fulfilled after the birth of Isaac (21:1–7), and all the internal tensions disappear. If one wants to include Gen 22:1–19 in the picture, the conclusion is identical. After the Akedah, the Abraham cycle does not contain any internal dramatic tension that requires a concrete and urgent answer. The promises of the land and of numerous descendants create another kind of expectation since they can be fulfilled only in an indeterminate future.47  R. L. Cohn, “Narrative Structure and Canonical Perspective in Genesis,” JSOT 25 (1983), 3–16, republished in The Pentateuch: A Sheffield Reader (ed. J. W. Rogerson; BibSem 39; Sheffield: Sheffield Academic, 1996), 89–102. 45  Sutherland, “Organization” (see n. 34), 337–384; Helyer, “Separation” (see n. 34), 77–88; Fokkelman, “Time and Structure” (see n. 34), 96–109. 46  For more details, see F. C rüsemann , “Die Eigenständigkeit der Urgeschichte: Ein Beitrag zur Diskussion um den ‘Jahwisten,’ ” in Die Botschaft und die Boten: Festschrift H. W. Wolff (ed. J. Jeremias and L. Perlitt; Neukirchen-Vluyn: Neukirchener Verlag, 1981), 11–29. 47  For a different opinion on the function of the promises, see J. S. Baden, The Promise to the Patriarchs (Oxford: Oxford University Press, 2013). His analysis is based on very different 44

214

Jean Louis Ska

The same holds true for the Abraham and the Jacob cycles. Here again it is possible to read the whole of Gen 25–35 without any knowledge of what happened in Gen 12–25.48 Even when Jacob goes back to his motherʼs family, we are not told that he is going back to Abrahamʼs homeland. Allusions to what precedes are limited to very few texts, promises in general, and these texts have little bearing on the development of the plot. We may even notice that Gen 27, the story of Jacob stealing the blessing, does not allude to the famous patriarchal blessings. When Isaac unknowingly blesses Jacob, he does not speak of the promise of the land or the promise of numerous descendants. He speaks of the fertility of the land and of his power over nations and the members of his family (27:28–29). This comes as a surprise, since this would be the place to mention the famous promises renewed in favor of Isaac in Gen 26:2–5. In short, the patriarchal blessings are certainly mentioned in the Jacob cycle (see 28:3–4; 28:13–14; 35:11–12), but they do not impinge on the plot since the conflict between Jacob and Esau is about birthright (25:29–34) and a blessing different from the so-called patriarchal blessings (27:27–29).49 The Jacob cycle comes to its natural conclusion after Jacobʼs covenant with Laban in Gen 31:43–32:1 and his reconciliation with Esau in 33:17. All the conflicts and problems are solved; the narrative is rounded off well and does not require any continuation.50 The peculiar nature of the Joseph story has been underlined by most authors.51 The promises disappear almost completely and are mentioned only at the end of the whole narrative, when all the conflicts are solved and all dramatic tensions have abated (Gen 50:24). As in Gen 27, we may notice that Jacob does not mention the classical promises to the patriarchs when he blesses his grandsons (48:20) or even in the blessing of his twelve sons (Gen 49). The conflict between Joseph and his brothers is caused by Josephʼs privileged position in the family and his dreams, not by the patriarchal blessings, and has little to do with the Jacob-Esau-Laban narrative. This conflict is a new element that has no connection with what precedes, and nothing in the former narrative cycles contributes to its solution. We are in a new narrative world, with its new setting, its own set of characters, and its own plot. At the other end of the Joseph story, when Joseph premises, however. 48  For an analysis of Gen 25–35 from this point of view, see F okkelman , Narrative Art (see n. 35). 49  For more details, see, among others, Blum, Vätergeschichte (see n. 1), 86–88, 149–151. 50  See, among others, B lum , Vätergeschichte (see n. 1), 147–148: “Gen 33,17: Ein Erzählschluß.” 51  See, for instance, G unkel , Genesis (see n. 32), liii-liv, 395–401; C. W estermann , Genesis 37–50 (BK I/3; Neukirchen-Vluyn: Neukirchener Verlag, 1982), 8–16; B lum , Vätergeschichte (see n. 1), 230–234; K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 56–63.

What Do We Mean by Plot and by Narrative Continuity?

215

reconciles with his brothers, the story reaches its conclusion, and the reader has answers to all the questions raised at the beginning of the narrative in Gen 37. It is always possible to ask questions such as “What happened next to Joseph and his brothers in Egypt?” but this question receives the usual answer: “This is another story, which we will tell some other time.”

7 Continuity or Discontinuity between Genesis and Exodus?52 Is there, from a narrative point of view, a logical link between Genesis and Exodus–Numbers? Leaving aside many discussions on the topic, let me treat this question from a different viewpoint, that of the narrative structure of the Pentateuch.53 The following simple observations may shed some light on this thorny problem, because we may reach a level of the stories that is preserved in all the sources or independently of the sources. As will be investigated in what follows, three points deserve attention: the presence of the Hebrews in Egypt, their proliferation in this foreign country, and their oppression at the hand of the Egyptians. 7.1 Why Are the Hebrews in Egypt?54 The answer to this question is obvious for the reader of the book of Exodus in its present shape because it is given in Exod 1:1: “These are the names of the sons of Israel who came to Egypt with Jacob, each with his household.” The verse is placed in a strategic position since it orients the reading of the whole book and makes all the Hebrews present in Egypt in the book of Exodus descendants of Jacob. This is clear enough, but it does not solve the problem in a completely satisfactory way. Joseph explains why Jacob and his family came to Egypt in Gen 45:6: “For the famine has been in the land these two years, and there are yet five years in which there will be neither plowing nor harvest.” Joseph repeats this explanation in 50:19–21. In other words, Jacobʼs family settles initially in 52

5–8.

 In these paragraphs, I develop some ideas presented in K. Schmid, Erzväter (see n. 51),

53  Among the recent contributions to the debate, see the dialogue between L. Schmidt, “Die vorpriesterliche Verbindung von Erzvätern und Exodus durch die Josefsgeschichte (Gen 37; 39–50*) und Exodus 1,” ZAW 124 (2012), 19–37, and E. Blum, “Zwischen Literarkritik und Stilkritik: Die diachrone Analyse der literarischen Verbindung von Genesis und Exodus – im Gespräch mit Ludwig Schmidt,” ZAW 124 (2012), 492–515; J. S. Baden, “The Continuity of the Non-Priestly Narrative from Genesis to Exodus,” Bib 93 (2012), 161–186; and K. Schmid, “Genesis and Exodus as Two Formerly Independent Traditions of Origins for Ancient Israel,” Bib 93 (2012), 187–208. 54  See the still very instructive study by H. E ngel , Die Vorfahren Israels in Ägypten (Frankfurt am Main: Knecht, 1979).

216

Jean Louis Ska

Egypt to survive a five-year famine. In no way do we hear about a plan to settle in Egypt for generations, even for centuries. On the other hand, it seems that the Hebrews would have remained in Egypt forever had one pharaoh not decided to oppress them because they were too numerous. Now, according to the Joseph story, the sojourn of the Hebrews in Egypt was temporary, not definitive, and nothing prepares the reader for a stay that lasted 430 years (Exod 12:41). We are not told why Jacobʼs family did not go back to Canaan after the five-year famine, which would have been quite sensible. Here is a first “Why?” that does not receive a clear answer in the texts at our disposal. Moreover, the reader may wonder why the Hebrews in Egypt did not remember any of the patriarchal promises of the land for 430 years. Again, it seems that the Israelites had settled permanently in Egypt and had even forgotten that their ancestors had come down with Jacob and his family from Canaan, the land promised by God to their ancestors. When reading Exod 1–15, the reader may even have the impression that the Hebrews are part of the population of Egypt, as the Scots are in Great Britain, the Bretons in France, the Basques, the Catalans, and the Galicians in Spain, or the Calabrians in Italy. Never do the Egyptians or Pharaoh tell them: “Go home!” On the contrary, Pharaoh and his ministers try by all means to prevent them from leaving the country because they are useful, even indispensable. This impression is substantiated by several other summaries of the origins of Israel. Ezekiel 20, for instance, says that God revealed himself to Israelʼs ancestors in Egypt: “Thus says the Lord YHWH: On the day when I chose Israel, I swore to the offspring of the house of Jacob, making myself known to them in the land of Egypt; I swore to them, saying, I am YHWH your God” (Ezek 20:5). For Jeremiah as well, Israelʼs origins are to be found in Egypt, when he rebukes his people, saying: “[Your fathers] did not say, ‘Where is YHWH who brought us up from the land of Egypt, who led us in the wilderness?’ ” (Jer 2:6). Neither Ezekiel nor Jeremiah mention the patriarchsʼ descent to Egypt. Hosea 12:10 and 13:4 go along the same lines when God says, “I am YHWH your God from the land of Egypt.” The history of Israel as a nation begins in Egypt, it seems. We may notice that Hos 12 speaks of several episodes from the Jacob cycle but does not mention his going down to Egypt with his family.55 The same holds true for Amos 3:1, 9:7, and Mic 6:4. God “brings Israel up from Egypt,” he does not bring Israel back from Egypt. The exodus is not presented as a journey back home.  On Hosea, see H.-D. N eef , Die Heilstraditionen Israels in der Verkündigung des Propheten Hosea (BZAW 169; Berlin: de Gruyter, 1987); D. R. Daniels, Hosea and Salvation History: The Early Traditions of Israel in the Prophecy of Hosea (BZAW 191; Berlin: de Gruyter, 1990); T. B. Dozeman, “Hosea and the Wilderness Wandering Tradition,” in Rethinking the Foundations: Historiography in the Ancient World and in the Bible; Essays in Honour of John Van Seters (ed. S. L. McKenzie and T. Römer; BZAW 294; Berlin: de Gruyter, 2000), 55–70. 55

What Do We Mean by Plot and by Narrative Continuity?

217

7.2 Why Do the Hebrews Multiply in Egypt? The proliferation of the Hebrews in Egypt is described in Exod 1:7, a text usually attributed to the Priestly writer: “But the people of Israel were fruitful and increased greatly; they multiplied and grew exceedingly strong, so that the land was filled with them.” In Exod 1:9, the event is reported by Pharaoh in a short discourse and in a different vocabulary: “Behold, the people of Israel are too many and too mighty for us.” It comes as a surprise, however, that none of these texts refer to the promises to the patriarchs. The vocabulary in Exod 1:7 is certainly similar to that of other Priestly texts, but even there we miss the typical Priestly sentence, “as YHWH said,” that we find, for example, in Gen 21:1. This exceptional increase could be interpreted, for instance, as the fulfillment of a promise made to Jacob in Gen 46:3, where we read, “Then [God] said, ‘I am God, the God of your father. Do not be afraid to go down to Egypt, for there I will make you into a great nation [‫]גוי גדול‬.’ ” But there is no allusion to this promise in Exod 1. Even the vocabulary is different. The reason why the Hebrews suddenly multiply in Egypt is not given. We have a typical and then and no therefore or because, and hence we have a story and no plot. 7.3 Why Are the Hebrews Oppressed in Egypt? Examining the reason why the Israelites were oppressed in Egypt will advance our argument because the sequence between the end of Genesis and the beginning of Exodus is more chronological than logical. The transition from Genesis to Exodus is actually summarized in Exod 1:8: “Now there arose a new king over Egypt, who did not know Joseph.” The verse contains two basic pieces of information: a new situation is created by the arrival of another pharaoh, and this pharaoh does not know Joseph and what he did for Egypt. Pharaoh may not know Joseph, but the narrator and the reader cannot ignore him, as he is mentioned by name in the text. Moreover, the oppression in Egypt is a consequence of this pharaohʼs lack of knowledge, as the text clearly suggests: Pharaoh did not know Joseph and, we may say, therefore he took all the initiatives described in the following section. This verse is surely crucial for a study of the relationship between Genesis and Exodus, since it presupposes a logical connection between the Joseph story and the oppression in Egypt. The verse is isolated, however, and the name Joseph never reappears in the following chapters. For instance, Joseph is never used as an argument to convince Pharaoh that the Israelites came from Canaan and that their presence, at least in the person of Joseph, was beneficial to the country. The real and the only “cause” of the oppression is the proliferation of the Israelites, a phenomenon that the Egyptians consider threatening (1:9) but which is not explicitly connected with events recounted previously.

218

Jean Louis Ska

Neither is it ever said that the Hebrews were oppressed because they were Hebrews or because they were foreigners coming from a foreign region. The Egyptians never accuse the Hebrews of having invaded their country or being illegal immigrants. The story would be the same were the Hebrews one of the ethnic groups of Egypt. It is again difficult to speak of a plot in the technical sense of the word in this case. Let us add one element to these observations. As Konrad Schmid has argued, the revelation of God to Moses in Exod 3:14–15 and 6:2–8 is presented as something completely new and unexpected.56 The Hebrews in Egypt did not know YHWH at all. If there were continuity in the sources, this phenomenon would at least require an explanation. The evidence provided here points, therefore, to a lack of real logical connection between Genesis and the beginning of Exodus.

8 The Episodic Plot of the Wanderings in the Wilderness A second group of stories is organized according to the principle of the itinerary. I have in mind the account of the wanderings in the wilderness that starts more or less in Exod 12 and finishes either at the end of the book of Numbers or at the end of the book of Deuteronomy, since the setting of the book of Deuteronomy is the same as the setting of the last chapters of the book of Numbers.57 To be sure, I do not deny that there are tensions, contradictions, and signs of the presence of different sources in these two blocks of narrative. My only contention is that the books are organized in such a way that this principle is more important than the sources. This is an overarching structure and, I think, a meaningful one. We may even say that all the elements, or almost all the elements, that are part of these “wanderings in the wilderness” fit into the structure. In other words, in the books of Exodus–Numbers/Deuteronomy, we have stages of an itinerary through the wilderness, and all the variations are variations around this theme, namely, different versions of the sojourn in the desert and of the wanderings in the wilderness. The journey starts with the exodus, when Israel comes out of Egypt, and ends when Israel reaches the Jordan Valley near Jericho. The final destination is the land of Canaan, and it is always the land of Canaan. And the way goes through the wilderness. There can be differences along the road, of course, and some may discuss, for instance, the possibility of an itinerary that did skip Sinai/Horeb. But this is “part of the game.” In an episodic plot such as the wanderings in the desert, it is possible to tell different  K. Schmid, Erzväter (see n. 51), 7–8, 90–91, 186–188, and passim.  On these texts, see, for instance, W. W. L ee , Punishment and Forgiveness in Israelʼs Migratory Campaign (Grand Rapids, MI: Eerdmans, 2003); C. D. Kupfer, Mit Israel auf dem Weg durch die Wüste: Eine leseorientierte Exegese der Rebellionstexte in Exodus 15:22–17:7 und Numeri 11:1–20:13 (OTS 61; Leiden: Brill, 2011). 56

57

What Do We Mean by Plot and by Narrative Continuity?

219

versions of the same story (Exod 17 and Num 20, Num 13–14) or to offer shorter or longer versions of the same journey, for instance, adding more stages to the list. This does not change the nature of the overarching plot that describes wanderings from Egypt to the land of Canaan through the wilderness. It is possible – and several exegetes have proposed this solution in the recent past – to consider the block Exodus–Deuteronomy as a biography of Moses.58 This may be the case, but if so it is only to a certain extent. There are two major objections to this opinion. First, Moses does not appear immediately in the story after Josephʼs death. There is at least one full chapter, Exod 1, between the end of the patriarchal narratives, the genealogical part of the Pentateuch, and the birth of Moses. This means that the birth of Moses does not coincide with the beginning of the next long section of the Pentateuch. Exodus 1 relates the origins of Israel as a “people” in Egypt and its oppression by the Egyptians. The main character or the protagonist of the narrative is not Moses but the people of Israel. The main element of the narrative, from Exod 1 on, is not Mosesʼs fate but Israelʼs destiny. Moses enters onto the stage as potential savior of Israel, at least in the Pentateuch as it stands now. Moses is instrumental in the liberation of Israel. This is perhaps a question of nuances, but the nuance is essential. The main focus of the narrative, from Exod 1 on, is not Moses but Israel. What matters is Israelʼs destiny, not exactly Mosesʼs career. Second, Moses is not present in all the episodes of Israelʼs history in Exodus–Deuteronomy. He is surely present in most of them, even in a majority of them. The ending of the story coincides with his death in Deut 34. But he is not present in a few instances, as in the Balaam–Balak episode (Num 22–24) or a few battle reports in Num 21:1–3, 10–30.59 Israel is present in these pages, not Moses. This means that the unifying element of the whole complex is not Moses as such but Israel. Moses is at the service of Israel; he is instrumental in its liberation from Egypt and its welfare in the wilderness. Only in Exod 2 is the narrative focused mostly on Moses and his fate, although Israel is not completely absent from the context (see Exod 2:11, 13, 22). For these reasons, it seems difficult to consider Moses as the really unifying element in the Exodus–Deuteronomy complex.

 See, for instance, the significant work of J. Van Seters, The Life of Moses: The Yahwist as Historian in Exodus-Numbers (Louisville, KY: Westminster John Knox, 1994); for more details on this topic, see, among others, G. W. C oats , Moses: Heroic Man, Man of God (JSOTSup 57; Sheffield: JSOT Press, 1988); idem , The Moses Tradition (JSOTSup 161; Sheffield: Sheffield Academic, 1993). 59  On Num 21:31–35 not mentioning Moses, see A. Rofé, Introduction to the Composition of the Pentateuch (BibSem 58; Sheffield: Sheffield Academic, 1999), 118–119. This text is a late interpolation that comes from Deut 3:1–3. For more details and a bibliography, see U. F istill , Israel und das Ostjordanland: Untersuchungen zur Komposition von Num 21,21–36,13 im Hinblick auf die Entstehung des Buches Numeri (ÖBS 30; Frankfurt am Main: Peter Lang, 2007), 50–51, with n. 10. 58

220

Jean Louis Ska

In this long series of stories, the chronological element prevails most of the time. The question is certainly “What next?” and most of the episodes are very loosely connected to each other. This is the reason it was possible to modify the itinerary, to add stages and stories, to introduce variants and comments. In one case, however, we have a logical element, an explanation of the long wanderings of Israel in the wilderness, and this is Num 13–14.60 The rebellion of the Israelites in this episode is the reason why the people had to sojourn so long – forty years, according to Num 14:34 – in the wilderness. These passages (Num 14:21–25, 26–35) create a plot in a long series of stories.

9 The Narratives about Patriarchs and the People of Israel After these reflections on the composition of the Pentateuch, it is possible to see two main parts in it, namely, the history of the origins in Genesis, structured as a “genealogy” of the universe and of Israelʼs ancestors on the one hand and the long journey of Israel from Egypt to the banks of the Jordan on the other. Exodus 1–15 is a long narrative describing first the oppression, then the decision to leave Egypt forever and the exodus itself. We may ask whether we have here only stories or also a plot. The answer, in my opinion, comes from only one text that creates a link of causality between the patriarchal narratives on the one hand and the Exodus and the wanderings in the wilderness until reaching the promised land on the other. This text, as noted by several exegetes, is Exod 6:2–8.61 This short divine discourse, which summarizes the main events of Israelʼs theological history, provides a logical link between the promises to the patriarchs and the events of the exodus and the wilderness. This link is the oath sworn to the patriarchs (‫)ברית‬. The oath is the cause of Godʼs intervention on behalf of Israel in Egypt and of the other events preceding the entrance into the land promised by oath to Abraham, Isaac, and Jacob (Exod 6:8; and see 6:4).62 60  On this text, see, among others, R. Achenbach, “Die Erzählung von der gescheiterten Landnahme von Kadesch Barnea (Numeri 13–14) als Schlüsseltext der Redaktionsgeschichte des Pentateuch,” ZABR 9 (2003), 56–123; S. B oorer , “The Place of Numbers 13–14* and Numbers 20:2–12* in the Priestly Narrative (Pg),” JBL 131 (2012), 45–53. 61  Among the many studies on this important text, see Blum, Studien (see n. 1), 232–242; J. C. G ertz , Tradition und Redaktion in der Exoduserzählung: Untersuchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000) 237–250; A. A. D iesel , “Ich bin Jahwe”: Der Aufstieg der Ich-bin-Jahwe-Aussage zum Schlüsselwort des alttestamentlichen Monotheismus (WMANT 110; Neukirchen-Vluyn: Neukirchener Verlag, 2006), 95–118. 62  See the fundamental analysis of this text by K. Schmid, Erzväter (see n. 51), 170–171, 255–266. On the independence of the early traditions on the patriarchs and those on the exodus, see W. S taerk , Studien zur Religions- und Sprachgeschichte des Alten Testaments

What Do We Mean by Plot and by Narrative Continuity?

221

Exodus 6:6, the verse that most interests us in this discussion, is the linchpin between the two parts of the divine discourse, the one dedicated to the past and to the patriarchs (6:2–5) and the one dedicated to the future of Israel, extending as far as the entrance into the promised land (6:6–8).63 The particle ‫לכן‬, “therefore,” in the short formula ‫ אני יהוה‬. . . ‫לכן אמר‬, “Therefore say . . . ‘I am the Lord’ ” (6:6), is an essential element in this text and in the exegesis of the Pentateuch. Because of this particle and the content of the oracle, the Pentateuch becomes a plot in Forsterʼs sense. This element is certainly a creation of the Priestly writer and is absent from the other sources or layers of the Pentateuch. It is juxtaposed to other series of stories but is not a substitute for the other chronicles or series of narratives about the origins of Israel. The novelty of Exod 6:2–8, with ‫ לכן‬in 6:6, is twofold. First, patriarchs and exodus are no longer two parallel traditions. They are logically linked in a single plot, since the exodus is the fulfillment of an oath sworn to the patriarchs. Second, the god of the patriarchs is the same as the god of Exodus. Only the names are different.

10 Conclusion One can certainly speak, of course, of plot and narrative continuity in reference to the Pentateuch and its different parts or sources. The Pentateuch presents itself as a relatively coherent series of narratives about the origins of Israel. We find in Genesis a very simple type of story, that of a genealogy, fleshed out here and there with narratives about some personages, especially in the case of the patriarchs. In Exodus–Deuteronomy, the narrative structure is that of a journey, with a long section dedicated to the departure from Egypt (Exod 1–15) and the

(2 vols.; Berlin: Reimer, 1899), 1: 48–51; K. G alling , Die Erwählungstraditionen Israels (BZAW 48; Gießen: Töpelmann, 1928), 56–63; A. de P ury , “Le cycle de Jacob comme légende autonome des origines dʼIsraël,” Congress Volume Leuven, 1989 (ed. J. A. Emerton; VTSup 43; Leiden: Brill, 1991), 78–96, reprinted in Die Patriarchen und die Priesterschrift/ Les Patriarches et le document sacerdotal: Gesammelte Studien zu seinem 70. Geburtstag/ Recueil dʼarticles, à lʼoccasion de son 70e anniversaire (ed. J.-D. Macchi et al.; ATANT 99; Zürich: Theologischer Verlag, 2010), 93–108; idem, “Osée 12 et ses implications pour le débat actuel sur le Pentateuque,” in Le Pentateuque: Débats et recherches (éd. P. Haudebert; LD 151; Paris: Cerf, 1992), 175–207; idem, “Dos leyendas sobre el origen de Israel (Jacob y Moisés) y la elaboración del Pentateuco,” EstBib 52 (1994), 95–131; idem, “Le choix de lʼancêtre,” TZ 57 (2001), 105–114, reprinted in Macchi, Römer, and Schmid (eds.), Die Patriarchen und die Priesterschrift, 109–118. 63  See J. L. Ska, “Quelques remarques sur Pg et la dernière rédaction du Pentateuque,” in Le Pentateuque en question (ed. A. de Pury; 3rd ed.; MdB 19; Geneva: Labor et Fides, 2001), 95–125, esp. 98–102.

222

Jean Louis Ska

rest to the different stages in the wilderness up to the last days of Moses on the banks of the Jordan, in front of Jericho. It would be necessary, however, to introduce many differences and nuances. In most cases we have stories, and only in a few cases do we have plots, to use E. M. Forsterʼs handy distinction. To come back to the problem of the relationship between Genesis and Exodus–Deuteronomy, we saw how difficult it really is. The only text that creates a plot, in the technical meaning of the word, i.e., a link of causality between the patriarchal narratives and the journey of Israel from Egypt toward the promised land, is Exod 6:2–8. This means, in other words, that we have to make sure that we know what we are saying when we speak of narrative continuity, since a series of stories is not yet a plot, and we have different kinds of stories and different kinds of plots in the Pentateuch, as in world literature. In a few words, it seems to me that speaking of narrative continuity with respect to the different components of the Pentateuch, with the exception of the Priestly writer, becomes very problematic. We hardly find continuous sources in the Pentateuch, and instead of narrative continuity it would be better to speak of narrative sequences, since the connections between the main parts are more chronological than logical. This is especially true when analyzing the transition between Genesis and Exodus. We have many stories in the Pentateuch, but only Exod 6:2–8 creates a plot that unites its different parts. Moreover, the literary composition of the Pentateuch as a whole makes sense, as we saw, and it is still possible to read it as it stands, without necessarily dissecting it. This was possible for centuries and is still possible today. We have a chronological sequence, starting with the creation of the world and ending with Mosesʼs death. Within this chronological framework, there is obviously room for a plurality of voices and songs and even for some symphonies, such as that composed by the Priestly writer.

Travel Narratives and the Message of Genesis Yairah Amit Prologue The genre of travel narrative, characteristic in many cultures, has a substantial place in the Torah literature, most of which is dedicated to the description of Israelʼs journey from Egypt to the promised land.1 However, in this article I will focus on the book of Genesis and show that the book is formulated as a book of various kinds of journeys: of migration, of wanderings, and especially of exile. Moreover, I will argue that the travel narratives carry the message of the book of Genesis, and the use of the journeys is not simply a technical editorial means for linking the different stories. Rather, they have theological, national, and historical purposes. All of this evidence also points to the late date of the editorial work. Before we set out to follow the patriarchsʼ travels in the book of Genesis, let me state that in this article my loyalty is to the approach of the biblical editors, who regard the Joseph story (Gen 37:1–50:26) as an integral part of the Jacob stories (Gen 25:19–36:43).2 I take this view even though comparing the Joseph story to the patriarchsʼ stories (11:26–36:43) shows that the Joseph story is a diverse and independent composition with special literary qualities: It is distinct from all previous narratives because of its unusual length, [. . .] it is from beginning to end an organically constructed narrative. [. . .] This mastery of the material by a clear succession of scenes, this division of the massive and very complex occurrence into individual eventful waves, shows without question a very superior artistry in representation.3

 This article will not deal with the long travel narrative of the Exodus; see p. 228, below.  For loyalty to the same approach, see R. J. C lifford , “Genesis 37–50: Joseph Story or Jacob Story?,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; VTSup 152; Leiden: Brill, 2012), 213–229. 3  G. von R ad , Genesis: A Commentary (OTL; trans. J. H. Marks; London: SCM Press, 1963), 342. See also K. Schmid, Genesis and the Moses Story: Israelʼs Dual Origins in the Hebrew Bible (trans. J. D. Nogalski; Siphrut 3; Winona Lake, IN: Eisenbrauns, 2010), 50–53. There are of course more features that differentiate between the Joseph story and the patriarchsʼ cycles. For example, God is behind the scenes in the Joseph story, and there are different kinds of dreams in the Joseph story (for Gen 46:1–4 see notes 5 and 49 below). However, this is not the theme of this essay. 1 2

224

Yairah Amit

It is quite possible then that the Joseph story was an independent unit before the editors of Genesis and the Pentateuch as a whole decided to place it between the patriarchsʼ cycles4 and the book of Exodus. Naturally the new location required them to fit the story to what precedes it and what comes after it.5 The editorsʼ imprint and approach is prominent already in the peculiar phrasing of the two opening verses of the Joseph story (Gen 37:1–2): “Now Jacob was settled in the land where his father had sojourned, the land of Canaan. This, then, is the line of Jacob: At seventeen years of age, Joseph tended the flocks with his brothers.”6 On the transition from Jacob to Joseph, Gerhard von Rad writes: What is interesting about this redactorial process is that what we like to call the Joseph story is really conceived as a Jacob story [. . .] and Jacob dies only at the end of the “Joseph Story” (Gen 49:33).7

Without referring to the terms redaction or editing, Samuel David Luzzatto (Shadal), building on the insights of Rashi and Rashbam, argued: “The people mentioned below are the descendants of Yaakov.”8 He explains that the following stories include all that happened to the sons of Jacob. In other words, these commentators understand the Joseph story as an integral part of the story of Jacobʼs family. Hence we find that Jacobʼs familyʼs journeys are followed past chapter 36, as far as chapter 50.9 In addition, my point of view in this article is  On the term cycle see Y. Amit, The Book of Judges: The Art of Editing (trans. J. Chipman; BibInt 38; Leiden: Brill, 1999), 222. 5  See for example the connection between Gen 46:8–27 and Exod 1:1–5; Gen 47:27 and Exod 1:7; Gen 50:24–26 and Exod 13:19 and Josh 24:32. Scholars do not agree about the extent of the redactorsʼ work in the Joseph story; see J. Skinner, A Critical and Exegetical Commentary on Genesis (ICC; Edinburgh: T&T Clark, 1963 [first ed., 1910]), 438–439, and compare to von Rad, Genesis (see n. 3), 342–343; see also n. 7, below. 6  Emphasis added. Here and elsewhere I use the English translation of the Jewish Publication Society. 7  V on R ad , Genesis (see n. 3), 345. According to S kinner , Genesis (see n. 5), 443, the redactor used “[t]hree disjointed fragments of P” in these two verses. Claus Westermann emphasizes that the Joseph story was combined into Jacobʼs stories and that Gen 37:1–2 are part of the Jacob cycle. Thus some units from Joseph story in Gen 46–50, including Gen 46:1–27, 48:1–19, 21–22, and others are the ending of the Jacob stories; C. W estermann , Genesis 37–50: A Commentary (trans. J. J. Scullion; London: SPCK, 1987), 22–24, 35–36. See also D. B. Redford Egypt, Canaan, and Israel in Ancient Times (Princeton: Princeton University Press, 1992), 422–429. Gordon Wenham thinks that Gen 37:1 ends the genealogies of Esau in Gen 36 and that Gen 37:2 is the beginning of the Joseph story; G. J. Wenham, Genesis 16–50 (WBC; Waco, TX: Word Books, 1994), 334–350. 8  S. D. Luzzatto, Shadal – Torah Commentary by Shmuel David Luzzatto (trans. E. Munk; 4 vols.; Jerusalem: Lambda, 2012 [Italian orig., Padua; F. Sacchetto, 1871–1876]), 2:355. 9  Actually journeys begin from the expulsion from Eden (Gen 3:23–24) but this article will not discuss most of the exposition to the Patriarchal stories (i.e., Gen 1:1–11:32), only the journey of Terah to Haran (Gen 11:31–32). 4

Travel Narratives and the Message of Genesis

225

synchronic and therefore I rarely discuss literary sources but rather take a birdʼs eye view of the final edited work before us.

Is “Travel Narrative” an Influential Genre? It seems that moving into places distant and unknown, places that will offer unexpected encounters, has always aroused human curiosity and imagination. There have always been people who dared to look for new and different places and those who told about their real or imaginary travels. In every era, listeners and readers of all ages drank up these tales. Journey stories appear in many literary genres.10 Sometimes the tales were assembled in travel books where the travel narrative itself became the main story; in others the journey appears only as part of the plot. Travel narratives appear in the literature of different peoples and in stories for children too. We know about the journey of Odysseus in Homerʼs epic (eighth c. BCE),11 the travels of Benjamin of Tudela who visited Europe, Asia, and Africa in the twelfth century CE,12 and the adventures of Gulliver in Jonathan Swiftʼs novel (1726).13 Reference to travel is made in nonliterary writing such as historical and geographical diaries, for example Herodotusʼs descriptions of the East in his Histories (fifth c. BCE)14 and the travel diaries of Fan Chengda in China (twelfth c. CE).15 Often travel narrative does not appear as a genre in lexicons of literary terms,16 which makes formal definitions difficult, save the observation that a journey is

10  See further in this volume, with attention to travel narratives and itineraries, J. L. Ska, “What Do We Mean by Plot and by Narrative Continuity?,” pp. 201–222. 11  Homer, The Odyssey (trans. A. T. Murray; LCL 104; Cambridge, MA: Harvard University Press, 1919). 12  B. T udela , The Itinerary of Benjamin of Tudela: Travels in the Middle Ages with Introductions by M. A. Singer (1983), M. N. Adler (1907), and A. Asher (1840) (Malibu, CA: Joseph Simon/Pangloss Press, 1987), 162–166. 13  J. Swift, Gulliverʼs Travels (ed. D. Womersley; The Cambridge Edition of the Works of Jonathan Swift; Cambridge: Cambridge University Press, 2012). As W. R. Benet, The Readers Encyclopedia (New York: Crowell, 1951), 467, notes: “Gulliverʼs Travels, frequently looked upon as a mere childrenʼs book, is in reality a biting social and political satire.” 14  R. B. Strassler (ed.), The Landmark Herodotus: The Histories (New York: Pantheon, 2007). 15  J. M. H argett , On the Road in Twelfth Century China: The Travel Diaries of Fan Chengda (1126–1193) (Münchener ostasiatische Studien; Stuttgart: Steiner, 1989). 16  E.g., J. A. Cuddon, A Dictionary of Literary Terms (rev. ed.; London: Andre Deutsch, 1979), 715–721 includes an entry “Travel Book” that opens with these words: “A neglected and much varied genre of great antiquity to which many famous, more or less professional or ‘full-time’ writers have contributed, but which has also been enriched by a number of occasional writers. For most part these have been diplomats, scholars, missionaries, soldiers

226

Yairah Amit

the background for travel narratives. Pardes explores what characterizes a travel narrative: The limits of the genre are somewhat blurred, a few definite characteristics can be noted: there is the search for the unknown, entry into marvelous scenes, wonderment on accepted patterns that go with departure from the familiar culture, encounter with inhabitants of foreign lands, blurring boundaries between the external and the figurative journey, between the geographical map and the imaginary map, the internal. In addition, the ancient travel narratives – the main corpus that leads me in reading the book [of Jonah] – deal with complicated relations between god(s) and mortals. This is true whether the journey is formulated against the god(s) with longing to encounter the divine, with obedience to the divine command, or with any combination of these possibilities. The divine-earthly axis is integral to the map of the journey.17

Even without a specific travel narrative genre, plot sequences that move from one place to another appear in all the basic literary genres including prose, poetry, and drama. We can then define the travel narrative as a subgenre that serves the structure and advances the plot. The reader must examine its use within the complete work and ask if the journey is simply a technical means for linking different stories and story elements or if the journey is significant and purposeful in itself. Our observation of some ancient travel narratives begins with the report of Wenamun from the eleventh century BCE. This story tells of the business travel of an emissary sent to Lebanon to bring cedar wood for building the sacred ship of the god Amun.18 The narrative explains how a task that should have taken two weeks to a month became a journey of two years full of trials and adventures that created problems regarding Wenamunʼs Egyptian identity. We do not know whether this story had a place in the literary repertoire of Egyptian culture and how many people were acquainted with it. However, numerous copies of the story of Sinuhe, written some thousand years before Wenamun, were found in Egyptian schools for scribes, and one may assume this indicates interest in travel stories.19 Sinuhe was a senior official of fortune, doctors, explorers and sailors.” The continuation of the entry is a list of books with a short description or information relating to the books or to their authors. 17  I. Pardes, “The Belly of the Underworld: The Book of Jonah as Travel Narrative,” in Wool from the Loom: The Development of Literary Genres in Ancient Literature (ed. N. Wasserman; Jerusalem: Magnes, 2002), 14 (Hebrew). 18  See H. Goedicke, The Report of Wenamun (Baltimore: Johns Hopkins University Press, 1975). Scholars note that Wenamunʼs account spans the genres of report and story. See, e.g., J. B. Pritchard (ed.), ANET, 25: “The story is almost picaresque in its atmosphere and must be classed as a narrative.” 19  “The tales of the Eloquent Peasant, of the Shipwrecked Sailor [1850 BC], and of Sinuhe are travel narratives or at least literary treatments of the travelling-abroad motif”; G. Moers, “Travel as Narrative in Egyptian Literature,” in Definitely: Egyptian Literature: Proceedings of the Symposium “Ancient Egyptian Literature: History and Forms,” Los Angeles, March 24–26 (ed. G. Moers; Studia Monographica 2; Göttingen: Seminar für Ägyptologie und

Travel Narratives and the Message of Genesis

227

of Egyptʼs Middle Kingdom who became a political exile and found shelter in Qedem, which was Syria or northern Transjordan. His stories tell readers of experiences on his journey into exile and in his new geographical and human surroundings. Going away to a foreign country is seen as a double confrontation – one with the foreigners there and one with the travelerʼs own identity. Sinuhe becomes aware of his Egyptian identity, of the general concept of human life and of Egyptian life in particular, so that Sinuheʼs travels add personal and intellectual value.20 At the same time we have no evidence that this literature was known to the Hebrew reader or listener. The situation is different in the case of Mesopotamian literature. A fragment of the Gilgamesh Epic from the Late Bronze Age (fourteenth century BCE)21 discovered at Megiddo may indicate that ancient Israelite scribes of the first millennium familiar with different oral traditions from the second millennium, such as the Ugaritic epics, could have been acquainted with the Gilgamesh Epic and possibly with other Canaanite material too. Moreover, if scholars find echoes of Gilgamesh not only in Genesisʼ flood story, but also in the Jacob stories22 from a late period,23 it proves that after the Assyrian occupation (eighth c. BCE) this epic had the status of classical literature. Gilgamesh tells of two journeys. First is the tale of the perilous adventures of Gilgamesh and Enkiduʼs journey to the cedar forest, the habitation of its monstrous guardian Humbaba. Gilgamesh and Enkidu wanted to kill Humbaba and make their names world famous. Then after additional tales and Enkiduʼs death, Gilgamesh sets out alone on a long, dangerous journey to meet Utnapishtim, hero of the flood, and to obtain from him the secret of eternal life.24 The evidence

Koptologie, 1999), 51. Moers also compares the story of Sinuhe to the report of Wenamum in order to emphasize their differences. He presents two other examples of the Egyptian travel narrative genre: the Letter of Wermai, better known as The Tale of Woe, where the foreign travel motif is transposed into Egypt itself, and the narrative of the Bentresh stela, where he takes with him an idol of Amun-of-the-Road. 20  See G. Moers, “Travel as Narrative” (see n. 19), 53–54; see also A. Loprieno, “Travel and Fiction in Egyptian Literature,” in Mysterious Lands (eds. D. OʼConnor and S. Quirke; London: UCL Press, 2003), 31–51. 21  See Y. Goren et al., “A Provenance Study of the Gilgamesh Fragment From Megiddo,” Archeometry 51, no. 5 (2009), 763–773, and the bibliography there. The authors conclude that the tablet was written in Gezer and that “the Gezer scribal school also studied the classical literary texts of the Mesopotamian scribal tradition and distributed them to other Canaanite centers.” 22  See E. J. Hamori, “Echoes of Gilgamesh in the Jacob Story,” JBL 130 (2011), 625–642. 23  See N. Na‍ʾaman, “The Jacob Story and the Formation of Biblical Israel,” Tel Aviv 41 (2014), 95–125, who argues that the pre-Priestly Jacob story was written in Judah in about the mid-sixth century BCE. 24  B. R. Foster, The Epic of Gilgamesh: A New Translation, Analogues, Criticism (New York: Norton, 2001), 30–37 and 66–83.

228

Yairah Amit

from Gilgamesh should suffice to establish that the travel genre was not foreign to the ancient Hebrew reader or listener. Did the Hebrew reader know of the long journey home of Odysseus, king of Ithaca, that lasted almost ten years? Whether this work that Homer wrote in the eighth century BCE reached the ears of Hebrew narrators is hard to ascertain.25 Besides, one may even doubt whether Mesopotamian or even Egyptian travel tales influenced Hebrew writers. It appears that this narrative subgenre has met human needs from ancient times to this day, so it may appear in different cultures without any traceable sources of influence. However, even if Hebrew writers were influenced by previous travel narratives, they used this subgenre to serve their own special needs. Thus I shall not focus on comparisons but go directly to the patriarchs in Genesis. In this article I set aside the long travel narrative of the exodus from Egypt that tells of a forty-year journey from which emerged the unique identity of Israel as a people connected by covenant to its God. In any event, the sequence from Genesis to the end of the exodus journey that concludes the book of Deuteronomy, including the journey of conquest in the book of Joshua, shows that the whole Torah (Pentateuch) with the description of the conquest of the land (Hexateuch) are based on travel narratives. From now on, however, I shall focus on the travels of the patriarchs.

What Kind of Travels Characterize the Patriarchs? The journeys of the patriarchs are characterized as physical transitions from place to place at a defined time within a specific area. These stories raise the question of the extent to which physical change of place and new and different encounters might change a travelerʼs mentality or identity and affect those close to him and their common future. The many and varied physical journeys of the Genesis characters will teach us how they coped with change and their evolving identity. The skeleton supporting the stories of the patriarchs is the recurring tales about journeys of various types. Three kinds of travel occur: (1) travels of migration; (2) travels of wandering; (3) and travels to exile that always have a return, even after death. What were the journeysʼ purpose or purposes? Perhaps the journey was a way of life and one need not look for a purpose. This does not seem to be the case in the stories of the patriarchs. So we must ask: if a man has a native place, what causes him to leave it? What does he hope to achieve on the way or after reaching his destination? When the journeyʼs outset is linked to physical difficulties like hunger or threat of foreign rule, the motive may be determined 25

 See n. 11.

Travel Narratives and the Message of Genesis

229

as circumstantial, i.e., to improve oneʼs conditions or even to survive. When by contrast the motive is a royal or a divine command, the purpose is usually contained in the command itself. The heroes of Genesis set out wandering toward a destination and the question is, why? They go into exile for longer or shorter periods, but for what motive? I shall try to answer these questions by examining each of the three kinds of travel that the patriarchs undertook. Travels of Migration Abrahamʼs journey to Canaan was preceded by his father Terahʼs journey of migration (Gen 11:26–32). Terah left Ur of the Chaldeans for Canaan with his family that included Abraham, Sarai, and Lot,26 but on his way he stopped at Haran and remained there (Gen 11:31). The reason for leaving Ur of the Chaldeans and the purpose of that journey are not mentioned, nor is it clear why he decided to stay at Haran instead of proceeding to Canaan. Haran became Terahʼs place of immigration because he chose to stay there. Abraham, though he did not know in advance that he would go on to Canaan, fulfilled his fatherʼs first migration destination and went to the place his god had destined for him.27 The journeys of Terah and of Abraham are described as being for immigration, that is, a “process by which non-nationals move into a country for the purpose of settlement.”28 Usually, the decision to begin this process is conditioned by the choice of where to immigrate and the decision about when to move. Moreover, 26  Some MSs of the Septuagint add to v. 31 the words “and his son Nahor” after “his son Abram.” The Samaritan version cites the names of Nahor and Malcah after mentioning Sarai. These names appear later in Gen 22:20, 23 and 24:10, 15, 24 and 47. In this connection see also Gen 27:43 and 29:5. The additional names provide expositional data that contribute to the narrative as it continues and emphasize that Nahor, like Abraham, was a native of Haran. It seems that these additional names are exegetical harmonization; see also C. Westermann, Genesis 12–36: A Commentary (trans. J. J. Scullion; London: SPCK, 1987), 133. 27  Other texts (Gen 15:7; Neh 9:7) tell of another tradition about why God caused Abraham to leave Ur of the Chaldeans; according to this tradition, it was divine grace. In Acts 7:2–4, Abraham received the divine command in Ur and settled in Haran, and God brought him to Canaan only after the death of his father Terah. According to Yair Zakovitch, the problem of these sources lies in three traditions that combine them. The first, which is mainly related to P, presents Terah as the initiator of the journey from Ur to Canaan (Gen 11:27–32+12:4b–5); the second describes the journey as Abrahamʼs response to Godʼs test, and the point of departure is Haran (Gen 12:1–4a); in the third, the point of departure is again Ur, and it is presented as Godʼs will to redeem Israel, found in the story of the Covenant between the Pieces (Gen 15:7) and in Nehemiahʼs speech (Neh 9:7). See Y. Zakovitch, “The Exodus from Ur of the Chaldeans: A Chapter in Literary Archaeology,” in Ki Baruch Hu: Ancient Near Eastern, Biblical, and Judaic Studies in Honor of Baruch A. Levine (ed. R. Chazan et al.; Winona Lake, IN: Eisenbrauns, 1999), 429–439. 28  This definition is from the Key Migration Terms of the International Organization for Migration (IOM), see http://www.iom.int/cms/en/sites/iom/home/about-migration/key-migration-terms-1.html.

230

Yairah Amit

this decision is associated with the varied physical, social, and psychological difficulties of being a newcomer who is foreign to his new surroundings: “in the aggregate it commonly raises questions about national identities and social membership.”29 These questions are expressed in the relationship between Abraham and his offspring to the local population, the Canaanites, and in their efforts to avoid assimilation by intermarriage in the case of Isaac (Gen 24) and Esau (Gen 26:34–35, 27:46–28:9).30 Abraham expresses his choice to leave Haran in deciding to obey the divine command to “go forth,” a founding event that raises his awareness of a need for change. Following the divine command is the motive for the journey in which he must sever himself from all that is known and familiar. Abrahamʼs choice is to obey and to set out with no doubt or hesitation, or at least we are not told about such feelings. The language of the divine command, a graduated threefold presentation that reverses the continuity anticipated in a case like this (Gen 12:1),31 conveys to Abraham and the reader the difficulty of such a separation.32 The command moves from the more general (“your land”) to the more distant relatives (“your kindred”), followed by those close at hand (“your fatherʼs house”).33 The hardship of separation is accompanied by uncertainty about the destination: “To the land that I will let you see.” Easing the hardship, however, are the divine blessing and promise (Gen 12:2–3). Godʼs words and Abrahamʼs immediate response express the journeyʼs purpose, which is to leave for a new life centered on obedience and keeping faith with the God who calls him. The condition for such a life is severance and distance from the past by moving to another place. That purpose highlights the difference between Abrahamʼs journey, motivated by divine command, and his fatherʼs, which was not, or at least is not reported as 29  D. B artram et al., Key Concepts in Migration (Sage Key Concepts Series; London: Sage 2014), 4. 30  See also Gen 34. This approach characterizes later texts such as Deut 7:1–4, Ezra 9–10, Neh 13:23–30, and others. A different approach is found in Gen 38; 46:10, in Chronicles, and elsewhere. On the different approaches to foreigners and aliens in Chronicles and other writings, see S. J aphet , The Ideology of the Book of Chronicles and Its Place in Biblical Thought (Frankfurt: Lang, 1989), 334–351, and Y. A mit , “Narrative Analysis: Meaning, Context, and Origins of Genesis 38,” in Method Matters: Essays on the Interpretation of the Hebrew Bible in Honor of David L. Petersen (ed. M. J. LeMon and K. H. Richards; Atlanta: Society of Biblical Literature, 2009), 271–291. 31  On the double effect of gradation on the person of the narrated world and on the reader, see Y. A mit , “Progression as a Rhetorical Device in Biblical Literature,” JSOT 28 (2003), 3–32. Specific discussion of this verse is on pp. 5–8. 32  Gordon Wenham, like most commentators, accepts “this divine imperative as a test of faith” and criticizes Westermannʼs opinion “that for a nomad it was easy to leave land and family”; G. J. Wenham, Genesis 1–15 (WBC; Waco, TX: Word Books, 1987), 274. 33  In this case I preferred the translation of E. Fox, The Five Books of Moses (New York: Schocken, 1995); cf. von Rad, Genesis (see n. 3), 153–154.

Travel Narratives and the Message of Genesis

231

such. Terahʼs purpose could have been, for example, to live better, and this may be why he had no difficulty in changing his destination and settling in Haran. Quite differently, Abrahamʼs journey was the result of Godʼs direct message to him, which could have been a revelation or an inner revelation34 that led him to the choice of belief. Possibly that choice filled some lack for Abraham, a lack satisfied by the movement of a journey to the new place where his faith would evolve. That lack, which remains a gap in the narrative, served later writers who explained it by the idol worship in Haran35 and the possibility that Abraham was against worshipping other gods.36 In any case, because Abraham departs for a distant place chosen for him, leaving his familiar place and settling where his new identity will crystallize and where he will live out his faith, migration is the purpose of his journey. According to the above definition of immigration, all of the matriarchs were immigrants. Sometimes they simply accompanied their husbands or their masters, as did Sarah and the two maidservants, Bilhah and Zilpah. Other times they chose to leave their native place, as did Rebekah, Rachel, and Leah. Looking for Rebekahʼs motive to emigrate, it seems that she was impressed by the jewelry and other gifts she received from Abrahamʼs emissary (Gen 24:22, 34, 47, 53). She was young and possibly she was also motivated by curiosity and by the wish to change something in her life. Rachel and Leah identified themselves with their husband and protested against their father, his attitude toward them, and his attempt to abuse and exploit their husband (Gen 31:1–16). In other words, these three matriarchs seem to have been interested in improving their circumstances, especially their economic circumstances. Whatever their motives, the description of all the matriarchs as immigrants highlights their participation and importance in creating the new nation and emphasizes the status of Abraham as the only one who immigrated because of religious motivation. Travels of Wandering Now we explore the two other travel narratives in Genesis involving Abraham and his descendants. Abrahamʼs arrival in Canaan does not halt his journeying;  See Skinner, Genesis (see n. 5), 491; H. Gunkel, Genesis (trans. M. E. Biddle; Macon, GA: Mercer University Press, 1997), 439; von Rad, Genesis (see n. 3), 154. Westermann Genesis 12–36 (see n. 26), 147, notes that “there is no account of a revelation; the stylization shows that J does not intend to narrate an event, but to pass on a divine oracle whose situation he leaves open.” 35  See Josh 24:2: “Then Joshua said to all the people, ‘Thus said the Lord, the God of Israel: In olden times your forefathers – Terah, father of Abraham and father of Nahor – lived beyond the Euphrates and worshiped other gods.’ ” 36  See the rabbinic midrashim about how Abraham treated the idols, collected in L. G inzberg , The Legends of the Jews (trans. H. Szold; 7 vols.; Philadelphia: The Jewish Publication Society of America, 1942), 1: 209–215; 5:217–218, and nn. 49–50. 34

232

Yairah Amit

it continues to mark his life and the life of his descendants. His other journeys, however, are no longer travels of immigration but belong to other subpatterns. One of these subpatterns, travels of wandering, is frequently assumed to reflect the wanderings of nomadic pastoralists.37 However, these are not nomadic wanderings because Abrahamʼs groupʼs nomadic situation is temporary and their real wish is the promised land.38 Therefore the question is what the patriarchsʼ wanderings signify. Wanderings involve choosing a permanently unsettled lifestyle, as do Bedouins, gypsies, or lately even high-tech experts and artists. Abrahamʼs motive for wandering in Canaan is the result not of lifestyle choice but of divine command: “Up, walk about the land, through its length and its breadth, for I give it to you” (Gen 13:17). The patriarchs wander in their destined land not as typical nomads or seminomads seeking pastures for their flocks but as travelers or even tourists who have to reach specific towns. The patriarchs mark their destinations by ritual acts and sometimes remain in their vicinity. Also, “walking about the land” is an act that symbolizes taking possession of it.39 In other words, Abrahamʼs wanderings and the wanderings of his descendants appear to have both cause and purpose. A close examination of the patriarchsʼ travels shows that the journeys had three purposes: 1. Getting to know the land and its boundaries. Abraham reaches Dan in the north (Gen 14:14) and Beersheba in the south, where he plants a tamarisk tree (21:33) and settles down for a period (22:19). In these journeys he appears to be literally determining the borders of the land appropriate to the geographic merism: “from Dan to Beer-sheba.”40

37  Thomas Barfield explains: “[they] live in societies in which the husbandry of grazing animals is viewed as an ideal way of making a living and the regular movement of all or part of the society is considered a normal and natural part of life”; T. Barfield, “Pastoral Nomads or Nomadic Pastoralists,” in The Dictionary of Anthropology (ed. T. Barfield; Malden, MA: Blackwell, 1997), 348–349. 38  See J. Van Seters, Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975), 13–38, esp. p. 38: “These journeys can hardly be construed as nomadic migrations. [. . .] There is nothing in this presentation of the ‘nomadic’ patriarchs which is inappropriate to the portrayal of pastoral life in the period of the late Judean monarchy or exilic periods, but there is much that speaks against the choice of any earlier period.” 39  N. Wazana, All the Boundaries of the Land: The Promised Land in Biblical Thought in Light of the Ancient Near East (Winona Lake, IN: Eisenbrauns, 2013), 92. 40  See Judg 20:1; 1 Sam 3:20; 2 Sam 3:10, 17:11, 24:2, 15; 1 Kgs 5:5; 1 Chron 21: 2; and 2 Chron 30:5. Dan is also the northernmost spot that Moses sees (Deut 34:1). According to Zecharia Kallai, Dan and Beersheba are sacred centers in the north and south and therefore actually they do not designate geographical borders – the border is beyond them. Z. Kallai, Historical Geography of the Bible: The Tribal Territories of Israel (Jerusalem: Magnes, 1986),

Travel Narratives and the Message of Genesis

233

2. Possessing the land at the stage when “the Canaanites were then in the land” (Gen 12:6b). Possession here is mostly symbolic and is manifested by walking through the length and the breadth of the land 41 and by purchasing tracts of land. Abraham bought land in Hebron for a family burying ground (Gen 23),42 and Jacob purchased land in Shechem (Gen 33:19).43 3. Abrahamʼs journeys have a clear and visible faith-oriented aspect. His first sacred station is the site of Shechem (Gen 12:6–7). From there he moves to the hill country east of Bethel, where he builds another altar to the Lord and invokes him by name (Gen 12:8). After coming back from Egypt and separating from Lot he reaches Hebron, where he settles (Gen 13:18) and is granted several revelations.44 He builds an additional altar at Mount Adonai Yera‍ʾeh, a name that hints at Jerusalem.45 That series of ritual acts structures Abrahamʼs religious identity, and this identity continues in Isaac his son and Jacob his grandson according to the principle that characterizes the patriarchal stories: “What happened to the patriarchs is a sign for their children.”46 It appears then that in his wanderings Abraham

309. For the use of this expression designating the territory of Israel, see Wazana , All the Boundaries of the Land (see n. 39), 58–82. 41  According to Umberto Cassuto the patriarchs symbolize “the ideal conquest”; U. Cassuto, A Commentary on the Book of Genesis, Part II: From Noah to Abraham, Genesis V19–XI32 (trans. I. Abrahams, Jerusalem: Magnes, 1964), 301. On p. 305 he adds: “the Book of Joshua portrays for us the actual subjugation in a manner paralleling the ideal conquest by the patriarchs – even the wording is similar – as though to say, the possession of the land gained in the days of Joshua was already implied, in essence, in the symbolic conquest that the first patriarchs had effected in their time, and that it was all predestined and foretold from the beginning in accordance with the Lordʼs will.” 42  See also Gen 25:9–10; 49:29–32; 50:13. 43  See Josh 24:32. 44  These revelations are meaningful to the crystallization of his religious identity and are linked to the covenant between the pieces (Gen 15), the circumcision covenant (Gen 17:1–22), and tidings of the son to be born (Gen 18:1–33). 45  Jerusalem is not mentioned in the book of Genesis nor in the entire Torah. See Y. Amit, “Is Jerusalem also an Issue for Polemic?” in Hidden Polemics in Biblical Narrative (trans. J. Chipman; BibInt 25; Leiden: Brill, 2000), 131–168. 46  The rabbinic Sages in Tanhuma, Lekh lekha 12, said: “The Holy One transmitted a sign to Abraham that whatever had happened to him would happen to his children.” See Midrash Tanḥuma (S. Buber Recension), vol. 1, Genesis (trans. J. T. Townsend; Hoboken, NJ: Ktav, 1989), 72. Ramban in his commentary on Gen. 12: 6 adds: “It is indeed a great matter, which our Rabbis mentioned briefly, saying ‘Whatever has happened to the patriarchs is a sign to the children,’ It is for this reason that the verses narrate at great length the account of the journeys of the patriarchs, the digging of the wells, and other events. Now someone may consider them unnecessary and of no useful purpose, but in truth they all serve as a lesson for the future: when an event happens to any one of the three patriarchs, that which is decreed to happen to his

234

Yairah Amit

prepares the ground for his descendants to settle, be buried in, and retain and strengthen their religious identity. Abrahamʼs wanderings are not described as pasturing the flocks, like those of Josephʼs brothers who reached Shechem (Gen 37:12), nor as a search like that of Joseph, who left the Hebron valley and found his brothers at Dothan (Gen 37:13–17).47 Rather, Abrahamʼs wanderings determine the land holdings and prepare the ground for serving God who brought him from Haran. Isaacʼs life, too, returns to the mode of wandering.48 He roams from Beer Lahai Roi in the desert (Gen 25:11) until he reaches Beersheba, where a revelation is granted him. There he calls on the name of God and builds an altar (Gen 26:23–25). Similarly, Jacob is vouchsafed a revelation at Bethel, on his way to Haran, sets up a stone for a pillar, and makes a vow (Gen 28:10–22).49 On his return journey he sets up a pillar at Mizpah (Gen 31:45) and makes a sacrifice there as well (Gen 31:54).50 Before crossing the Jordan, he is granted two revelations: at Mahanaim (Gen 32:2–3)51 and at Penuel (Gen 32:22–33).52 At Shechem, after buying a parcel of land, he erects an altar and calls it El children can be understood.” See Nahmanides, Commentary on the Torah (trans. C. B. Chavel; 5 vols.; New York: Shilo, 1971–1976), 1:169. 47  These examples belong to a Joseph story that was first independent and then added to the patriarchsʼ cycles, as discussed in the prologue of this work. 48  The patriarchs have been depicted as semi-nomads, but Isaac is also shown as a landholder in Gen 26:12. Therefore G unkel , Genesis (see n. 34), 295 argues: “Farming, rarely explicitly attributed to the patriarchs [. . .] appears here as a particular exception, but is also occasionally undertaken by nomads (analogies in Dillmann) [. . .] Here the element, however, is only a means for the narrator to avoid a dilemma: he must explain Isaacʼs wealth somehow.” Jon Levenson writes in J. D. L evenson , “Genesis: Introduction and Annotations” in The Jewish Study Bible (ed. A. Berlin and M. Z. Brettler; 2nd ed.; New York: Oxford University Press, 2014), 50: “The image of Isaac as a landholder and farmer (v. 12) is in tension with the picture of him as a semi-nomad (cf. v. 17).” See my discussion in this paper and in Van Seters, Abraham in History and Tradition (see n. 38). 49  On the special place of this dream-revelation in the book of Genesis and beyond, see Y. I. Peleg, Going Up and Going Down: A Key to Interpreting Jacobʼs Dream (Gen 28.10–22) (trans. B. Rozen; London: Bloomsbury, 2015). 50  Mizpah is mentioned in Judg 10–11 in the context of Jephthahʼs war with the Ammonites. The city was a holy center (Judg 11:11). As to the problems of its identification, see I. F inkelstein et al., “The Biblical Gilead: Observations on Identifications, Geographic Divisions and Territorial History,” in UF 43 (2012), 141–143, and the bibliography there. 51  Mahanaim was an important city east of the Jordan. Its identification is not clear. See history and suggestions in R. A. Coughenour, “A Search for Mahanaim,” BASOR 279 (1989), 57–66 and Finkelstein et al., “Biblical Gilead” (see n. 50), 146–148 and the bibliography there. 52  For further information on Penuel as a sacred place, see M. Arneth, “Der Gotteskämpfer am Jabbok,” ZABR 14 (2008), 350–364 and J. B arton , “Jacob at the Jabbok,” in Die Erzväter in der biblischen Tradition. Festschrift für Mattias Köckert (eds. A. C. Hagedorn and H. Pfeiffer; BZAW 400; Berlin and New York, 2009), 187–195. For the dispute about

Travel Narratives and the Message of Genesis

235

Elohei Israel (Gen 33:18–19), buries the alien gods under the oak that was near Shechem (Gen 35:2–4),53 and continues to Bethel. There he builds another altar, calling the place El Bethel, “for it was there that God had revealed himself to him when he was fleeing from his brother” (Gen 35:7). Deborah, Rebekahʼs nurse, was buried there under the oak below Bethel (Gen 35:8), and Jacob is again granted a revelation in Bethel, where he sets up a pillar, offers a libation on it, and pours oil upon it (Gen 35:9–15). From Bethel Jacob proceeds, as his grandfather did, to Hebron, but on the way he has to bury Rachel on the road to Ephrath – now Bethlehem54 – and on her grave he sets up a pillar (Gen 35:16–21). When he comes to Hebron he meets his father, but after some time he has to bury him, too (Gen 35:27–29). In the time of severe famine Jacobʼs sons, like Abraham and Isaac, go abroad to obtain food (Gen 42:1–47:27). In the last wandering journey of Jacob and his sons to Egypt, Jacob stops in Beersheba, offers sacrifices to the God of his father Isaac, and has one more revelation (Gen 46:1–4).55 The purpose of this concise survey is to show that the wandering travels of the patriarchs are usually to the same places where they perform similar acts, like building altars. The wanderings thus construct affinity to the land whether by land purchase, by burial places, by consecrating ritual sites, or by returning to them. In the wanderings beyond the borders of the land there are no ritual acts. Moreover, the model of three patriarchs who have similar wandering travels creates a family memory that is the basis for national history. These travels may be called typological routes: “the route followed by Abraham becomes a typological one, to be followed by his descendants.”56 The authors/editors57 are the identification see F inkelstein et al., “Biblical Gilead” (see n. 50), 148–149, and the bibliography there. 53  On the ambivalent attitude toward Shechem in Biblical literature, see Y. Amit, “How to Slander the Memory of Shechem,” in Memory and the City in Ancient Israel (eds. D. V. Edelman and E. Ben Zvi; Winona Lake, IN: Eisenbrauns, 2014), 231–243. 54  About the location of Rachelʼs tomb, see N. Na‍ʾaman, “The Settlement of the Ephrathites in Bethlehem and the Location of Rachelʼs Tomb,” RB 121 (2014), 516–529 and the bibliography there. 55  Some scholars, with whom I agree, argue that this episode is secondary to the Joseph story. See, e.g., D. B. R edford , A Study of the Biblical Story of Joseph (Genesis 37–50) (VTSup 20; Leiden: Brill, 1970), 19; G. W. Coats , From Canaan to Egypt: Structural and Theological Context for the Joseph Story (CBQMS 4; Washington, DC: Catholic Biblical Association of America, 1976), 48; Westermann, Genesis 37–50 (see n. 7), 154, also sees in 46:1–5 as “part of the patriarchal story.” 56  See Y. A mit , Reading Biblical Narratives: Literary Criticism and the Hebrew Bible (trans. Y. Lotan; Minneapolis: Fortress Press, 2001), 124. 57  My assumption is that “it is difficult to distinguish or to separate in an absolute way the role of the editor and that of the author during the prolonged process of composition, transmission and editing of the biblical work. The knowledge that different editors altered, deleted, added, innovated and corrected this work, explain why it is today impossible to separate the

236

Yairah Amit

not interested in the wanderings themselves, or in the roads, or the difficulties of the people who traveled them. They are interested mainly in those places that were the purpose of the wanderings and the events that occurred during them, mainly rituals that are of importance to Israelite belief. In conclusion, the purpose of the wanderings is ideological. They do not reflect the life of the ancestors; they reflect the means to keep and preserve the connection to the land and its holy places. These travels and wanderings do not shed light on the reality or the history of the patriarchs, but they do show the thoughts and beliefs of those editors who “sent” them on these similar and repeated wanderings. Travels to and from Exile Exile of an individual or a group is an enforced transfer to another country due to natural causes like famine or human causes like war.58 Exile can be temporary or long lasting. Generally it ends with assimilation in the land of exile and loss of affinity with the exilesʼ place of origin. The deportees then retroactively become immigrants, like those exiled by the Assyrians from Israel59 (in contrast to those from Judah in the Babylonian period).60

original work of the author from the handling of the editor” (Amit, Judges [see n. 4], 15). See also J. Barton, “Redaction Criticism” in ABD 5:644–647. 58  According to the IOM terminology (see n. 28), exile can be defined as forced migration: “A migratory movement in which an element of coercion exists, including threats to life and livelihood, whether arising from natural or man-made causes.” See also J. J. A hn , Exile as Forced Migrations: A Sociological, Literary, and Theological Approach on the Displacement and Resettlement of the Southern Kingdom of Judah (BZAW 416; Berlin: de Gruyter, 2011). Often biblical scholars connect exile only with military punishment, for example, D. R om S hiloni , “Deuteronomic Concepts of Exile Interpreted in Jeremiah and Ezekiel,” in Birkat Shalom: Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Judaism presented to Shalom M. Paul on the Occasion of His Seventieth Birthday (eds. C. Cohen et al., Winona Lake, IN: Eisenbrauns, 2008), 102. 59  More details about the Assyrian deportations from the north are discussed below. Biblical literature does not include information about the deportees form the north. In the apocryphal midrash The Assumption of Moses it is told that they were assimilated: “And the ten tribes will increase and multiply among the Gentiles during the time of their captivity” (4:9); R. H. Charles, The Assumption of Moses (London: Adam and Charles Black, 1897), 16. According to the Sages too (b. San 110:2), those deportees are not expected to return, reflecting the assumption that they were assimilated in Assyria. No wonder they were the inspiration for many legends (see, e.g., b. San 94:1). On the disappearance of the ten tribes see, e.g., S. Talmon, “Exile and Restoration in the Conceptual World of Ancient Judaism” in Restoration: Old Testament, Jewish and Christian Perspectives (ed. J. M. Scott; JSJSup, 72; Leiden: Brill, 2001), 119–121 and B. Oded, The Early History of the Babylonian Exile (8th –6th Centuries B.C.E.) (Haifa: Pardes, 2010), 451–458 (Hebrew). 60  Further information about the Judean exiles can be found below. See also Ahn, Exile as Forced Migration (see n. 58).

Travel Narratives and the Message of Genesis

237

The exilic travel of the patriarchsʼ experience was forced, but exile is depicted not only as a threatening place – it is an advantageous one as well. In Genesis there are six examples of exilic travel experienced by the patriarchs: 1./2. Famine exiles Abraham to Egypt (Gen 12:1). After a time, without any stated reason, he returns and then is exiled to Gerar (Gen 20:1), the land of the Philistines,61 where he remains for a long time (Gen 20:15; 21:34), although he is not forced to go there. In both places his wealth increases (Gen 12:16, 13:1–2, 20:14). 3. Isaac is also exiled by famine to Gerar (Gen 26:1). His plan appears to be to go down to Egypt, but a vision of God changes that (Gen 26:2). He too remains in Gerar for a long time, tills the land, and like his father digs wells and increases his wealth (Gen 26:12–22). 4. Jacob is exiled to Haran for two reasons: fear of his brother Esauʼs vengeance and to find an Aramean bride who would please his parents, rather than a Canaanite bride (Gen 27:41–28:5).62 Although his mother thought the exile would be for a short while (Gen 27:43–45), Jacob remained in Haran for a long time (Gen 31:38, 41). After twenty years, he returned with the family he established there and the wealth he had achieved. 5. Joseph was exiled to Egypt and was sold there as a slave (Gen 37:28, 39:1), but after some time he rose to greatness in the land of his exile (41:38–57). Thus Joseph was able to prepare the way for the long sojourn of his father, his brothers, and their families. On his deathbed he declared that the children of Israel would one day go up from Egypt back to their promised land, and he asked them to take his bones with them for burial in Canaan (Gen 50:24–25).63 6. In his old age Jacob had to go into exile again, this time with his entire family, who wanted to leave Canaan because of the severe famine and the good conditions they were promised by Joseph and Pharaoh (Gen 45:16–20). On the way a revelation came to Jacob at Beersheba legitimizing the descent into Egypt and even promising a return to the land (Gen 46:1–4). From the outset, then, this was exile, not emigration or 61  For further information about Gerar see E. M. Yamauchi, “Abraham and Archeology: Anachronism or Adaptations?” in Perspectives on Our Father Abraham (ed. S. A. Hunt; Cambridge: Eerdmans, 2010), 15–32, esp. 25–28. 62  See p. 230, above, and n. 30 there. 63  Cf. Jacobʼs request from Joseph (Gen 47:29–31) and from his sons (Gen 49:29–32). Many scholars are convinced that these verses (Gen 50:24–25) are secondary to the Joseph story and represent a late layer of Torah editing. See, e.g., R. Rendtorff , Das überlieferungsgeschichtliche Problem des Pentateuch (BZAW 147; Berlin: de Gruyter, 1977), 75–79. Westermann, Genesis 37–50 (see n. 7), 208–210 argues that these verses connect the Joseph story to the patriarchsʼ stories and to the story of the Exodus. See also J. Van Seters, Prologue to History: The Yahwist as Historian in Genesis (Louisville: Westminster John Knox, 1992), 323.

238

Yairah Amit

wandering. Jacob died in Egypt but was buried in his ancestral grave in Hebron (Gen 49:29–33). His descendants, as foretold in the covenant between the pieces, would remain in Egyptian exile for four hundred years (Gen 15:13). Despite the difficulties, then, the patriarchsʼ journeys of exile end successfully, and it is not right to examine them through the eyes of Deuteronomy, where exile is depicted as a horrible punishment (Deut 4:27–28; 28:36–37, 64–68; 29:21–27; 30:1–7). According to Deuteronomy the punishment of exile is expressed not only by physical conditions but also by “the double loss of Israelʼs identity: Dispersion of the population dissolves its political identity, and idol worship dissolves its religious identity.”64 As opposed to these descriptions, the situation in the patriarchal narratives is different, and there is some allure in their exilic situation. After overcoming the initial period of being migrants, the patriarchs became established, and from an economic point of view they even succeed in exile. The travels of exile contain on the one hand the threat of oppression and assimilation through loss of independent ethnic entity, but on the other hand they offer tempting chances of material success without loss of unique identity. Stories of the father and the sister-wife (Gen 12:10–13:2; 20:1–18; 26:1–14) highlight the two sides of the issue.65 With the threat of taking away the wife, hinting at assimilation, there is also the return with great wealth or economic success in the place of exile, as enjoyed by Abraham, Isaac, Jacob, and Joseph. Furthermore, the patriarchs learn to know that their God does not forsake them but protects them in exile too, and they pass this faith on to their descendants. The result is that the identity of the forefathers as believers is shaped in exile. The story of the encounter between Jacob and Laban in Gilead is a declaration of Jacobʼs Israelite identity, separate and different not only from that of the Canaanites but also from that of Nahorʼs family in Haran, whose daughters are Jacobʼs wives. The Gilead encounter (Gen 31:25–54) after Jacobʼs decision to leave his place of exile and return to Canaan emphasizes both Jacobʼs separation from Haran and the signs of Israelite identity he preserves throughout those many long years of exile. These signs are:66 64  B. M. L evinson , “Deuteronomy: Introduction and Annotations” in The Jewish Study Bible (ed. A. Berlin and M. Z. Brettler; 2nd ed.; New York: Oxford University Press, 2014), 411. See further Rom-Shiloni, “Deuteronomic Concepts of Exile” (see n. 58), 103–109. She concludes that “The major deviation between Deuteronomy and the prophetic books rests on what may be the literary character of Deuteronomy or its preexilic perspective, presenting exile as a total and inclusive event” (122). 65  Wenham, Genesis 1–15 (see n. 32), 291, following Cassuto, Genesis (see n. 41 above), 336, agrees that these parallel stories “foreshadow in the tales of the Patriarchs the history of their descendants.” Both relate only to the exile in Egypt and not to the later exiles. 66  See D. R om -S hiloni , “When an Explicit Polemic Initiates a Hidden One: Jacobʼs Aramaean Identity,” in Words, Ideas, Worlds: Biblical Essays in Honour of Yairah Amit (ed. A. Brenner and F. Polak; Sheffield: Sheffield Phoenix, 2012), 206–235.

Travel Narratives and the Message of Genesis

239

1. Belief in the existence of God wherever Jacob stays and loyalty to him, not to Nahorʼs god, who is the god of Laban and is represented by the teraphim, the household idols (Gen 31:29–30, 34–35, 42, 49, 53). 2. Keeping and not repressing memories of the past, meaning loyalty to Jacobʼs legacy, which is the legacy of his father, Isaac. This can be seen from the way Jacob tells his wives about the revelation he had (Gen 31:3). The short revelation becomes a full speech (Gen 31:5–13) in which he mentions the god of his father (Gen 31:6) and explains that what happened with the flocks resulted from the intervention of this god (Gen 31:7–10). He tells them about the revelation at Bethel, about the commitment to his vow, and about the connection to his native land (Gen 31:11–13). With these stories he continues to transfer his legacy to the members of his family. 3. Preserving his language. Though Jacob lives many years in an Aramean area, he calls the pillar gal ed, which is a Hebrew name, and not yegar shahadutha, the Aramaic name Laban uses (Gen 31:47). 4. Respecting geographical boundaries at Mizpah in Gilead. Jacob accepts that the mound is a border point between his territory and Labanʼs (Gen 31:51–52).67 Laban the Aramaean (Gen 31:20, 24) regards his daughters who married Jacob as outsiders belonging to a different place (Gen 31:14–16). These episodes show that Jacob preserves his Israelite identity and imparts it to those who accompany him, as also indicated by burying the remnants of idolatry in Shechem (Gen 35:2–4),68 hinting back to the teraphim Rachel concealed. Two questions now arise: Why were the editors of Genesis not content with accounts of immigration and wanderings and so added this third exilic type of travel? Why did the editors decide to repeat the travels of exile six times? In my opinion the answer to the first question is that they were interested in many journeys of exile to highlight the idea of returning from exile. All the patriarchs are portrayed as returning from exile, if not in their lifetime then after their deaths, as it is with Jacobʼs second exile (Gen 49:29–50:13) and, after hundreds of years, with Joseph (Josh 24:32, following Gen 50:24–26 and Exod 13:19). The message is clear: exile is a valid option, but the evolving attitude to it in the patriarchal stories emphasizes that exile always continues for a limited time and that from every exile, distant or close by, one must return to the promised land. The second answer relates to the number of experiences of exile. As shown above, six exiles are mentioned in the patriarchal stories and the people of Israel 67  “This story reflects an Iron II reality of the border between Israelite and Aramean populations that lived in proximity in northern Transjordan”; Finkelstein et al., “Biblical Gilead” (see n. 50), 141. 68  See n. 53.

240

Yairah Amit

underwent six exiles from the eighth to the sixth centuries BCE. The six exiles in the patriarchal narratives stories echo the six historical mass deportations:69 1. Exile of the Northern Kingdom of Israel that included Gilead and Galilee in 733–732 BCE by Tiglath-Pileser III (2 Kgs 15:29; Judg 18:30; 1 Chr 5:6; 26).70 2. The exile of Samaria in 722–720 BCE by Sargon II, who ended the siege of Shalmaneser V. Sargon is also responsible for turning Samaria into an Assyrian province (2 Kgs 17:1–6; 18:9–12).71 3. Exile by Sennacherib in 701 BCE of the southern Judean cities, including the cities of the Shephelah, during the reign of Hezekiah. There is no account of this in the Bible save for the verse about capturing all the fortified cities of Judah (2 Kgs 18:13; 2 Chr 32:1, 9) and a hint about the intention to exile in Rabshakehʼs speech (2 Kgs 18:32). Reports come from Sennacheribʼs annals72 and reliefs from his palace at Nineveh.73 It seems that the biblical editors were so impressed that Jerusalem was not destroyed that they ignored and did not mention explicitly the destructive results of Sennacheribʼs campaign in Judah.74 4. Exile of Jehoiachin by Nebuchadnezzar II (2 Kgs 24:8–17; 2 Chron 36:9–10; Jer 22:24–30; 24:1; 27:20; 28:4; 29:2; Ezek 1:1, 17:12; Esth 2:5–6) in 597 BCE.75 Nebuchadnezzar “impaired the kingdomʼs economic  It is always possible that one mass deportation was actually some groups that were taken separately. 70  See N. Na‍ʾaman, “Population Changes in Palestine Following Assyrian Deportations,” Tel Aviv 20 (1993), 104–106; K. L. Younger, “The Deportations of the Israelites,” JBL 117 (1998), 201–227; and J. K iefer , Exil und Diaspora (Leipzig: Evangelische Verlagsanstalt, 2005), 51–54 and the bibliography there. 71  On problems of who conquered Samaria and when, see N. N a ‍ʾaman , “The Historical Background to the Conquest of Samaria (720 BC),” Biblica 71 (1990), 209–220; idem, “Population Changes in Palestine,” (see n. 70), 106–108; Younger, “Deportations of the Israelites,” (see n. 70); and B. Becking, The Fall of Samaria: An Historical and Archaeological Study (SHANE 2; Leiden: Brill, 1992). 72  D. D. L uckenbill , The Annals of Sennacherib (The University of Chicago Oriental Institute Publications 2; Chicago: University of Chicago Press, 1924), 32–34. On varied issues relating to Sennacheribʼs campaign, see L. L. Grabbe (ed.), Like a Bird in a Cage: The Invasion of Sennacherib in 701 BCE (JSOTSup 363; Sheffield: Sheffield Academic, 2003). 73  For example, see the reliefs of the Lachish deportees in D. Ussishkin, The Conquest of Lachish by Sennacherib (SMNIA 6; Tel Aviv: Tel Aviv University Institute of Archaeology, 1982), 84–85. 74  About the atmosphere in the remnant of Judah we can learn not only from historiography (2 Kgs 19:35–37// Isa 37:36–38// 2 Chron 32:21–23) but from prophecy too, as in, e.g., Isa 9:1–6; 37:21–32// 2 Kgs 19:22–32. 75  See A. K. Grayson, Assyrian and Babylonian Chronicles (Locust Valley, NY: J. J. Augustin, 1975), 102. See also A. Malamat, “The Twilight of Judah,” in Israel in Biblical Times: Historical Essays (Jerusalem: The Bialik Institute and The Israel Exploration Society, 1983), 69

Travel Narratives and the Message of Genesis

241

strength and deported part of the nationʼs social, religious, military, and economic elite.”76 Jehoiachin is also mentioned in the epilogue of the book of Kings, where the positive change in his status in exile is related (560 BCE). King Evil-merodach, the son of Nebuchadnezzar II,

released him from prison. He spoke kindly to him, and gave him a throne above those of other kings who were with him in Babylon. His prison garments were removed, and [Jehoiachin] received regular rations by his favor for the rest of his life. A regular allotment of food was given him at the instance of the king – an allotment for each day – all the days of his life (2 Kgs 25:27–30; cf. Jer 52:31–34).

There are similarities between Jehoiachinʼs release from prison in Babylon and Josephʼs release from prison in Egypt. In both stories prison clothes were replaced by fine apparel, and both enjoyed a high position at the royal court and personal contact with the king. This connection is also strengthened by the common expression “raised the head of” (‫ ;נשא ראש‬2 Kgs 25:27; Gen 40:13, 20), which in light of the Akkadian we understand in this context as “took note of,” meaning to remember. Thus both were able to look after their brethren in exile. It seems that all this is not coincidental and that a hidden polemic in the Joseph story hints at Jehoiachin.77 5. Nebuchadnezzar II in 586 BCE exiled Zedekiah and most of the Judeans and destroyed Jerusalem and its temple (2 Kgs 25:1–21; Jer 1:3; 38:28–40:1; 52:1–30; 2 Chr 36:17–20).78 6. After the murder of Gedaliah ben Ahikam, some people who feared punishment by Babylon went into exile in Egypt (2 Kgs 25:25–26; Jer 41:16–18). According to Jer 52:30 and Josephus (Ant. 10:181–183),79 some were also exiled by the Babylonians in the twenty-third year of Nebuchadnezzarʼs reign (582 BCE).80 263–265 (Hebrew). Malamat solves textual problems and inconsistencies by suggesting that this deportation had two stages. 76  O. Lipschits, The Fall and Rise of Jerusalem: Judah under Babylonian Rule (Winona Lake, IN: Eisenbrauns, 2005), 55. 77  See the doctoral thesis of my student O. Biderman, “The Story of Joseph: Gen. 37–50: Its Time, Context and Purpose” (PhD diss., Tel Aviv University, 2011), 240–257 (Hebrew); and see also pp. vii–viii of the English abstract. 78  There is no evidence of this crucial event in the inscriptions of Nebuchadnezzar II, nor in the chronicle of his campaigns till 594 BCE. Cf. Oded, Early History of the Babylonian Exile (see n. 59), 36; Lipschits, Fall and Rise of Jerusalem (see n. 76), 75 and n. 135. Therefore the sources for reconstruction are mainly biblical and also archaeological excavations. 79  See P. R. Ackroyd, Israel under Babylon and Persia (London: Oxford University Press, 1970), 36–37. According to Lipschits, Fall and Rise of Jerusalem (see n. 76), 119–122, the group of people who went down to Egypt was relatively small. 80  Although no reason is given for the deportation by the Babylonians, Lipschits, Fall and Rise of Jerusalem (see n. 76), 122 is convinced that “this event should not be linked to the murder of Gedaliah.”

242

Yairah Amit

It seems then that the numerous journeys into exile in the patriarchal stories have a purpose: to convey the idea that what happened to the patriarchs is a sign for their children. As the fathers were exiled, so the sons are expected to be exiled; as the universal God was with the fathers in exile, so he will be with the sons; and as God brought their fathers back from exile to their promised land, so he will do with the sons. The historical experience of exiled populations from northern Israel, assimilated and lost, accentuated the motif of return from exile, so prominent in the stories of the forefathers and in the history of the Jewish people.

Conclusion At the beginning of this article I presented some standard characteristics of the travel narrative subgenre. Close examination of the three types of travels the patriarchs experienced demonstrates that most of these defining characteristics are absent from the travel stories of the forefathers. The subgenre provides only the skeleton of the plot on which to hang the stories that together explain how the patriarchsʼ identity was crystallized vis-à-vis God and humans. In conclusion, the formulation of Genesis as a book of journeys of migration, of wanderings, and especially of exile is functional. The journeys are not simply a technical means for linking the bookʼs varied stories. The journeys have theological, national, and historical purposes. The authors are not concerned with describing the journeys for their own sakes, and difficulties of the road are not discussed. The conduct of the patriarchs when their destination was reached, their developing attachment to the land, and their awareness of their new identity are the emphases of the stories. The journeys serve the redaction process in establishing the idea of Israelʼs right to the land of their fathers. The book shows that that idea is based not only on the divine promise but also on physical contact expressed in journeys through the land, land purchases, consecration of sacred sites, and, in addition, by establishing journeys of return to the land at any time, from any land, as a purpose and goal. The theme of exile as related to the experience of the Israelites and Judahites confirms my argument that the principal redaction of Genesis took place after the Babylonian exile, and to be more exact, in the Persian period. The bookʼs redaction only makes sense once the possibility of return to the land of Israel was opened (538 BCE).81 The need to convince deportees to return to Judah then became a meaningful issue. Second Isaiah did it in his prophetic way, and the book of Genesis did it by means of travel narratives. Thus Genesis becomes the credo of Zionism, and one must be deeply impressed by the editorial work of the ancient authors who succeeded in formulating a message that remains relevant after thousands of years. 81  This conclusion has become increasingly convincing; see for example K. Schmid, “Genesis in the Pentateuch,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; VTSup 152: Leiden: Brill, 2012), 27–50.

Why Is the Pentateuch Unreadable? Or, Why Are We Doing This Anyway? Joel S. Baden I invite the reader to imagine for a moment a Pentateuch very different from the one we have now. Imagine a text that read as fluently as a Jane Austen novel – a text that told the sweeping epic of Israelʼs history from creation through the patriarchs through the exodus and the wilderness up to the death of Moses. Imagine a text in which the broader themes and the smallest details of plot were all in perfect agreement, in which the language was consistent but not redundant, in which the theology was definable and yet also realistically complex.1 What could biblical scholars do with such a text? Almost everything, really: scholars could inquire as to the historicity of the events it describes; could explore its use of rhetoric; could try to use it as a datum for understanding the history of Israelite religion; could approach it from every possible postmodernist angle, be it feminist, queer, postcolonial, black, liberationist . . .; could investigate the constituent elements of the narrative with an eye toward identifying what possible earlier materials, be they oral or written, the author may have used in composing this text. One approach that would not be justified would be the attempt at any literary-historical analysis. Why, with such a text, would anyone ever think to divide it into sources, or layers, or redactions, or glosses? Source criticism – and here I use the word in its broadest sense, to include the sort of literary-historical textual surgery that most every pentateuchal scholar performs to a greater or lesser degree – is fundamentally unlike other biblical methods, such as form, rhetorical, feminist, or postcolonial criticism. Most critical methods are applicable to any text, not only biblical texts. One could apply any of these methods to, say, Jane  Although there are, of course, scholars who maintain that the Pentateuch can be read as a unified text, they do so often from intellectual positions that require such a claim: either theological, as in the classic anticritical works of W. H. Green, The Higher Criticism of the Pentateuch (New York: Charles Scribnerʼs Sons, 1902) or U. C assuto , The Documentary Hypothesis (Jerusalem: Magnes, 1959), from the Christian and Jewish perspectives, respectively; or literary-critical, taking the initial position that the form of the text to be interpreted is the final form, as in the case of T. Mann, The Book of the Torah: The Narrative Integrity of the Pentateuch (Atlanta: John Knox, 1988). Even these scholars, however, do not maintain that the Pentateuch is easy to read or untroubled by contradictions and doublets; they simply have different sorts of explanations for these phenomena. 1

244

Joel S. Baden

Austen (and one can be sure that enterprising English PhDs across the world have done so). Source criticism, however, cannot – or at the very least should not – be used on any and every text.2 In large part this is because, despite its common name, source criticism is not, in fact, a critical method at all. There is no broad theoretical basis that can be shared by other disciplines, even though other disciplines may do something similar. One could not write an abstracted description of source criticism without reference to the text as one could with, say, form criticism.3 That is, there is no source-critical approach that is common to all literature, or even to all ancient literature, or even to all ancient Israelite literature. The literary-historical investigations of Homer, or of the Pali canon of Buddhism, or of the hadith literature of Islam, are all distinct in nature from one another and from the literary-historical investigation of the Pentateuch.4 They may all be source-critical, but one could not derive a handbook of source criticism from their commonalities. Source criticism is not a method or an approach. It is, more accurately, the name we give to a set of results – in the case of the Pentateuch, to the common scholarly conclusion that it is the product of multiple hands. These results, this conclusion, emerge from a particular question, one that it is not feasible to ask of just any text, namely, why is this book, this Pentateuch, so difficult to read? There are obviously many different types of difficulty when it comes to reading a literary work. This is true even when there are no questions of authorship: childrenʼs books are easier to read than books written for adults; books written for a popular audience are easier to read than those written for the academy. This means, in part, that we have to know what cultural expectations we can place on any given piece of writing before we start wondering whether maybe, just maybe, it is so difficult as to require a literary-historical explanation. Fortunately, we are blessed to have a relative abundance of comparative material against which to judge ancient Israelite writing. We know what fluid, coherent biblical prose looks like because we have examples of it elsewhere in the Bible: entire books, like Ruth; large swaths of corpora, like the so-called Succession Narrative; even 2  This is not to say that it cannot or should not be used on any text other than the Pentateuch, of course. Myriad texts, from Homer to Hamlet, have been rightly subjected to something akin to a source-critical process. 3  One may note the relative abundance of introductions to form criticism, which tend to roam freely and widely over the biblical corpus and which almost universally begin with discussions of universal concepts such as genre. See, e.g., M. Buss, Biblical Form Criticism in Its Context (JSOTSup 274; Sheffield; Sheffield Academic, 1999); J. H. Hayes, Old Testament Form Criticism (TUMSR 11; San Antonio: Trinity University Press, 1974); K. K och , The Growth of the Biblical Tradition: The Form-Critical Method (New York: Scribnerʼs Sons, 1969); G. Tucker, Form Criticism of the Old Testament (GBS; Philadelphia: Fortress, 1971). There are, to my knowledge, no similar overviews of “source criticism” divorced from its pentateuchal (or otherwise text-specific) content. 4  My thanks to Bernie Levinson for these examples.

Why Is the Pentateuch Unreadable?

245

individual chapters, such as Gen 24 or 38. We also know what the literature of Israelʼs neighbors looked like. So we know that we do not require Israelite writing to be childrenʼs literature, but we also know that it would be anachronistic and culturally ridiculous to expect it to look like Joyce or Faulkner.5 So let us return to our imaginary pristine Pentateuch. It is my contention that in the absence of any difficulty reading such a text, we would be unjustified in attempting to find a literary-historical solution to it. After all, what problem would we be solving? If the text presents no literary difficulties, we have no reason to pick it apart. It would be like doing source criticism on a Jane Austen novel. Could one try it? One can try anything. But without any grounds for doing it, the results are going to be ugly: a perfectly fluent text broken up into less fluent pieces, on completely subjective grounds, without any real connection to the text itself. Of course, that is not the Pentateuch we have, not by a long shot. But the recognition that there are texts that we are happy to leave in unified peace leads to the fundamental question of interest in this essay: where the line is between a perfect text that resists literary-historical analysis and a text like the Pentateuch that, virtually all scholars agree, demands it. What if we adjusted the perfect text so that it was a little more uneven in its use of language? This is obviously a bit of a silly example. Most scholars would agree that among the most easily discarded elements of early pentateuchal scholarship is the propensity for lengthy word lists, attributing words and phrases to individual sources.6 The notion that a given word belongs to a particular author and to no other is wrongheaded in theory and in practice. The ancient Israelite authors all wrote perfect Biblical Hebrew; we are in no position to say that one knew a word that another did not. We also cannot say, at least in the abstract, that one author uses a particular word and another does not. All we can say is that, as far as we can tell, one author used a particular word and another did not. It is the fallacy of statistics: just because something appears to be the case in every example we have seen so far does not mean that it will be the case in the next example. So to identify a word as belonging to one author in ninety-nine of its appearances does not mean that it will belong to that author also in its one hundredth. In his writings, my former teacher and now colleague Harold Bloom uses the word “uncanny” with what is, to my mind, alarming frequency. I have

5  Contra, in a way, the extensive comparison of P precisely with childrenʼs literature by S. M c E venue , The Narrative Style of the Priestly Writer (AB 50; Rome: Biblical Institute Press, 1971). 6  See in particular J. E. Carpenter and G. Harford-Battersby, The Hexateuch according to the Revised Version (2 vols.; New York: Longmans, Green, 1900), 1:185–221, as well as H. H olzinger , Einleitung in den Hexateuch (Freiburg i.B.: J. C. B. Mohr, 1893), 93–110, 181–190, 283–291, 338–349.

246

Joel S. Baden

maybe used it once in everything I have published.7 This statistical observation does not mean that my sentence should be attributed to Harold Bloom. The idea that language variation can be a determining factor for dividing a text into separate authors breaks down completely whenever we encounter analyses that try to proceed on such grounds. These were common in early pentateuchal scholarship, where a verse, or often even a single clause, was stripped out of its context and assigned to another author merely because it used a word that had been determined, on the basis of a statistical word list, to belong to someone else. See, for example, Carpenter and Harford-Battersbyʼs vivisection of Gen 39, according to which the story is almost entirely a unity, with the exception of the words “he made him his personal attendant” and “Joseph was well-built and handsome; after these things” in verses 4, 6, and 7. Because the word “attendant,” ‫משרת‬, the expression “well-built,” ‫יפה תאר‬, and the phrase “after these things,” ‫ויהי אחר הדברים האלה‬, were all believed to belong exclusively to E, they were ripped from the narrative, which was otherwise attributed to J. Of course, when this was done we were left not with one beautiful text, as we started with, but with two impossible texts, including the one that reads, simply, “He made him his personal attendant. Joseph was well-built and handsome. After these things.”8 The most famous example along these lines, of course, is Jean Astrucʼs division of Genesis purely according to the distinction between Yahweh and Elohim.9 His results were fine in some parts, but they were disastrous overall.10 Today there is no one, I think, who would argue that the divine names are a reasonable sole basis for literary division.11 It should be remembered, however, that the divine names were not really the rationale for Astrucʼs literary division either, though they may have been the means. Astruc did not come to Genesis thinking that he had a perfect text with the exception of the alternative divine names. He came to Genesis knowing the already quite extensive scholarship that had recognized all of the famous literary problems, particularly in the creation and flood accounts. He knew his Simon well, and Astruc was merely recognizing that the doublets and contradictions that scholars had seen for a century hap-

 Now, perhaps, twice.  Carpenter and Harford-Battersby, Hexateuch (see n. 6), 2:61–62. 9  J. Astruc, Conjectures sur les mémoires originaux dont il paroit que Moyse sʼest servi pour composer le Livre de la Genèse (ed. P. Gibert; Paris: Éditions Noêsis, 1999). 10  See, e.g., his unified view of Gen 37 (Astruc, Conjectures, 300–305 [see n. 9]), or, on the other side, his division of the story of the binding of Isaac in Gen 22 between vv. 1–10 and vv. 11–19 (Conjectures, 225–227). 11  See E. Blum, Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 471–475; J. Van S eters , Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975), 156. 7 8

Why Is the Pentateuch Unreadable?

247

pened to line up, in crucial moments, with the alternation of divine names.12 It is mistaken to claim that Astruc divided Genesis into parallel narratives in order to solve the issue of the divine names. He divided Genesis using the divine names in order to solve the narrative problems. Variations in terminology and style are no reason to divide up an otherwise perfectly good text. So let us take a step further. What if this text also contained multiple genres? Not only narrative but also some poetry, some law, some genealogies – would this be a reason to start asking whether it might not have had multiple authors? Again, I think not, and it really requires little argumentation. Genres do not belong to individual authors; they belong to individual settings in a given culture. Everyone in that culture is able to utilize any number of genres, in any combination. We all do it constantly. The notion that the ancient Israelite author was capable of expressing himself in only one genre, or form, or Gattung, is among the most famous and most famously erroneous aspects of Gunkelʼs otherwise largely brilliant contribution to our field.13 We know better than to treat the ancient Israelite mind as somehow more primitive. What this means, then, is that, in the absence of any other pressing evidence, there is no reason to think, simply because a text moves from narrative to law to poetry to genealogy and back again, that it could not be the work of a single hand. In fact, such fluency with contemporary genres and forms is common today, and probably would have been then, and certainly should be, from the scholarly perspective, evidence of a certain sort of genius. We see this in the prophetic corpus to be sure, and often also in psalms that bring together divergent generic elements into new artistic units – so too with our imagined Pentateuch. Of course, genres are often linked with style, as is frequently noted. It is often the case that fluctuations in style occur precisely where the text transitions from one genre to another. If the text were perfect in every sense, except that a different set of terms and a different literary style were used when it shifted from narrative to ritual law, we would not be in any position to think that we were dealing with multiple authors.14 So let us take yet another step: what if this text also took up a variety of themes over its length? What if, in the course of telling its story, it shifted its thematic emphasis from passage to passage – here concentrating on the theme of sibling rivalry, there on, say, corporate punishment for sin? Unless we were to state from the outset that an ancient Israelite author can treat only a single theme  See the open acknowledgement of Simon, le Clerc, and others in Astruc, Conjectures (see n. 9), 135–137. 13  See his disparaging statements about the intellectual capacity of the ancient Israelite in H. Gunkel, “Israelite Literary History,” in Water for a Thirsty Land: Israelite Literature and Religion (ed. K. C. Hanson; Minneapolis: Fortress, 2001), 31–41, esp. 34. 14  This was, in fact, one of the standard objections to the entire source-critical enterprise among its early critics. See the comment of Cassuto, Documentary Hypothesis (see n. 1), 54: “Change of style depends on change of subject-matter, not on difference of sources.” 12

248

Joel S. Baden

in any given work, there is no obvious reason why this would be unacceptable. (We would also have to declare impossible such multithemed literature as the book of Ruth.) And if the two themes were somewhat closer in nature – in one passage emphasizing the authority of prophetic figures, for example, while concerned with accentuating priestly power in another – would this sort of thematic variance, on its own, demand a literary-historical solution? The instinct might be to say yes – we would have there evidence of one pro-prophetic author and one pro-priestly author. But would we say that if the text were otherwise totally coherent? Or, more pertinently, would these two emphases render the text somehow incoherent? There is nothing impossible about a single author justifying both prophetic and priestly power. There is not even anything unlikely about it, unless we assume from the outset – as, admittedly, many do – that there was some constant battle between the two sides and that every ancient Israelite author had to align himself with one of the two camps. It seems reasonable to accept that limiting ancient authors to a single theme is unfair and, moreover, would result in rather boring literature. In even the most restrictive analyses of the Priestly writings, one would be hard-pressed to say that there is only a single theme present. Of course there are themes that are more central than others, and perhaps there are even overarching thematic concepts, but virtually every biblical text encompasses multiple concepts and focuses. Again, if it were perfectly coherent in every way, yet also contained multiple themes, our imaginary Pentateuch would still be impervious to literary-historical analysis. It would simply be a manifestation of good, complex ancient Israelite literature, like the other examples mentioned already. If we further tweaked this imaginary Pentateuch so that it contained a variety of theological views, would we finally have arrived at the breaking point? Can we imagine a text in which different theological questions and positions are presented without feeling the burning need to tear it to pieces? Obviously there are different levels of theological difference. In one passage God is imagined as speaking to people in dreams (e.g., Gen 20:3), whereas in another God speaks face to face (e.g., Exod 33:11), and in yet another God speaks from heaven (e.g., Gen 21:17). I think most readers could live with such a text – at least no one would think to identify the “God-speaks-in-dreams” source or the “God-speaksfrom-heaven” layer. More difficult, perhaps, would be a text in which in one passage Godʼs control of history from behind the scenes is emphasized (e.g., Gen 45:5), whereas in another it is clear that God takes direct action on behalf of Israel (e.g., Exod 3:8). Still, even in this case one author could well put forward both ideas. They are not contradictory; they are complementary. God sometimes acts this way and sometimes that way. It might not be neat and clean, but it has the advantage of being more true to human experience. Even if in one place the text said that God punishes to the third and fourth generation (Exod 34:7) and in another it said that God does not visit the sins of the fathers on the sons (Deut 24:16) – if everything else about the text were

Why Is the Pentateuch Unreadable?

249

unproblematic, I think that we could live even with this.15 We would understand it as we do so much wisdom literature, as expressing ostensibly contradictory ideas in the service of representing more accurately the range of humanityʼs interaction with God. Theologies are, I suggest, almost always capable of standing beside one another in a singly-authored text.16 The only sorts of exceptions I can think of would be claims that Israel must worship only one god, but in one text that god is identified as Yahweh while in another it is Baal. That would be pretty well irreconcilable – but we do not find that in the Pentateuch, either the real one or the imaginary one being constructed in this essay. Diversity of language and style, of genre, theme, and theology – none of these reach the tipping point, the moment when it is necessary to search for a literary-historical solution to the problems of the text. None of these render the text unreadable, either in our imaginary Pentateuch or in the real one. Moreover, I would argue that not until the twentieth century did anyone ever claim that they did. Astruc, Eichhorn, Vater, Hupfeld, Graf, Kuenen, Wellhausen – none of them divided the text into its constituent parts because it contained a variety of styles, or themes, or genres. Before them, Simon and Spinoza were not compelled to challenge the unity of the text on the basis of disparate terminology or theology. These may have been some of the means by which analysis proceeded, but they were not the rationale for the analysis in the first place. From the very beginning the impetus for source-critical analysis, the reason that the text was considered so unreadable as to require a literary-historical solution, was always and ever the fact that the narrative, on the level of plot – who, what, when, where, why, and how – is self-contradictory, repeatedly and incontrovertibly.17 It is not the names for God that render the flood story unreadable, nor was that ever thought to be the case. It is the blatant contradictions in narrative claim at virtually every stage of the story. It is not the differing theological views related to Godʼs position vis-à-vis the world that render Gen 1 and 2 impossible to read as being from a single hand. It is the narratively untenable sequence of events. It is not the differing emphases on priestly and prophetic authority that eliminate the possibility of reading Num 16 as a unified text. It is the utter confusion on the basic level of the plot, of who is doing what where and when.

15  Though see the extensive commentary on this ostensible contradiction and its dispersion throughout biblical literature in B. M. L evinson , Legal Revision and Religious Renewal in Ancient Israel (Cambridge: Cambridge University Press, 2008), 57–88. 16  See the self-consciously canonical but certainly theologically diverse interpretive treatment of W. Brueggemann, Theology of the Old Testament (Minneapolis: Fortress, 1997). 17  I contend that this is true even for Graf and Kuenen, who concentrated their efforts primarily on the contradictions and duplications among the legal corpora of the Pentateuch. Those legal corpora are only contradictory because they are in a narrative setting, in which the existence of multiple different legal positions, all given by the same deity to the same messenger at the same historical moment, is logically problematic.

250

Joel S. Baden

No one is going to argue that there is not any terminological variation in the two flood stories, or that there is no difference in theology between Gen 1 and 2, or that there are not different thematic emphases in the two stories of Num 16. But if it were only that – if there were no problems on the level of plot – then I think we would not come to the same conclusions. The issue that demands resolution – that demands a literary resolution in particular – is the issue of plot consistency. If that sounds too strong, consider: is an analysis of the flood narrative that left us with two (or more) texts that made absolutely no sense on the level of the plot even remotely imaginable? If it is not the problems of plot that drive the analysis, then why does every scholar try as hard as possible to end up with texts that are readable on that level? We may disagree on stylistic or thematic or theological points, but no one has yet to put forward a flood story in which the water lasts both forty and one hundred and fifty days or in which God twice promises Noah that he will not destroy the earth again. Even scholars who are inclined to isolate the smallest literary units in the Pentateuch inevitably find that those smallest units are narratively coherent – indeed, that coherence is one of the defining features of the smallest literary unit.18 The plot is fundamental; it is irreducible. If the text is sick, it requires surgery. If the text were not sick, no surgery would be required. Cutting up a text even if it shows no signs of being sick is called elective surgery. It can be done, but one should not expect insurance to cover it. It is not necessary, not safe, and when it is over, the patient usually looks worse than before the procedure began. Our text is sick, and that illness is exclusively the literary contradictions on the level of plot. That is the level we are operating on. It is not sick because it uses different terminology and style in its various parts, or because it contains a variety of themes, or because it is theologically intricate – and it is not sick because it contains a variety of traditions, or because it covers a large temporal range, or because we can imagine some of its constituent parts existing in some independent form. These descriptions are true of most complex texts, ancient and modern alike, and we do not slice all of them up into sources or layers or redactions. Jane Austen is full of stylistic and thematic variety, and she, like every author, relies on an abundance of previous oral and written traditions to construct her narrative. But we refrain from dissecting Jane Austen because Jane Austen is perfectly coherent on the level of plot. If it is the contradictions in plot that drive us to the literary-historical analysis of the text – and that is the claim being put forward here – then it is only logical 18  This is true in nondocumentary work as divergent in theory and result as that of, e.g., B lum , Vätergeschichte (see n. 11), and C. L evin , Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993). Later literary developments render the earlier text difficult to read; the earliest levels are always narrative jewels (though sometimes not particularly multifaceted).

Why Is the Pentateuch Unreadable?

251

that our literary-historical solutions should also proceed on the basis of resolving those contradictions in plot. This is because, sensibly enough, if we try to divide the text on other grounds – terminological, stylistic, generic, thematic, theological – then we are not actually addressing the basic problem. This is why Astrucʼs separation of Genesis failed – he was trying to solve the plot difficulties by means of terminology. So too with any such mismatched mechanism, examples of which have been all too common in scholarship of all stripes from the early twentieth century down to the present. Now if one thinks that the problem is not plot but is actually theology, then by all means one may attempt to separate the text according to its various theological viewpoints – but one cannot then expect the narrative problems to be resolved. If one thinks that what makes the Pentateuch unreadable is its conflicting themes, then it is certainly in line with those themes that the text should be divided – but the contradictions in the plot will remain. Similarly, of course, if one divides the text according to the plot contradictions, then one cannot expect all of the terminology, theme, and theology to similarly break down into absolutely neat lines. But that is precisely the point: there is no need for terminology, theme, and theology to break down into absolutely neat lines. In the imaginary Pentateuch constructed above, variations in style, theme, and theology do not necessarily demand literary-historical solutions. They do not render the imaginary Pentateuch unreadable. And so too it is with the texts we identify as making up the real-life Pentateuch. Once the plot contradictions are resolved, if we are left with a narratively coherent text, that text can, like any text ancient or modern, accommodate stylistic and thematic and theological complexity. The reverse is not true: a stylistically or thematically or theologically uniform text cannot accommodate plot contradictions. What makes the Pentateuch unreadable is its thorough-going internally contradictory plot. The analysis that explains that unreadability is, by necessity, grounded in the resolution of those plot contradictions. That is why source criticism exists – that is why anyone ever thought to enter into this sort of analysis hundreds of years ago. And if one does not think the Pentateuch is fundamentally unreadable, then one ought to stop performing elective surgery on it. It has been under the knife for long enough, and surely it could use a rest.

Pentateuchal Coherence and the Science of Reading Jeffrey Stackert Source-critical analysis of the Pentateuch is a response to frustrated attempts to achieve an intelligible reading of this text as a single, unified work. The procedure of contemporary pentateuchal source criticism thus ever recapitulates its historical origins: stymied efforts to read pentateuchal texts as unified compositions motivate alternative explanations of the textsʼ geneses and compositional histories, including the identification of sources, strata, redaction, and compilation. Yet even if the Pentateuchʼs incomprehensibility is well established, having been demonstrated repeatedly and in its various parts,1 it is also the case that for much of its history the Pentateuch has been understood as a unified text. Moreover, though readers who understand the Pentateuch as unified do at times acknowledge various issues in the text that diminish its coherence, they nonetheless achieve a coherence for the whole that they deem satisfactory – or, at least, they claim as much. The question may thus rightly be posed: what contributes to the reception of the Pentateuch as a coherent, unified text? For historical-critical scholars, to ask this question is to consider an issue regularly set aside. If it has been demonstrated that the Pentateuch is a composite work, why examine how it can be understood otherwise? To do so is arguably to move beyond the bounds of biblical studies proper – either a short step into the history of interpretation or a more substantial one into the anthropology and sociology of religion. In either case, the motivations for unified reading are oftentimes taken to be religious and thus external to the text itself. Yet it should be acknowledged that there are multiple factors that contribute to unified readings of the Pentateuch. In addition to external religious factors, there are internal linguistic features that influence such readings; external nonreligious factors also play an important role. In this article, I will focus especially on these latter contributors to perceptions of the text as coherent, namely, linguistic features of the pentateuchal texts themselves and nonreligious, external factors. Specifically, I will apply the findings of empirical research on reading comprehension, textual cohesion and coherence, and cognition, especially in the fields of psychology, linguistics, and education, to the question of how readers understand pentateuchal texts 1  See, for example, Joel S. Badenʼs essay, “Why Is the Pentateuch Unreadable? Or, Why Are We Doing This Anyway?,” in this volume.

254

Jeffrey Stackert

as unified and coherent. I will argue that the various cohesive ties within and between pentateuchal texts, coupled with regular processes of human cognition and reading comprehension, make it possible for readers to perceive even internally contradictory, demonstrably composite texts as coherent. I will focus on two psychological experiments. The first examines the juxtaposition and interrelation of two types of coherence, termed local coherence and global coherence, and their breakdown when a reader encounters content contradictions. The second experiment examines cases of blatant content inconsistencies that go undetected by research participants. I will then turn to three pentateuchal examples that are illuminated by the social-scientific evidence reviewed. My discussion will focus on modern interpretation of biblical texts because the psychological research that I am citing is focused on the modern period. Yet the history of biblical interpretation suggests that similar strategies for creating coherence can be observed across time. I will thus also make some reference to the history of interpretation in my discussion of pentateuchal examples.

Cohesion, Coherence, and Contradiction Before turning to the first experiment, let me briefly define several terms that I will employ. Cohesion refers to the meaningful connections within language, or, more specifically, the internal semantic linkages between sentence elements. In their foundational study on cohesion, Michael A. K. Halliday and Ruqaiya Hasan argue, “Cohesion occurs where the INTERPRETATION of some element in the discourse is dependent on that of another. The one PRESUPPOSES the other, in the sense that it cannot be effectively decoded except by recourse to it.”2 Halliday and Hasan identify five types of cohesion: reference, substitution, ellipsis, conjunction, and lexical cohesion. As a characteristic of the text itself, cohesion may be distinguished from coherence, which is properly an achievement of the reader, even as it is highly dependent upon a textʼs cohesive ties.3 Local coherence refers to a readerʼs attainment of meaningful relations (i.e., successful comprehension of words, phrases, and sentences in terms of causality and/or referentiality) between the constituents of single and adjacent sentences.4 2  M. A. K. H alliday and R. H asan , Cohesion in English (London: Longman, 1976), 4 (capitalization in original). See also, inter alia, L. J. C hapman , Reading Development and Cohesion (London: Heinemann Educational Books, 1983). 3  A. J. Moe, “Cohesion, Coherence, and the Comprehension of Text,” Journal of Reading 23 (1979), 16–20. On the differentiation of referential from relational coherence, see T. Sanders and H. Pander Maat, “Cohesion and Coherence: Linguistic Approaches,” in Encyclopedia of Language and Linguistics (ed. K. Brown; 2nd ed.; 14 vols.; Amsterdam: Elsevier), 2:591–595. 4  J. D. Murray, “Logical Connectives and Local Coherence,” in Sources of Coherence in Reading (ed. R. F. Lorch, Jr. and E. J. OʼBrien; Hillsdale, NJ: Lawrence Erlbaum Associates,

Pentateuchal Coherence and the Science of Reading

255

Local coherence might be extended beyond adjacent sentences but generally not beyond the paragraph. Global coherence refers to the achievement of meaningful relations beyond the adjacent sentence or the paragraph boundary, up to and including at the level of an entire text. This distinction between local coherence and global coherence is built on a theory of activated information in the readerʼs mind – in other words, working memory.5 Local coherence is achieved by accessing and relating information currently activated in a readerʼs mind to make inferences that satisfy coherence standards, whereas global coherence requires the reactivation of such information.6 A contradiction effect refers to the readerʼs measurable recognition of inconsistencies in characterization, causality, or other details in a text, either at a local or global level. Local Coherence and Global Coherence Experiments that test reading comprehension and the construction of local and global coherence often employ problematic textual phenomena reminiscent of pentateuchal examples. As expected, in such studies readers easily recognize contradiction locally; that is, they readily recognize contradiction when problematic features are set side by side within a sentence or in adjacent sentences. Yet experimentation shows that readers also recognize contradiction at the global level. In a study designed to test global coherence, Jason E. Albrecht and Edward J. OʼBrien presented the following narrative to research participants, with either the consistent elaboration or the inconsistent elaboration: Today, Mary was meeting a friend for lunch. She arrived early at the restaurant and decided to get a table. After she sat down, she started looking at the menu. Consistent elaboration: This was Maryʼs favorite restaurant because it had fantastic junk food. Mary enjoyed eating anything that was quick and easy to fix. In fact, she ate at McDonalds

1995), 107, and P. van den B roek et al., “The Role of Readersʼ Standards for Coherence in the Generation of Inferences During Reading,” in Sources of Coherence in Reading (ed. R. F. Lorch, Jr. and E. J. OʼBrien; Hillsdale, NJ: Lawrence Erlbaum Associates, 1995), 356–367. 5  For discussion of working memory see, e.g., A. Baddeley, “Working Memory,” Science 31 (1992), 556–559; idem , “Working Memory and Conscious Awareness,” in Theories of Memory (ed. A. F. Collins et al.; Hove, Engl.: Lawrence Erlbaum Associates, 1993), 13–28. In relation to reading and varying working-memory capacities, see esp. M. D aneman and P. A. C arpenter , “Individual Differences in Working Memory and Reading,” Journal of Verbal Learning and Verbal Behavior 19 (1980), 450–466. 6  Van den Broek et al., “Readersʼ Standards for Coherence” (see n. 4), 354; J. E. Albrecht and E. J. OʼBrien, “Updating a Mental Model: Maintaining Both Local and Global Coherence,” Journal of Experimental Psychology: Learning, Memory, and Cognition 19 (1993), 1061; I. Tapiero, Situation Models and Levels of Coherence: Toward a Definition of Comprehension (Mahwah, NJ: Lawrence Erlbaum Associates, 2007), esp. 85–113.

256

Jeffrey Stackert

at least three times a week. Mary never worried about her diet and saw no reason to eat nutritious foods. Inconsistent elaboration: This was Maryʼs favorite restaurant because it had fantastic health food. Mary, a health nut, has been a strict vegetarian for 10 years. Her favorite food was cauliflower. Mary was so serious about her diet that she refused to eat anything which was fried or cooked in grease. After about ten minutes, Maryʼs friend arrived. It had been a few months since they had seen each other. Because of this they had a lot to talk about and chatted for over a half hour. Finally, Mary signaled the waiter to come take their orders. Mary checked the menu one more time. She had a hard time deciding what to have for lunch. Mary ordered a cheeseburger and fries. She handed the menu back to the waiter. Her friend didnʼt have as much trouble deciding what she wanted. She ordered and they began to chat again. They didnʼt realize there was so much for them to catch up on.7

The research participants read each narrative one sentence at a time on a computer screen, advancing to the next sentence by means of a manual click of a keyboard spacebar. As the participants performed these activities, the researchers measured the reading times for each sentence. Albrecht and OʼBrien found that reading times for the target sentence, “Mary ordered a cheeseburger and fries” (above, in italics), as well as equivalent, contradictory sentences in other, comparable test narratives, were on average more than 15 percent higher for readers who had encountered the preceding inconsistent elaboration than for those who encountered the consistent one (consistent: 2010 milliseconds; inconsistent: 2319 milliseconds). Such increases in reading time indicate a contradiction effect at the global level.8 In other words, the posited explanation for the variant reading times is that readers read more slowly when they recognized the inconsistency between the elaboration that described Mary as a vegetarian with special concern for eating healthy food and the subsequent notice that she ordered a cheeseburger and fries for lunch. This study, which has been replicated,9 argues against claims that readers either do not recognize contradictions at the global level or concern themselves with global coherence only if local coherence has broken down.10 The experiment of Albrecht and OʼBrien has significant implications for understanding claims to pentateuchal coherence at the global level. As is readily apparent, the contradiction in this Mary story corresponds closely with the types  Albrecht and OʼBrien, “Updating a Mental Model” (see n. 6), example test narrative from p. 1070. Also included in the study was a neutral elaboration to serve as a control. 8  Albrecht and OʼBrien, “Updating a Mental Model” (see n. 6), 1064. 9  See, e.g., D. L. Long and J. L. Chong, “Comprehension Skill and Global Coherence: A Paradoxical Picture of Poor Comprehendersʼ Abilities,” Journal of Experimental Psychology: Learning Memory, and Cognition 27 (2001), 1424–1429. 10  G. M c K oon and R. R atcliff , “Inference during Reading,” Psychological Review 99 (1992), 440–466. 7

Pentateuchal Coherence and the Science of Reading

257

of conflicting characterizations and contradictory narrative claims regularly attested in the Pentateuch. Among many comparable pentateuchal examples, I might highlight (1) the sequences of creation in Gen 1:1–2:4a and Gen 2:4b–25; (2) the conflicting claims that the Ishmaelites brought Joseph to Egypt in Gen 37:28 and that the Midianites sold Joseph to Potiphar in Gen 37:36 (an example to which I shall return below); and (3) the conflict between Exod 4:1, where Moses possesses a rod that Yahweh then instructs him to use to perform signs, and Exod 4:17, where the deity gives such a rod to Moses, implying that Moses did not possess a rod previously. Each of these examples contains one or more conflicting claims at the global level, just as the Mary/cheeseburger story does. The Mary story experiment suggests that modern readers should recognize these and other similar inconsistencies in plot and characterization in the Pentateuch. Even so, many readers do not recognize them and thus do not identify incoherence in pentateuchal narratives. The Moses Illusion Additional studies offer possible explanations for readersʼ inhibition or suppression of contradictory or irrelevant data.11 For example, researchers have identified specific instances in which a contradiction effect is repeatedly thwarted, even at the local level. Perhaps the most famous such case is one sometimes termed semantic illusion or, in line with its initial demonstration and especially appropriate to its application to pentateuchal studies, the Moses Illusion. The Moses Illusion experiment poses to participants the question, “How many animals of each kind did Moses take on the ark?” as well as other questions like this one. The Moses question is only answerable if the test subject has prior knowledge of the biblical flood tradition, in which case, however, the question is nonsensical. Researchers consistently find that this question does not seem nonsensical to test subjects. (I should note that the correct answer that the researchers are looking for is two animals [Gen 7:2b, 9, 15]; they do not deal with the seven pairs of clean animals in the Yahwistic account [Gen 7:2a, 3].12) Even when various controls are introduced into the testing – for example, for participant compliance, incomplete encoding, mode of reception, phonological similarity between the correct and discrepant term, participant expectations,

11  For discussion of readersʼ inhibition or active suppression of contradictory or irrelevant data, see Tapieroo, Situation Models and Levels of Coherence (see n. 6), 85–113. 12  The researchersʼ identification of “two” as the correct answer to the question is itself an interesting instance of suppression within their experimental design. It likely reflects what their participants perceive the correct answer to their question to be. This scenario points to the abstracted, coherent story that readers/listeners retain from the biblical narrative and the various retellings of it that they have encountered.

258

Jeffrey Stackert

focus, verbal articulation of the text with the discrepant term, and so on – test subjects still regularly fail to detect distortion.13 Yet there are also limits to the distortion that test subjects will tolerate. As indicated in the label semantic illusion, these limits oftentimes (though not always) relate to word meanings. The semantic relationship between the distorted term and the expected term is a key predictor of the illusion. Thus, for example, the Moses Illusion is quite robust when names of biblical characters are substituted for Noah (e.g., Adam, Abraham, Moses). By contrast, all test subjects recognize the distortion in the question, “How many animals of each kind did Nixon take on the ark?” Similarly, increased relatedness of the distorted term to the larger context in which it is embedded (and not simply to the expected term) increases the occurrence of the illusion.14 Though pentateuchal examples do not replicate the characteristics of the Moses Illusion precisely, explanations for the illusionʼs occurrence are relevant for understanding coherent readings of the Pentateuch. Some psychologists explain the Moses Illusion in light of a more general model of human cognition and comprehension as processes of partial matching. According to this model, new information – words, sentences, and ideas – are introduced into an alreadyexistent network of data in a personʼs memory. A person makes matches between various aspects of the new information and existing memory data, and once a threshold of overlap is reached, even inconsistent data is perceived as coherent.15 According to this model, it is human nature to “do a ‘sloppy’ job of matching,”16 most likely because of the significant benefits that partial matching affords. Humans regularly encounter incomplete information and must draw inferences on the basis of it. If they were not to do so, they would be significantly inhibited in their responses to the information and stimuli encountered and, in some instances, would suffer for it.17 It is thus relatively easy to posit an evolutionary advantage to partial matching in cognition.18 Moreover, as Anthony J. Sanford  T. D. E rickson and M. E. M attson , “From Words to Meaning: A Semantic Illusion,” Journal of Verbal Learning and Verbal Behavior 20 (1981), 540–551; E. N. Kamas et al., “Partial Matching in the Moses Illusion: Response Bias Not Sensitivity,” Memory and Cognition 24 (1996), 687–699; H. Park and L. M. R eder , “Moses Illusion: Implications for Human Cognition,” in Cognitive Illusions (ed. R. F. Pohl; Hove, Engl.: Psychology Press, 2004), 275–291. For discussion of the Moses Illusion in relation to the coherence of biblical texts (with special focus on the book of Judges), see M. Z. Brettler, “The ‘Coherence’ of Ancient Texts,” in Gazing on the Deep: Ancient Near Eastern and Other Studies in Honor of Tzvi Abusch (ed. J. Stackert et al.; Bethesda, MD: CDL Press, 2010), 411–419. 14  Kamas et al., “Partial Matching in the Moses Illusion” (see n. 13), 688, 697; Park and Reder, “Moses Illusion: Implications for Human Cognition” (see n. 13), 282–284. 15  Park and Reder, “Moses Illusion” (see n. 13), 285–286. 16  Park and Reder, “Moses Illusion” (see n. 13), 281, and see p. 285. 17  Kamas et al., “Partial Matching in the Moses Illusion” (see n. 13), 698. 18  For example, if a small animal in the forest hears a rustling of leaves, this rustling may or may not signal the presence of a predator. Should the animal retreat to a safe place? If the 13

Pentateuchal Coherence and the Science of Reading

259

and Patrick Sturt emphasize, both underspecified expression and underspecified interpretation remain very effective in communication in spite of the ambiguities that necessarily attend them.19 The existence of partially matched/underspecified cognition, which expresses a prior assumption about the meaningfulness of stimuli, suggests that readers of the Pentateuch, like readers of any text, exhibit a natural bias toward finding coherence in the text. It also suggests that the ability to find coherence in a pentateuchal text does not correlate with that textʼs actual internal consistency and cohesion. Note that the psychologists who have conducted Moses Illusion experiments do not redefine the contradictions and distortions in their prompts as not contradictory based on their test subjectsʼ inability to recognize them as contradictions and distortions. This is because they have good, empirical data to suggest that they should not. In Moses Illusion experiments, when upon completion of a test researchers have confronted their subjects with an unobserved distortion, those test subjects have immediately recognized the nonsensical quality of the question posed to them. They have also expressed surprise at having failed to detect it.20 Contradiction clearly does not necessarily lead to a conclusion of incoherence. It should also be recognized that there are several features of the Pentateuch that abet the partial matching process by relieving the need for it. For instance, in addition to the inconsistencies among them, there are many instances of consistency – i.e., cohesive ties – among and across the sources combined in the Pentateuch: the protagonist in the flood accounts is always Noah; the patriarchs are always Abraham, Isaac, and Jacob – and in that order; Moses is consistently portrayed as a prophetic mediator between God and Israel and the leader of the Israelites in the wilderness; and so on. In addition, many smaller units in the Pentateuch are not composite and thus readily reward the inclination toward coherence that attends the reading process. The consistency in these units then provides a framework of coherence in which inconsistencies can be more easily addressed or suppressed. Likewise, the chronological arrangement of the pentateuchal plot lends a cohesiveness to the whole and thus engenders a sense of coherence that accompanies and even outweighs the sense of incoherence

animal were to retreat only after confirming the predatorʼs presence, it might well be too late to retreat successfully. By retreating at the first sound of potential danger, the small animal will actually retreat both when a real predator is lurking and when the leaves are disturbed by the wind (or in any other way). The animal will thus in some instances incorrectly interpret the rustling sound as coming from a predator, yet by doing so, the animal will be much more likely to survive. 19  A. J. S anford and P. S turt , “Depth of Processing in Language Comprehension: Not Noticing the Evidence,” Trends in Cognitive Sciences 6 (2002), 382–386. 20  Erickson and Mattson, “From Words to Meaning” (see n. 13), 541.

260

Jeffrey Stackert

generated by the textʼs various contradictions.21 As I will show below, it is also the case that the compilerʼs specific choices in combining source material can contribute to perceptions of the text as coherent. Yet it must also be acknowledged that the partial matching process is inherently creative and thus readily creates connections between materials not otherwise so associated or anticipated.

Partial Matching and Pentateuchal Texts I turn now to three brief pentateuchal examples to show how, through partial matching, coherence can be inferred from specific texts that contain contradictions created by the combination of sources. Each example has been recognized to some extent in the history of interpretation. In the first case, the compiler creates a cohesive tie at the local level between the two sources that he combines by introducing a plot element into the compiled account that does not appear in either of the underlying sources. This new plot element offers a basis for the reader to build further coherence as the narrative proceeds and to understand anew foregoing plot elements. In the second case, the reader is able to make an inference across source boundaries that is consistent with the combination of elements from the two sources but not attributable either to the compiler or to the underlying sources. In the third example, the compiler reassigns an action from one agent in his source text to another in his composite narrative in an attempt to maintain the chronological progression of the plot. In so doing, he creates a possible logical inconsistency that requires the reader, through a partial match, to make an inference about the storyʼs characters not present in the text and not required by the underlying sources. Genesis 37:28, 36 The first example comes from Gen 37, the account of Josephʼs descent into Egypt referenced already. As demonstrated by Baruch J. Schwartz, this text can be divided as follows: P: v. 1 and the first three words of v. 2 (‫;)אלה תלדות יעקב‬ E: vv. 2 (except the first three words), 11b–18, 21–22, 24, the first three words of 25 (‫)וישבו לאכל לחם‬, 28aα, 29–30, 36; J: vv. 3–11a, 19–20, 23, 25 (except the first three words), 26–27, 28aβ–b, 31–35.22 The compiler judges the E and J  See J. Stackert, “Before and After Scripture: Narrative Chronology in the Revision of Torah Texts,” JAJ 4 (2013), 168–185. 22  B. J. Schwartz, “Josephʼs Descent into Egypt: The Composition of Genesis 37 from its Sources,” Beit Mikra 55 (2010), 1–30 (Hebrew); idem, “How the Compiler of the Pentateuch Worked: The Composition of Genesis 37,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; VTSup 152; Leiden: Brill, 2012), 263–278. As Schwartz notes, ‫ חברון‬in v. 14 should likely be attributed to the compiler (replacing a different geographical designation). The compiler apparently introduced it here to harmonize Eʼs 21

Pentateuchal Coherence and the Science of Reading

261

accounts as descriptions of the same event and thus interweaves them. In v. 28, the compilerʼs arrangement of parts of the two stories creates the sale of Joseph by the Midianites to the Ishmaelites (with agents and corresponding actions marked below): Genesis 37:28 ‫ויעברו אנשים מדינים סחרים וימשכו ויעלו את יוסף מן הבור וימכרו את יוסף לישמעאלים בעשרים‬ ‫כסף ויביאו את יוסף מצרימה‬ Midianite traders passed by, and they pulled Joseph up from the pit, and they [the Midianites] sold Joseph to the Ishmaelites for twenty silver pieces, and they [the Ishmaelites] brought Joseph to Egypt.

This verse may be divided between J and E as follows (J underscored; E unmarked): Genesis 37:28 ‫ויעברו אנשים מדינים סחרים וימשכו ויעלו את יוסף מן הבור וימכרו את יוסף לישמעאלים בעשרים‬ ‫כסף ויביאו את יוסף מצרימה‬ Midianite traders passed by, and they pulled Joseph up from the pit, and they sold Joseph to the Ishmaelites for twenty silver pieces, and they brought Joseph to Egypt.

At the local level, the compilerʼs combination is a remarkable and even ingenious means of merging conflicting claims in the two narratives. Unlike the combined narrative, in the J narrative (to which the sale-to-the-Ishmaelites plot element belongs), it is Josephʼs brothers who sell him, as is seen very clearly in the description of their plans in vv. 26–27: Genesis 37:26–27 ‫ויאמר יהודה אל אחיו מה בצע כי נהרג את אחינו וכסינו את דמו לכו ונמכרנו לישמעאלים וידנו‬ ‫אל תהי בו כי אחינו בשרנו הוא וישמעו אחיו‬ Judah said to his brothers, “What profit is there if we kill our brother and cover up his blood? Come, let us sell him to the Ishmaelites, but let not our hand be against him, for he is our brother – our flesh.” And his brothers heeded him.

The compilerʼs introduction of the sale from the Midianites to the Ishmaelites then creates a tension with v. 36, where it is the Midianites who transport Joseph to Egypt and sell him there, not the Ishmaelites: Genesis 37:36

‫והמדנים מכרו אתו אל מצרים לפוטיפר סריס פרעה שר הטבחים‬ The Midianites sold him into Egypt, to Potiphar, Pharaohʼs official, the chief steward.

What I would like to highlight is how the compilerʼs combination of J and E impacts the readerʼs understanding of v. 36. In light of the claims regarding the Midianites and Ishmaelites in v. 28, achieving coherence in v. 36 requires a particular and creative reading along the following lines: Josephʼs brothers account with the location of Jacobʼs settlement preserved from P in Gen 35:27 (“How the Compiler of the Pentateuch Worked,” 266–267 n. 10).

262

Jeffrey Stackert

planned to sell Joseph to the Ishmaelites, but the Midianites got to him before they could do so. The Midianites then sold Joseph into Egypt by selling him to the Ishmaelites, who then took Joseph to Egypt and sold him to Potiphar.23 This reading, of course, is well attested in the history of interpretation, perhaps most famously in the commentary of the medieval Jewish exegete Rashi.24 Yet what is apparent is that, even as it arguably resolves the inconsistency between vv. 26–28 and v. 36, this interpretation does nothing to resolve the other contradictions created by the compilerʼs amalgamation of sources in this chapter: are Jacobʼs sons farmers or shepherds? Is the cistern a place to hide Joseph after he is killed or a place to kill him (or actually save him)? Why is Reuben so surprised and despondent if he participated with his brothers in their plan?25 Whatever harmonization is achieved between vv. 26–28 and v. 36 proves the exception to the rule of inconsistency in the compiled narrative. Even so, the fact that the reader may successfully explain the various references to Josephʼs sale into Egypt does contribute, at least modestly, to an assessment of the chapterʼs coherence. Exodus 7:24 A second example of the creation of coherence by means of partial matching across pentateuchal source divisions appears in the blood plague in Exod 7:14–25. This unit may be divided between J and P as follows: J: vv. 14–18, 20 beginning from ‫וירם‬, 21a, 23–25; P: vv. 19–20 (through ‫)כאשר צוה יהוה‬, 21b–22.26 According to the first half of v. 21 (J), after Moses struck the Nile with his staff, the Egyptians could not drink the bloodied water in their conventional water source, the Nile. According to the second half of v. 21 (P), in fact all of the water of Egypt was affected by the plague (in this case, accomplished by Aaron, not Moses), not just the Nile.

23  For the notion of selling “into” a nation, see Gen 45:4 (‫מצרימה‬, “into Egypt”) and Joel 4:8 (‫אל גוי רחוק‬, “into a distant nation”), which is also the plain sense of Gen 37:36. The issue is the subject of the verb in this verse (Midianites vs. Ishmaelites). 24  See already Gen. Rab. 84:22. For discussion of various traditional Jewish interpretations and modern attempts at a coherent reading of the Midianites/Ishmaelites issues, see J. S. B aden , The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven, CT: Yale University Press, 2012), 3–12. 25  For a full listing of contradictions and disruptions in the narrative of Gen 37, see Schwartz, “How the Compiler of the Pentateuch Worked” (see n. 22), 263. 26  For this source division, see, e.g., M. Greenberg, “The Redaction of the Plague Narrative in Exodus,” in Near Eastern Studies in Honor of William Foxwell Albright (ed. H. Goedicke; Baltimore: Johns Hopkins University Press, 1971), 248; E. Blum, Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 242–256 (see esp. the text layout on pp. 247–248); J. Stackert, “Why Does the Plague of Darkness Last for Three Days? Source Ascription and Literary Motif in Exodus 10:21–23, 27,” VT 61 (2011), 660.

Pentateuchal Coherence and the Science of Reading

263

The combination of these two claims in the compiled narrative impacts the readerʼs understanding of v. 24: Exodus 7:24

‫ויחפרו כל מצרים סביבת היאר מים לשתות כי לא יכלו לשתת ממימי היאר‬ All Egypt dug around the Nile for water to drink, for they could not drink from the waters of the Nile.

The J source, from which the digging element derives, implies that the Egyptians successfully found water to drink when they dug around the Nile. In the combined narrative, however, this is impossible, for in light of v. 21b, they would have simply dug up more blood. Thus, to make sense of v. 24 in the combined narrative, the reader might reasonably infer irony or humor along the following lines: “Those stupid Egyptians – they thought that they could dig up fresh water, but they couldnʼt! They canʼt outfox Yahweh!”27 Such a reading even finds some confirmation as the compiled narrativeʼs plot continues to unfold. According to Exod 10:7 (J), after the hail the Egyptians plead with their king to let the Israelites depart, imploring Pharaoh, ‫הטרם תדע כי אבדה מצרים‬, “Do you not yet see that Egypt is destroyed?” What the Egyptians did not realize before – say, when they foolishly dug around the Nile to find water – they now recognize clearly based on their continued experience of the plagues: Yahwehʼs power is irresistible. Yet as in the example of Josephʼs descent into Egypt, the possibility of making sense of the Egyptiansʼ digging around the Nile in the blood plague fails to address the various other inconsistencies in this unit (e.g., Did Moses or Aaron effect the wonder? Did he strike the water or hold out his hand over it? How long did the wonder last?). To the extent that the meaningfulness that can be constructed in relation to the Egyptiansʼ digging contributes to a sense of coherence for the whole episode, it does so by directing the readerʼs attention away from other inconsistencies in the account or otherwise mitigating the effect of such inconsistencies. In other words, the interpretation of the digging plot element considered here may contribute to the readerʼs ability to reach a satisfactory coherence threshold, but it does so not by fully reconciling the two accounts but by facilitating a partial match. Exodus 8:3 A third example appears in the next plague in Exodus, the plague of frogs (Exod 7:26–8:11). In this text, the source division requires more discussion than in the previous two examples. In my view, Exod 7:26–29 and 8:3b–11aα belong to J, and Exod 8:1–3a and 11aβ–b belong to P. It is also likely that P once included  For review and discussion of both ancient and modern attempts to understand the Egyptiansʼ digging for water, see C. Houtman, Exodus (trans. J. Rebel and S. Woudstra; 4 vols.; HCOT; Kampen: Kok Pharos, 1993–2002), 2:29–30. 27

264

Jeffrey Stackert

a report of Pharaohʼs stubbornness (‫ )ויחזק לב פרעה‬immediately prior to the P portion of Exod 8:11, as in Exod 7:13, 22; 8:15; 9:12, 35; 10:20, 27; 11:10.28 Potentially controversial is the division of Exod 8:3 between two sources.29 Moreover, even for those who would admit compositeness in this verse, the source ascription of its parts might be debated. Arguing against the division of the verse is the fact that v. 3b, ‫ויעלו את הצפרדעים על ארץ מצרים‬, “They brought up the frogs upon the land of Egypt,” seems to follow naturally upon v. 3a, ‫ויעשו‬ ‫כן החרטמים בלטיהם‬, “The magicians did the same with their charms.” For this reason, it might be claimed, there is no reason to posit a source division between the verseʼs two halves. Yet when the frog episode is set within the larger P plague narrative, it becomes clear that an interpretation of v. 3b as a notice of the magiciansʼ replication of the frog wonder can only be understood as such based on the compilerʼs combination of J and P. A first step toward clarifying the source ascription of Exod 8:3b is an examination of Pʼs use of the clause ‫ויעשו כן‬, “They did so/the same.” This clause appears six times in the P plague account (Exod 7:10, 11, 20, 22; 8:3, 14).30 Exodus 7:10 and 11 are part of the account of dueling staffs (7:8–13). In each of these verses, the ‫ ויעשו כן‬clause is followed by an elaboration of the specific actions performed. The examples in Exod 7:20 and 22 are part of Pʼs blood account; the first refers to Mosesʼs and Aaronʼs actions, and the second refers to the magiciansʼ actions. Yet in neither case does P elaborate the specific actions performed (as it does in the dueling-staffs episode).31 The instance in Exod 8:3  The notice that Pharaoh made his heart “heavy” (‫ )כבד‬is uncontroversially J. If P did not once include a comparable statement, the verb ‫ שמע‬in 8:11b would have no proximate antecedent noun to serve as its subject. In this case, the compiler retained the J heart statement (‫ )והכבד את לבו‬and omitted the virtually equivalent P statement (‫ )ויחזק לב פרעה‬because they would have been immediately adjacent to each other in his combined narrative. For discussion of the compilerʼs amalgamation of verses, including additional examples, see J. S. Baden, J, E, and the Redaction of the Pentateuch (FAT 68; Tübingen: Mohr Siebeck, 2009), 282–285, and J. Stackert, A Prophet Like Moses: Prophecy, Law, and Israelite Religion (New York: Oxford University Press, 2014), 86–87. 29  For recent ascriptions of v. 3 in its entirety to P, see, e.g., B lum , Studien (see n. 26), 245, 247; T. R ömer , “The Exodus Narrative according to the Priestly Document,” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions (ed. S. Shectman and J. S. Baden; ATANT 95; Zürich: Theologischer Verlag, 2009), 166 n. 37; C. B erner , Die Exoduserzählung: Das literarische Werden einer Ursprungslegende Israels (FAT 73; Tübingen: Mohr Siebeck, 2010), 177–178. 30  ‫ ויעשו כן‬in Exod 8:13 may be dismissed as a scribal error (vertical dittography) (Stackert, “Why Does the Plague of Darkness Last for Three Days?” [see n. 26], 662). See also the discussion of Pʼs command-execution formulae in T. Pola, Die ursprüngliche Priesterschrift: Beobachtungen zur Literarkritik und Traditionsgeschichte von Pg (WMANT 70; Neukirchen-Vluyn: Neukirchener Verlag, 1995), 116–146, esp. 116–118, 122–124. 31  Exod 7:21b does include the result of these actions but not a description of the actions themselves. 28

Pentateuchal Coherence and the Science of Reading

265

appears in the frog episode and refers to the magiciansʼ actions. There is no accompanying ‫ ויעשו כן‬clause in this episode that refers to Mosesʼs and Aaronʼs actions. P instead includes a specific description of Aaronʼs actions in v. 2 (whereas it must simply be assumed that Moses gives his commands to Aaron because Aaron performs the actions that Yahweh instructs Moses to command Aaron to do).32 The example in Exod 8:14, which is part of Pʼs account of lice (8:12–15), describes the magiciansʼ failed attempt to reproduce the wonder. From these examples a pattern emerges. Within its plague account, P elaborates on the actions referenced by the ‫ ויעשו כן‬clause when it first employs it, namely, in its dueling-staffs episode (Exod 7:10–11). This elaboration then serves as a guide for subsequent uses of this clause in the unfolding narrative. Thus, P either describes its charactersʼ actions explicitly (Exod 8:2) or employs the ‫ ויעשו כן‬clause without elaboration (Exod 7:20, 22; 8:3).33 In the case of Exod 8:14, because this verse relates the magiciansʼ unsuccessful attempt to reproduce the marvel, it also includes a brief description of their action and its infelicity. Comparing these different instances of ‫ויעשו כן‬, Exod 8:3a closely resembles Exod 7:22a, and with the inclusion of the reconstruction noted above immediately prior to Exod 8:11b, Pʼs blood and frog accounts are shown to have virtually identical conclusions: Exodus 7:22

‫ויעשו כן חרטמי מצרים בלטיהם ויחזק לב פרעה ולא שמע אלהם כאשר דבר יהוה‬ The Egyptian magicians did the same with their charms, and Pharaoh was emboldened, and he did not listen to them, just as Yahweh predicted. Exodus 8:3a+11b (reconstructed) ‫ויעשו כן החרטמים בלטיהם ולא שמע אלהם כאשר דבר יהוה‬ The magicians did the same with their charms, and he did not listen to them, just as Yahweh predicted.

Given the foregoing analysis, Exod 8:3b should not be assigned to P. It does, however, readily fit into the J account. As such, it recounts the activities of Moses (and Aaron), not those of the Egyptian magicians.34 In other words, it is the fulfillment of the commissioned threat in Exod 7:26–29. Omitted from this fulfillment is an actual description of Mosesʼs speech to Pharaoh, but such omis32  On the regular occurrence of partial reports of commissioned speech and action in biblical narrative, see A. M. Vater, “Narrative Patterns for the Story of Commissioned Communication in the Old Testament,” JBL 99 (1980), 365–382. 33  See n. 31, above. 34  The objection might be raised that Moses alone is presumed as the producer of the plague in Exod 7:26–29. The plural verb in 8:3b is then problematic. Yet note that, throughout, J presumes Aaronʼs presence alongside Moses (Exod 4:14–16). Moreover, Aaron appears alongside Moses unproblematically in v. 4 when Pharaoh pleads for their help in removing the frogs. There is thus no problem with assigning v. 3b, with its plural verb, to J. See H. Holzinger, Exodus (KHAT; Tübingen: Mohr Siebeck, 1900), 23.

266

Jeffrey Stackert

sions are characteristic of pentateuchal commission and fulfillment narratives generally, including Jʼs plague account in particular.35 Much as he does in the case of the sale of Joseph in Gen 37:28, the compiler dissociates an action from one character/party and assigns it to another. He does so in this case in order to avoid a doubled report of Moses and Aaronʼs imposition of frogs upon Egypt, an event that he includes from P in Exod 8:2. Though the compilerʼs work here is very clever and successfully facilitates the maximal preservation of his sources, the resulting story also contains a logical problem that readers may resolve through partial matching. This problem arises in the progression from v. 3 to v. 4: if the magicians have most recently plagued Egypt with frogs, why does Pharaoh plead with Moses and Aaron to remove them? Should he not command the magicians to remove the frogs that they just produced? The text offers no explicit answer to these questions. Before turning to possible solutions, it should be noted that the differing natures of the frog plague in J and P may contribute to the compilerʼs creation of the logical problem in vv. 3–4. In J, the plague has noticeable and lasting effects: the frogs are highly invasive, and after they die, they must be piled into heaps; they also produce a great stench (8:9–10). In P, by contrast, the frog wonder is momentary and without any durative impact. Thus, the frogs that the magicians produce are not added to the already-existing frogs that Moses and Aaron have produced; the latter are already gone when the magicians reproduce them. A similar situation obtains in the P blood plague: the magicians are able to reproduce the sanguination of the water in all the land of Egypt because the bloodiness resulting from Aaronʼs marvel is already resolved before the magicians act.36 Because the compiler had separate J and P frog accounts and, as shown already, understood well their narrative claims, he may have maintained Pʼs view of the magiciansʼ nondurative wonder when he juxtaposed Exod 8:1–3a with vv. 3b–4 (and following), leading to the incongruity in the progression from v. 3 to v. 4 in the compiled text. Whatever the origin of the illogic that the compiler introduces, there are several solutions that a reader might creatively infer from the text to overcome it. For example, the magicians have functioned heretofore in the narrative as mimics. Thus, if Moses and Aaron remove the frogs that they produced, the magicians may be counted on to do the same, even without explicit notice of their doing so. Alternatively, a reader might conclude that the magicians are simply unable to remove the frogs; only Moses and Aaron, the Israelite heroes, are capable of performing this feat. That Pharaoh calls on Moses and Aaron to seek divine assistance from Yahweh might be understood as confirming this  See Vater, “Narrative Patterns” (see n. 32), 366 n. 6 (pattern III), 373.  See Stackert, “Why Does the Plague of Darkness Last for Three Days?” (see n. 26), 661–662, and C. R. M oss and J. S tackert , “The Devastation of Darkness: Disability in Exodus 10:21–23, 27, and Intensification in the Plagues,” JR 92 (2012), 362–372. 35

36

Pentateuchal Coherence and the Science of Reading

267

view (even as Pharaohʼs plea tends to suggest that the frogs to be removed were summoned by Moses and Aaron through Yahwehʼs power [7:27] and not by the Egyptian magicians). The fact that the magicians fail to reproduce the lice of the immediately following plague (8:12–15) might be taken as further confirmation of the magiciansʼ inability to remove the frogs. If so, the magiciansʼ loss of power might be understood as incremental: they fully produce the crocodile (7:11–12) and blood (7:22), produce frogs but fail to remove them (8:3–4), fail to produce lice (8:14), and are entirely unable to appear when afflicted by boils (9:11). The perception of such patterning, even if only the result of creative partial matching, lends a sense of coherence in this case, not least because such patterns do exist in unified compositions.37

Concluding Remarks In one respect, the social-scientific studies reviewed above confirm scholarsʼ experiences with their students and other readers of the Bible: readers are fully capable of suppressing or otherwise overlooking significant internal discrepancies in biblical texts in their pursuit of coherence. Yet these social-scientific studies also offer important and underappreciated insights into the specific procedures that readers undertake to address discrepancies. The Mary/cheeseburger story example, which closely resembles examples of content discrepancies in composite pentateuchal narratives, demonstrates that readers are able to recognize content contradictions, even at the global level. The history of biblical interpretation confirms that such discrepancies have long been observed in pentateuchal texts: the same discrepancies that serve as data for modern source critics have long occasioned creative harmonizations, namely, interpretive solutions achieved through partial matching. For its part, the Moses Illusion illustrates how widespread and robust the propensity for partial matching is: readers regularly and readily overlook discrepancies in texts, even when they are attentive to their potential appearance. The common experience of underspecification in language requires humans to be on constant alert to “fill in the blanks” – to make partial matches – and they do so with great success. In some instances, however, this “success” is incomplete or even illusory, as it is with many pentateuchal texts.  The P account does employ a pattern in its depiction of the magicians that is very similar to the one imagined here. See Stackert, “Why Does the Plague of Darkness Last for Three Days?” (see n. 26), 661. Regarding the pattern hypothesized above, the omission of any reference to the magicians in the imposition of swarm and pestilence, which appear in the compiled account between lice and boils, presents a potential problem. For the most recent assertion of coherence in the final form of the Exodus plagues narrative by means of an identification of patterning (attributed to a redactor), see J. Grossman, “The Structural Paradigm of the Ten Plagues Narrative and the Hardening of Pharaohʼs Heart,” VT 64 (2014), 588–610. 37

268

Jeffrey Stackert

If the theory of partial matching in cognition is correct, an impulse toward coherence characterizes both noncritical and critical readers of the Pentateuch. It comes as little surprise, then, that scholars must continue to refine their compositional analyses of pentateuchal texts. The discussion of Exod 8:3 above demonstrates this point well. Compositional analysis requires a cultivated resistance to the human impulse toward coherence, even as this drive was fundamental to the initial identification of sources and strata in the Pentateuch and continues to be so in contemporary compositional analysis. This study thus sheds light on what all readers – critical and noncritical alike – are doing when they read the Pentateuch.

Does the Pentateuch Tell of Its Redactional Genesis? The Characters of YHWH and Moses as Agents of Fortschreibung in the Pentateuchʼs Narrated World Jean-Pierre Sonnet “Genuine works of art, while relating all sorts of things, are really telling of their own birth.”1

In current pentateuchal research, methods are often played off against each other: narrative approaches are regularly stigmatized for their lack of attention to the fractures in the text, and redactional investigations are criticized for failing to pay attention to narrative intent. This paper will address the debate the other way around, by focusing on cases in which the activities normally associated with redaction become embedded in the narrative.2 In the staging of the Pentateuch, both God and Moses are cast in hermeneutical and scribal ventures. In a substantial part of the five books, the plot is the story of a progressive yet discontinuous legal revelation and of an ongoing yet intermittent updating of regulations. The narrative process in question, I shall argue, both camouflages and mirrors the redactional process implied in the genesis of the biblical texts. In a certain sense, the authors of the Pentateuch have turned its formation and compositional growth into one of its main themes, thanks to appropriate characters: a revisionary God and an exegetical prophet. As the story of an organic yet discontinuous revelation, the pentateuchal plot astutely mimics its redactional making and by doing so legitimizes it. In other words, the authors of the Pentateuch have turned redaction into revelation: the redaction process, behind the scenes, is analogically mirrored on the front stage in the narrative unfolding of the revelation process.3 1  R. Jakobson, Language and Literature (ed. K. Pomorska and S. Rudy; Cambridge, MA: Belknap, 1987), 307. 2  In the present essay, the term redaction will refer comprehensively to the scribal genesis of the biblical texts and books. In this use and scope, redaction thus means more than the editing of an already formulated literary document or the compilation of a series of traditions and relates to the overall process of the Bibleʼs literary growth. 3  I owe the conceptual couple redaction–revelation to B. M. L evinson , “The Revelation of Redaction: Exodus 34:10–26 as a Challenge to the Standard Documentary Hypothesis,” forthcoming.

270

Jean-Pierre Sonnet

1 A Revisionary God The primary achievement in the redaction-as-revelation story of the Pentateuch is the casting of a revisionary God who engages in continuous updating and confirmation of his words by supplementing and reformulating them. These changes are achieved through direct speech – in other words, by a character speaking. In contrast, in the books of Ezra and Nehemiah, for instance, authoritative words are, from the start, authoritative texts (e.g., Cyrusʼs written edict, lists, authorized letters, genealogical records, etc., with the last such text being the Torah book) to be searched for in the archives, disclosed, brought in, read aloud, translated, or commented upon.4 In the Pentateuch, all the regulations are, in the first instance, words spoken by evolving characters – God and Moses. God in particular marks himself out as the character who will say it again, freely expanding and mending his own words (freely, for God is not bound to the so-called canon formula: “do not add to it or take anything from it” [Deut 13:1]). The liberal character of the speaking and self-quoting God is all the more surprising since God did not need to repeat himself in the act of creation when he declared, “Let there be light” (Gen 1:3). Yet, in the ensuing history, the same God stands out as the deity who keeps revisiting his own words, adjusting them to human contingency and fallibility. Direct speech – the first of Godʼs attributes – is a decisive channel and crucible of such changes in divine view and will. The first impressions of the efficiency of Godʼs word created in Gen 1:3 are, however, somehow vindicated: adding new words (after creation, after Sinai) is not, for Godʼs character, a way to evade or to defuse his original words; it is, rather, a way to bring them home. Godʼs Redirection of the Human Diet The first occurrence of divine amendment to known regulations appears in Gen 9. The scene here creates what will be an enduring impression: God speaks the way the scribes write, and Godʼs Fortsprechen matches the scribesʼ Fortschreibung. As can be seen in chart 1, Godʼs redirection of the human diet in Gen 9:3 makes the most of the original phrasing of the menu in Gen 1:29, ‫נתתי‬ ‫לכם את כל עשב‬, “I give you every plant.” In Godʼs reformulation, it now evolves into ‫כירק עשב נתתי לכם את כל‬, “just as the green plants, I give you every[thing]” (Gen 9:3). “The fact that the narrative traditio [in Gen 9:3] has indicated that the new edibles are a divine allowance ‘like’ those permitted in the older traditum [in Gen 1:29],” Michael Fishbane writes, “shows a deliberate attempt to relate the new dispensation to the original source and its authority.”5 Godʼs reformulation, moreover, is typically lemmatic: the elements of the original phrase  See D. Joseph, “Raconter lʼHistoire avec des documents: Mise en intrigue de lʼécrit dans le livre dʼEsdras,” NRTh 134 (2012), 531–547. 5  M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985), 318. 4

Does the Pentateuch Tell of Its Redactional Genesis?

271

“every plant” (‫ )כל עשב‬have been broken down and redistributed. The first, ‫עשב‬, “plant,” enters in the comparison introduced by ‫כ׳‬: “just as [I gave you] the green plant (‫)כירק עשב‬.” The second element, ‫כל‬, “every,” now defines the updated human menu: “I give you every[thing] (‫)כל‬.”6 Chart 1 Genesis 1:29

Genesis 9:3

A. “I give you every plant (‫)כל עשב‬ B. [. . .]”

B. “[. . .] A. just as [I gave you] the green plant (‫)עשב‬,   I give you every[thing] (‫)כל‬.”

The sweeping expansion to the menu, however, comes along with a restriction, as shown in chart 2, expressed through what amounts to a technical term, ‫אך‬: “However (‫)אך‬, flesh with its life – its blood – you may not eat; and, however (‫)ואך‬, for your lifeblood, I shall surely require [. . .]” (Gen 9:4–5). The priceless value of life, animal and human, is reasserted through the double ‫“( אך‬however”) sentence, in distinctive scribal fashion.7 The divine concession is thus a way to accommodate humans without giving up ethical standards. Chart 2 Genesis 9: 3–5 v. 3 v. 4 v. 5

“just as [I gave you] the green plant, I give you every[thing] ‫ אך‬however, flesh with its life-force – its blood – you may not eat; ‫ ואך‬and, however, for your lifeblood I shall surely require [. . .].”

There is no need to rehearse here the procedures of innerbiblical exegesis involved. What needs to be observed is the fact that God speaks this way. Godʼs oral procedures match the scribesʼ scriptural procedures in their hermeneutics of innovation.8 The redactional strategies of redirection, restriction, expansion,  For discussion of legal expansion through the generalizing formula ‫כל‬, “every,” see F ishbane , Interpretation (see n. 5), 170–177. The technique of lemmatic citation and reformulation is finely encapsulated in Bernard Levinsonʼs formula: “Reduced to a cluster of individual lemmas and then reassembled in a new context, the older doctrine becomes infused with new content”; B. M. Levinson, Legal Revision and Religious Renewal in Ancient Israel (Cambridge: Cambridge University Press, 2008), 80. 7  For discussion of legal restriction through the formula ‫אך‬, see Fishbane, Interpretation (see n. 5), 319–320. 8  Godʼs amendments actually illustrate the main scenarios of innerbiblical (legal) exegesis. The very case under consideration (Gen 9:4, 5) is an instance of legal restriction with introductory formula through the technical expedient ‫אך‬, “however, only” (see F ishbane , Interpretation [see n. 5], 319–320). Other cases exemplify the procedure of legal expansion with an introductory formula such as ‫כן תעשה‬, “you shall do likewise”; see, for instance, Exod 23:11b. Examples of legal expansion and restriction in Godʼs speech without introductory 6

272

Jean-Pierre Sonnet

recontextualization, and semantic specialization are prototypically met in Godʼs discourse. The procedures in question, conversely, provide the rational model that enables us to make fuller sense of Godʼs oral amendments, which, in most cases, include an imaginative reelaboration of the wording of previous utterances. One could say that this is just a truism, since Godʼs words are obviously a redactional creation by the scribes, one that fits their own agenda. Yet overlooking the staging of the revision process as the work of the deity misses an important aspect of the scribesʼ enterprise. Backstage and front stage are here in a close, reflexive relationship. In narrated history, the divine character is the one who, before anyone else, speaks the way scribes (will) write – inscribing the new within the old without revoking it. The revision of the human diet is, indeed, presented as an expansion, not as a cancellation, as Fishbane puts it: “For while Gen. 9:3 is a new divine revelation on edibles, its purpose is to extend the legal scope of Gen. 1:28.”9 A divine character who reasserts his word through changes and amendments and who does so “the lemmatic way”10 provides the figurative authority the scribes needed to legitimize their own venture. God, in that sense, is the prototype and the patron of the scribes in their revisionary endeavor. The dramatic representation of Godʼs amendments, however, is not limited to the logic of legitimation: the narrative portrayal of the God who “says it again” is also a tribute to a specific divine manner in the matter. God, the narrative claims, has a way to mend his previous words. In Gen 9, this way is associated with the logic of blessing (v. 1), ethical accountability (v. 5), and covenant (v. 9). Moreover, God is driven in the revision of his design by what can be called his repentance for repentance, that is, his repentance for his decision to bring the flood (see Gen 8:21).11 In other words, Godʼs revision takes place within a specific plot and involves Godʼs inner drama. The staging of God in Gen 9 thus creates determinate first impressions. When creating the world, God did not have to repeat himself; when legislating in history, the same God turns into a quoting God who reaffirms previous standards by changing them through regulated procedures. Genesis 9 actually creates a meaningful precedent: a great part of the ensuing pentateuchal narrative will be occupied with recounting similar divine amendments of previous divine words. formulae yet with appropriated and sophisticated scribal techniques are equally attested (see the analysis of Exod 34:19–20 in the present essay, pp. 273–275). 9  Fishbane, Interpretation (see n. 5), 318 (my italics). 10  I owe the expression to H. E. Wiegand, Semantics and Lexicography: Selected Studies (1976–1996) (Lexicographica Series Maior 97; Berlin: de Gruyter, 1999), 237–238. 11  See J.-P. S onnet , “Godʼs Repentance and ‘False Starts’ in Biblical History (Genesis 6–9; Exodus 32–34; 1 Samuel 15 and 2 Samuel 7),” in Congress Volume Ljubljana 2007 (ed. A. Lemaire; VTSup 133; Leiden: Brill, 2010), 469–484.

Does the Pentateuch Tell of Its Redactional Genesis?

273

After Sinai When it comes to Godʼs legal Fortsprechen, the richest point of departure is, of course, the Sinai revelation. The event has the form of an extended speech act by the deity.12 Yet the oral revelation of the Sinai law is not Godʼs last word on the matter. The Sinai God repeatedly revisits his own legal formulations, and he does so by (self-)quotation. The process of divine revision and supplementation of the Sinai words starts as early as Exod 34:14–26, i.e., the divine answer to the crisis provoked by the golden-calf episode. Godʼs reformulation of earlier laws in Exod 34 is particularly fitting in the narrative context of Exod 32–34: “the terms of the renewed covenant – prohibition of idolatry and a calendar of ‫ַחג‬ provisions,” Shimon Gesundheit writes, “all pertain to the two spheres in which the Israelites betrayed YHWH, celebrating ‫( ַחג‬32:5) and worshipping a molten calf (32:6, 8).”13 In the so-called Ritual Decalogue of Exod 34:14–26 God as a matter of fact updates and reformulates Sinai commandments, in particular the festival calendar revealed in Exod 23:14–19.14 The reuse in Exod 34:20 (in the regulations of the Feast of Unleavened Bread [vv. 19–20]) of the phrase “and my face shall not be seen empty-handed” is typical; this phrase was borrowed from Exod 23:15, where it did not mention a particular offering.15 The phrase is restated in Exod 34 after a significant interpolation of words spoken by God before Sinai: the rules pertaining to the first-born, animal and human, and the provision for their offering or redemption. The rules in question were originally 12  As against Mesopotamian law, which was uttered in most cases by the king; see M. Greenberg, “Some Postulates in Biblical Criminal Law,” in Yehezqel Kaufmann Jubilee Volume (ed. M. Haran; Jerusalem: Magnes, 1960), 5–28. 13  S. Gesundheit, Three Times a Year: Studies on Festival Legislation in the Pentateuch (FAT 82; Tübingen: Mohr Siebeck, 2012), 36. 14  The narrative staging of the Ritual Decalogue – located in the book after the giving of the Ten Words and of the Covenant Code – mirrors its redactional genesis. The regulations of Exod 34:14–26 have long been assigned to the earliest of the pentateuchal sources. Recent contributions by S. Gesundheit, E. Blum and B. M. Levinson, however, have shown, on the basis of close reading, that the text actually represents a revision of extant materials, in particular the festival calendar in Exod 23:14–19. The festival calendar in Exod 34:18–26, Shimon Gesundheit writes, is “but a midrashic revision of [Exod 23:14–19]. The inner-biblical midrashic process has solved difficulties, eliminated obscure words and phrases, and drawn conclusions based on the juxtaposition of disparate elements in the earlier text. Archaic linguistic usages have been replaced by later ones, and the discrepancies between the ancient festival calendar and those calendars found in later Pentateuchal texts have been harmonized” (Gesundheit, Three Times [see n. 13], 36). In Gesundheitʼs analysis, “the secondary nature of this revision is reinforced by the presence of Priestly influence and Deuteronomic style” (p. 37). See also E. Blum, “Das sog. ‘Privilegrecht’ in Exodus 34,11–26: Ein Fixpunkt der Komposition des Exodusbuches?” in Studies in the Book of Exodus: Redaction–Reception–Interpretation (ed. M. Vervenne; BETL 126; Leuven: Leuven University Press, 1996), 347–366; L evinson , “Revelation” (see n. 3). 15  See in particular Gesundheit, Three Times (see n. 13), 12–21.

274

Jean-Pierre Sonnet

formulated by God in his instructions before the first Passover (Exod 13:1–2, 11–15). By their insertion into the festival calendar, a new rule is created: on the Feast of Unleavened Bread, the sons of Israel will appear before YHWH, bringing along any first-born livestock as an offering.16 In the narrative staging, the scribesʼ phrasing in Exod 34:14–26 expresses the authority of the deity, who can make clear what was left opaque in a previous legal provision (Exod 23:15) and who can elaborate on it with a regulation he has uttered on yet another occasion (Exod 13:1–2, 11–15). Godʼs reformulations not only draw from previous stipulations, they also use words spoken in the dialogical context in the story world. In the legal and ritual reformulation in Exod 34, God thus instills a subtle echo of another episode into the drama that has led to the innovation of the offering of the first-born. Godʼs self-reference in Exod 34:23 as “the Lord, YHWH, the God of Israel,” in a variant to Exod 23:17 (see chart 3),17 actually puts an end to a dispute that has undergirded the golden-calf episode. Chart 3 Exodus 23:17

Exodus 34:23

“Three times during the year all your males “Three times during the year all your males shall appear before the Lord, YHWH.” shall appear before the Lord, YHWH, the God of Israel.”

The starting point (see chart 4) is the peopleʼs blasphemy in Exod 32:4, which turns the molten calf into Israelʼs deity: “These are your gods, Israel, who brought you up out of the land of Egypt!”18 God condemns the peopleʼs blasphemy (v. 8) in a move to estrange them (“your people,” he says to Moses in v. 7) and substitute Moses as the head of a new nation. In his intercession, however, Moses endeavors to revive the exclusive link between YHWH and the people of Israel, calling on God to “remember Abraham, Isaac, and Israel, your servants” (Exod 32:13). In this unprecedented expression,19 Moses deflects Godʼs foundational promise to the patriarchs and inserts Israel as the name of the third patriarch. In other words, Moses recontextualizes the name Israel in its appropriate theological frame of reference. The trope is efficient: God indeed addresses the people as “sons of Israel” (33:5; the one and only occurrence of “sons of Israel” in Godʼs speech in Exod 32–34). The next and final step in 16  See B. M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (Oxford: Oxford University Press, 1998), 69, 91–92. 17  See Gesundheit, Three Times (see n. 13), 22–23. 18  The point is aptly noted by G esundheit , Three Times (see n. 10), 23, who, however, attributes the quote to Aaron. 19  This distinctive syntactical sequence does not occur elsewhere either in Exodus or in the Pentateuch.

Does the Pentateuch Tell of Its Redactional Genesis?

275

the process of reconciliation is Godʼs self-reference as the “God of Israel” in the Ritual Decalogue: “Three times during the year all your males shall appear before the Lord, YHWH, the God of Israel” (Exod 34:23). What emerges is a demonstration of Godʼs rhetorical skill (next to Mosesʼs) in legal as in covenantal matters. Summoning words of the past, he opens a new future for the people: the revisionary God is visionary in his reformulations, as much as are the scribes in their legal Fortschreibung. Chart 4 Exodus 32:4 (Israel)

Exodus 32:8 (God)

Exodus 32:13 (Moses)

Exodus 33:5 (God)

Exodus 34:23

“These are your gods, Israel, who brought you up out of the land of Egypt!”

“They have cast for themselves an image of a calf, and have worshiped it and sacrificed to it, and said, ‘These are your gods, Israel, who brought you up out of the land of Egypt!’ ”

“Remember Abraham, Isaac, → “Say to the and Israel, your sons of Israel [. . .]” servants”

“Three times during the year all your males shall appear before the Lord, YHWH, the God of Israel”

Compare Exod 33:1 (God) “I swore to Abraham, Isaac, and Jacob”

Compare Exod 23:17 “Three times during the year all your males shall appear before the Lord, YHWH”

The Narrative Expedient of the Tent of Meeting The ritual regulations of Exod 34 were meant to preserve the reestablished covenant from any new major breach such as the golden-calf crisis that would compromise the very existence of the bond between YHWH and the people. These measures, however, were only the first step in a more comprehensive process of providing the covenant with regulations intended to update it and to adapt it to its fallible human partners. The procedures for sacrifice and atonement revealed in Leviticus are important examples of this phenomenon. The supplementation process initiated in Exod 34 thus extends beyond Exodus, and it does so through the expedient of the tent of meeting.20 The tent of meeting makes 20  The “tent of meeting” motif in Exod 33:7–11 has a long record in source and redactional criticism. For recent reassessments of the critical dossier, see J. S. Baden, “On Exodus 33,1–11,” ZAW 124 (2012), 329–340. In a synchronic reading, Gianni Barbiero has indicated that rather than being a “digression” in the dialogue between YHWH and Moses, the passage narrowly fits its narrative context. Taking advantage of the temporal determination in v. 6,

276

Jean-Pierre Sonnet

its entrance in Exod 33:7–11 as the material symbol of a conciliation process between the holy God and the sinful people. Beginning with its first appearance in Exod 33:7, the tent is linked to recurrent use,21 and this use extends, via Lev 1:1 and Num 1:1, up to Deut 31. The tent thus hosts an ongoing legal revelation that revises and supplements the Sinai provisions, equipping the covenant with built-in procedures that take into account the peopleʼs fallibility. Supplementation, though, is combined with reformulation, as seen particularly in Lev 17–26, the so-called Holiness Code. The God of Sinai is prone to revision not only to ensure the recovery of the covenantal relationship but also to grant it maximal intension and extension, with holiness now explicitly extended to the entire community of Israel.22 Supplementation and Reformulation in the Diet of the Sons of Israel In the process just mentioned, the case of dietary laws is worth considering again, in particular because here it highlights long-distance revision by the divine character. The diet meant for the sons of Noah, Fishbane writes, was “notably vague.”23 It amounted to “every moving thing [‫ ]כל רמש‬that lives shall be food for you” (Gen 9:3). Similarly, the stipulations meant for the sons of Israel in the original Sinai revelation remained particularly fragmentary as far as diet was concerned. The Covenant Code contemplated just two cases of diet (Exod 22:30 and 23:19).24 This indefiniteness called for the covenantal determinations that appear in Lev 11 and 17. These determinations are formulated after Sinai Barbiero further states that the tent of meeting is “der Ort, wo JHWH mēhar hôrēb Mose die Tora mitteilte” (G. Barbiero, “Ex xxxiii 7–11: Eine synchrone Lektüre,” VT 50 [2000]: 152–166, here 158) – a destination substantiated in Lev 1:1 and Num 1:1. The destiny of the tent of meeting is best described by Blum: “In seiner Entstehung ist das Zelt verknüpft mit der Krise der Beziehung zwischen Jhwh und Israel, in seiner weiteren Bedeutung erscheint es jedoch von dieser Herkunft unbelastet und dient als Ort der unmittelbaren Gottesbegegnung (für Mose) auch nach dem Aufbruch vom Gottesberg”; E. Blum, Studien zur Komposition des Pentateuchs (BZAW 189; Berlin: de Gruyter, 1990), 62. 21  Contrast the opposite view by M. Rogland, “ ‘Moses Used to Take a Tent’? Reconsidering the Function and Significance of the Verb Forms in Exodus 33:7–11,” JTS 63 (2012), 449–466, who construes the yiqtol and weqatal verb forms of the pericope as injunctions rather than as reports of a repeated or habitual activity by Moses. 22  The vocation to sanctity first expressed in Exod 19:6 (“you shall be for me a priestly kingdom and a holy nation”) can be formulated anew (Lev 19:2; 20:7–8, 26; 26:12; proleptic announcement in 11:44–45). Intimacy has been regained, to the point that God announces to the people that he would “walk in their midst” (Lev 26:12, cf. Gen 5:22, 24; 6:9); see C. N ihan , From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 106. 23  Fishbane, Interpretation (see n. 5), 318, n. 3. 24  Exod 22:30 prohibits eating “flesh torn by beasts in the field”; Exod 23:19 (repeated in 34:26) forbids “boiling a kid in its motherʼs milk.”

Does the Pentateuch Tell of Its Redactional Genesis?

277

and the resumption of the covenant in Exod 34 but also after the erection of the sacrificial altar in Exod 38:1–7 and its first use in Exod 40:29. In a first step, God, in Lev 11, enumerates a new range of discriminations within the “moving things” of Gen 9, and he does so by reusing the categories of animal species mentioned in Gen 1:28 in the definition of manʼs dominion (“the fish of the sea, the birds of the air, and every living thing that moves upon the earth”; for the last two, see also Gen 9:9–10). In Lev 17, God similarly resorts to a provision introduced in the primeval age, further elaborating on the criterion of the blood formulated in Gen 9:4: “However, flesh with its life – its blood – you may not eat (‫)אך בשר בנפשו דמו לא תאכלו‬.” In the reformulation of Lev 17:11 and 14 it now becomes: “For the life of the flesh is in the blood; and I have given it to you for making atonement for your lives on the altar; for, as life, it is the blood that makes atonement” (v. 11); and “You shall not eat the blood of any flesh (‫)דם כל בשר לא תאכלו‬, for the life of every flesh is its blood” (v. 14). The universal Noahide law has thus been applied to Israel with the additional provision that the animalʼs blood be collected on the altar.25 In other words, the regulations of Lev 17 make sense as a divine elaboration on previous divine words – previous in both Godʼs spoken pronouncements and the narrative.26  In a redactional perspective, Jacob Milgrom points out, “H has applied this universal Noahide law to Israel, but insisting that the altar must be the depository of the animalʼs blood”; J. Milgrom, Leviticus (AB 3A; New York: Doubleday, 2000), 1457. 26  Lev 17, Christophe Nihan writes, “is best understood as a complex but nevertheless outstanding case of inner-biblical exegesis which presupposes and reinterprets earlier legislation on sacrifices in P, D, and even CC [the Covenant Code]. Its main theme is the rejection of the permission of profane slaughter introduced by D and further adopted by P. Such slaughter is [now] limited to game animals (v. 13–14)”; N ihan , Priestly Torah (see n. 22), 429. By being featured in Leviticus, the regulation of the Holiness Code thus precedes, in the narrative sequence of the Pentateuch, one of its (plausible) redactional sources, the Deuteronomic Code. The narrative order of presentation, however, has a rationale of its own, which lies in the hierarchical sequence of the exegetical agents: in the Holiness Code, it is God who revisits hermeneutically the provisions of Gen 9; in Deut 12 and 14, it is Moses who elaborates his prophetic and prescriptive teaching. The rationale in question goes along with a spatial-temporal motivation: whereas Lev 17 is revealed during Israelʼs stay at Sinai, Deut 12 and 14 are part of Mosesʼs Torah speech, a legal instruction destined to accompany the second generation in its entry into the land. The narrative representation thus enables the compounding of hierarchy in authority (God in Leviticus, his prophet in Deuteronomy) by spatial-temporal momentum (Leviticus gives way to Deuteronomy, life in the desert yields to life in the land). About the view that Deuteronomy serves as a source for the Holiness Code, see in particular E. O tto, “Innerbiblische Exegese im Heiligkeitsgesetz Levitikus 17–26,” in Leviticus als Buch (ed. H. J. Fabry and H. W. Jüngling; BBB 119; Berlin: Philo, 1999), 125–196; C. N ihan , “The Holiness Code between D and P: Some Comments on the Function and Significance of Leviticus 17–26 in the Composition of the Torah,” in Das Deuteronomium zwischen Pentateuch und Deuteronomistischem Geschichtswerk (ed. E. Otto and R. Achenbach; FRLANT 206; Göttingen: Vandenhoeck & Ruprecht, 2004), 81–122; B.M L evinson , “The Birth of the Lemma: Recovering the Restrictive Interpretation of the Covenant Codeʼs Manumission Law 25

278

Jean-Pierre Sonnet

What emerges is a discontinuous yet intentionally driven process in which the divine first-person voice asserts itself: “I have given [blood] to you for [. . .]” (Lev 17:11); “I am YHWH your God; I have separated you from the peoples. You shall therefore make a distinction between the clean animal and the unclean” (Lev 20:24–25; see also 11:44–45). What, in the world of the scribes, was per force a bit-by-bit process becomes, on the narrative stage, an overall design subjectively unfolded by the deity. The progressive legal revelation by supplementation and reformulation stands out, more precisely, as a process of divine self-assertion, regularly marked out by the formula “I am YHWH.”27

2 An Exegetical Prophet In the narrative continuum of the Pentateuch, God is the first to be engaged in legal revision. When Moses undertakes a didactic update of the revelation at Horeb in Deuteronomy, he is just emulating the divine lawgiver, or, more precisely, he has been divinely instructed to do so. Moses relates his commission in Deut 5:25–31, a scene that could be termed the etiological scene of the Torah as Torah, that is, Torah as law enforced by way of teaching.28 In his report, as can be seen in chart 5, Moses circumstantiates the scene of Exod 20:19 where the people request a mediator; as a witness to this event, Moses provides details not given by the narrator in Exodus. In Godʼs answer to the people, an answer now fully reproduced by Moses, an important but subtle rephrasing of the peopleʼs petition can be observed.29 The people had asked Moses to tell them (‫)תדבר‬ whatever God told him (‫)ידבר‬, word for word.30 When God endorses the peopleʼs by the Holiness Code (Lev 25:44–46),” JBL 124 (2005), 617–639; J. Stackert, Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007). 27  The ending of the prescriptive unit of Num 15, where God elaborates on the command to wear fringes on oneʼs garments (Num 15:37–41), is typical: “so you shall remember and do all my commandments, and you shall be holy to your God” (v. 40; see Exod 19:6; 22:30; Lev 11:44; 19:2); “I am YHWH your God, who brought you out of the land of Egypt to be your God” (v. 41a; see Exod 20:2; Lev 22:33; 26:45); “I am YHWH your God” (v. 41b; see in particular the uses in Lev 17–26). 28  The mandate is first mentioned by the narrator in Deut 1:3 (“Moses spoke to the sons of Israel according to all that YHWH had commanded him unto them”), before being alluded to by Moses in 4:14 “And YHWH commanded me at that time to teach you statutes and ordinances for you to do in the land that you are about to cross into and occupy.” “At that time,” that is, during the Horeb theophany, as elaborated in Deut 5:23–31. 29  On this issue, see J.-P. S onnet , The Book within the Book: Writing in Deuteronomy (BibInt 14; Leiden: Brill, 1997), 37–38, 47–48; K. Finsterbusch, Weisung für Israel: Studien zu religiösem Lehren und Lernen im Deuteronomium und seinem Umfeld (FAT 44; Tübingen: Mohr Siebeck, 2005), 166–169. 30  See the double use of the verb in Exod 20:19 in chart 5.

Does the Pentateuch Tell of Its Redactional Genesis?

279

request, he modifies one of the terms: “But you, stand here by me, and I will tell (‫ )ואדברה‬all the commandments, the statutes and the ordinances, that you shall teach them (‫)תלמדם‬, so that they may do them in the land that I am giving them to possess” (Deut 5:31).31 Godʼs subtle and astute shift is nothing less than the rationale of Mosesʼs speeches in Deuteronomy. In reproducing Godʼs integral directive, Moses indeed provides the divine warrant of his own rhetorical enterprise. The word-for-word conveying of the Sinai words has already taken place (primarily in the Covenant Code, which records Godʼs words as spoken by God, but also in the ensuing divine legal revelations, as summarized in Num 30:1); it is now time for a didactic updating, a prescriptive teaching, based on the “told” revelation read in Exod 21–23.32 Chart 5 Exodus 20:19 (the people to Moses)

Deuteronomy 5:27 (the people to Moses)

“Go near, you yourself, and hear all that YHWH our God “You tell (‫ )תדבר‬us, and we will tell. Then tell (‫ )תדבר‬us everything that YHWH our will listen; but do not let God tell (‫ )ידבר‬us, or we will God tells (‫ )ידבר‬you, and we will listen and do it.” die.”

Deuteronomy 5:31 (God to Moses) “But you, stand here by me, and I will tell (‫)ואדברה‬ all the commandments, the statutes and the ordinances, that you shall teach them (‫)תלמדם‬, so that they may do them in the land that I am giving them to possess.”

In the Deuteronomic legal code, Moses intervenes repeatedly as a hermeneutical and didactic prophet, and he immediately shows the measure of his exegetical skill in the law about the centralized altar (Deut 12), a lemmatic reformulation of Exod 20:24–25.33 The theme of diet, examined twice already in this study, provides another example of this ability. Mosesʼs reformulation of the dietary laws in Deut 14 makes the most of the regulation uttered twice by God (Exod 23:19 and 34:26): “You shall not boil a kid in its motherʼs milk.” As in the divine 31  Saying so, God astutely brings into play the biblical principle of “man proposes, God disposes” (see Prov 16:1), i.e., the people propose that you tell us whatever God tells you; God answers that you shall teach them whatever I tell you. Other instances of this phenomenon are found in Gen 11:3, 7; 1 Sam 8:7–9; and 2 Sam 7:5–11. 32  Highlighting Mosesʼs teaching should not overshadow the prescriptive aspect of the mosaic Torah – hence my “prescriptive teaching.” On this issue, see Norbert Lohfinkʼs perceptive observations about the coupling of the verbs ‫למד‬, piel, “to teach,” and ‫צוה‬, piel, “to command,” in Mosesʼs Deuteronomic speeches in N. Lohfink, “Prolegomena zu einer Rechtshermeneutik des Pentateuch,” in Studien zum Deuteronomium und zur deuteronomistischen Literatur V (SBAB 38; Stuttgart: Katholisches Bibelwerk, 2005), 198–199, n. 59. 33  On the subtlety of the lemmatic reformulation of Exod 20:24–25 in Deut 12, see Levinson, Deuteronomy and the Hermeneutics (see n. 16), 23–52.

280

Jean-Pierre Sonnet

pronouncements of this regulation in Exodus, the provision is mentioned by Moses at the very end of a series of commandments (in Deut 14:21b, at the close of the unit 14:3–21). However, it is also detached from its previous context – in both cases, pilgrimage festival offerings – and is reinterpreted “as a general law of food preparation.”34 As an authoritative tag, it is now attached to what was the other dietary law of the Covenant Code – the prohibition of eating carrion (Exod 22:30; Deut 14:20; see also Lev 17:15). This double stipulation follows a distinct body of stipulations (Deut 14:3–20) about clean (“you may eat”) and unclean animals (“you shall not eat”) organized according to the three dominions mentioned in Gen 1:26 (land, water, air).35 For the reader of the Pentateuch, the Deuteronomic Moses thus stands out as the one who collects and harmonizes previous and separate divine regulations about diet and who does so for the sake of the second generation after the exodus – the generation that is about to live according to the complex legal revelation made to the fathers. In other words, the Deuteronomic Moses is narratively cast as a didactic prophet, responding to Godʼs comprehensive revelation and emulating God in his own reformulation: like God, and second to him, Moses is a character who says it again and anew. The prophet is able to do so because he has been the recipient of the revelation from the start. Moreover, he has been prompted to do so because of Godʼs command (see Deut 5:31), and he does so in a (Torah) speech that is a “last chance” speech: the crossing of the Jordan by the people (without Moses) is to take place “this very day” (Deut 9:1). Throughout his Torah speech in Deuteronomy (Deut 4:44–26:19), Moses is engaged in a reformulation process that mirrors the venture of the scribes behind the scenes. The scene in Deut 5:25–31 of Mosesʼs mandate to teach what God has told him is an ingenious expedient whereby the Deuteronomic scribes – engaged in a revision of the Covenant Code (Exod 21–23) – endow their own activity with a divine sanction through Moses. Yet as in the case of the divine character, the mimetic relationship goes in both directions. The ascription of the revised legal tradition to the character of Moses associates legal revision with

 B. M. Levinson, “Deuteronomy,” in The Jewish Study Bible (ed. A. Berlin and M. Z. Brettler; Oxford: Oxford University Press, 2004), 398. 35  The affinity with the more detailed laws in Lev 11 is obvious, even if scholarly inquiry has indicated that the two sets of laws probably represent independent recensions of a common tradition and that no direct textual dependence (in either direction) can be assumed between Lev 11 and Deut 14. See in particular W. S. M orrow , Scribing the Center: Organization and Redaction in Deuteronomy 14:1–17:13 (SBLMS 49; Atlanta: Scholars Press, 1995); C. Nihan, “The Laws about Clean and Unclean Animals in Leviticus and Deuteronomy and Their Place in the Formation of the Pentateuch,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman, K. Schmid, and B. J. Schwartz; FAT 78; Tübingen: Mohr Siebeck, 2011), 401–432. 34

Does the Pentateuch Tell of Its Redactional Genesis?

281

Mosesʼs personal journey.36 The scribes have viewed the Deuteronomic Moses “as their ancestors and their patron,” Karel van der Toorn writes, and in such a guise, as a prophet.37 In other words, the scribes are Moses-like.

Conclusion: Scribes in the Image The mirroring of the exegetical characters on the stage and the scribes behind the scenes turns all the more vivid since, in the venture, both God and Moses have progressively assumed explicit scribal features. God becomes a tablet writer (Exod 24:12) and rewriter (when he makes a duplicate of the first tablets in Exod 34:1, 28).38 Moses emerges as a ‫ספר‬-writer (Exod 17:14), particularly in Exod 24:3–8.39 In true ancient Near Eastern fashion, Moses provides a container – a tablet-box, the wooden ark – for the reinscribed divine words (Deut 10:1–5). And in Deut 31, Moses, who has already completed the writing down of the Torah (v. 9), is ordered to supplement it with the song ‫( האזינו‬vv. 19–22). What is staged is nothing other than a scene of supplementation, the interpolation of a written text within a written text.40 The scene is punctuated with expressions describing Mosesʼs activity as a scribal industry: “When Moses had finished (‫ )ככלות‬writing down in a book the words of this law to the very end (‫”)עד תמם‬ (v. 24).41 Once more, the narrative scenography mirrors redactional activities, the text offering a dramatized mise en abyme of its own making.42  See J. Nohrnberg, Like unto Moses. The Constituting of an Interruption (Bloomington: Indiana University Press, 1995), 325–332. 37  K. van der Toorn, Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007), 167; see also 169. 38  See Sonnet, Book (see n. 29), 42–51, 59–61. 39  As showed by Jean Louis Ska, the double commitment of the people in Exod 24 (“Everything YHWH has said we will do,” v. 3; “Everything YHWH has said we will do and listen,” v. 7) represents a passage from oral to written economy. Though both commitments are about the same legal revelation, the first is performed vis-à-vis Mosesʼs oral report and the second vis-à-vis Mosesʼs written book. See J. L. Ska, “From History Writing to Library Building: The End of History and the Birth of the Book,” in The Pentateuch as Torah (ed. G. N. Knoppers and B. M. Levinson; Winona Lake, IN: Eisenbrauns, 2007), 145–169, here 160–163. 40  See Sonnet, Book (see n. 29), 156–167. 41  Of the colophon “to the very end (‫)עד תמם‬,” Fishbane writes that it “is the precise Hebrew correspondence to the common colophonic notation used in cuneiform literature: qati,” that is, “the end” (M. Fishbane, “Varia Deuteronomica,” ZAW 84 [1972], 350). The verb ‫כלה‬, “to complete,” also has a colophonic use in biblical Hebrew; see H. Gevaryahu, “Set of Remarks about Scribes and Books in Biblical Times,” Beth Miqra 43 (1970), 368–370 (Hebrew). 42  The French term mise en abyme derives from heraldry and literally means “placed into the abyss.” In literature and art theory, mise en abyme refers to the phenomenon in which an image (or a work) contains a smaller copy of itself, in a sequence that can appear to recur infinitely. As pointed out by Lucien Dällenbach, the mise en abyme can specifically focus on 36

282

Jean-Pierre Sonnet

With its cast of scribal characters, its plot of continuity and discontinuity in revelation, and its variegated discursive settings, the Pentateuch is thus a narrative treatise about what biblical legal writing is all about. This trait may be typical of the Bible. The Greek epics and tragedies had accompanying theoretical works, such as the Poetics of Aristotle; Athenian laws had their theoretical counterpart in Platoʼs dialogues in the Laws and the Republic, Aristotleʼs Politics, or the Constitution of the Athenians. No equivalent treatise about the scribal art of Torah making ever accompanied the Pentateuch at its birth or during its growth. The Bibleʼs genius was rather to collapse both levels – the drama of the law and the treatise about it – into one, the one we read in the Pentateuch. Throughout the Pentateuch, God and Moses have done, before anyone else, what the scribes have always done: they have updated, amended, and supplemented. Through the projection of such characters in the narrated story, characters endowed with their own hermeneutical skill, the scribes have authorized – better, canonized – their own work, inscribing their practice in the texture of the founding events. They have cast God and his prophet as scribes in their own image, so as to become scribes in Godʼs and Mosesʼs images.

the production or the reception of the embedding work (L. Dällenbach, Le récit spéculaire: Essai sur la mise en abyme [Paris: Seuil, 1977], 100–122). The most conspicuous mise en abyme in the Hebrew Bible is the staging of the “book of the covenant” in Exod 24:4–7 as internal duplication of the embedding of the book of Exodus (or at least of Exod 20–23). The making, the reading aloud, and the reception of the “book of the covenant” in Exod 24 is a working analogy of the communication process associated with the book of Exodus. See J.-P. Sonnet, “Le Sinaï dans lʼévénement de sa lecture: La dimension pragmatique de Exode 19–24,” NRTh 111 (1989), 322–344; about mise en abyme in Deuteronomy, see Sonnet, Book (see n. 29), 78–83, 246–250. See also D. A. Bosworth, The Story within a Story in Biblical Hebrew Narrative (CBQMS 45; Washington, DC: Catholic Biblical Association of America, 2008).

Continuity between the Gaps The Pentateuch and the Kirta Epic Joel S. Baden In discussions of certain older, and recently rehabilitated, theories of the Pentateuchʼs literary history, one often encounters the same criticisms, whether one reads older anticritical tracts or contemporary alternative presentations. Many of these criticisms fall into the category that might be called perfectionist. That is to say, it is frequently suggested that in order for a documentary theory to hold water it must be flawless; one hole, one gap, in any of the purported sources or in the explanation for their existence and compilation, any place where the present text is difficult to explain, and the entire theory will drain away.1 The logical impetus behind such criticisms is easy enough to see. As one graduate student in a European doctoral program said to me recently, “Iʼd just prefer a theory that can explain every single word over one that doesnʼt.” This is, naturally, hard to argue with. Nevertheless, it might be cautiously suggested that a theory that can indeed explain every single word of the present text is probably a bit overly optimistic about our scholarly ability to reconstruct the complete process of composition. Which is not to say that we can rest content with the mere recognition of areas in the text that are difficult, if not in fact impossible, to explain; it is, rather, that a few of them, here and there, do not necessarily shake the foundations of any given hypothesis. They are reminders that we are dealing with what must have been a historically messy literary project and that perfection of explanation, like perfection of transmission, is probably too much to ask. 1  From the older critics, see especially W. H. Green, The Higher Criticism of the Pentateuch (New York: Charles Scribnerʼs Sons, 1902), 84: “The great body of the Elohim passages are given to the second Elohist [E], and nothing reserved for the first [P] but occasional disconnected scraps, which never could have formed a separate and independent record.” See also his broader comments on 106–9. More recently, Rolf Rendtorff made the same arguments, that the discontinuous state of the Priestly narrative is grounds for declaring untenable the theory that it was an independent document, beginning with this a priori definition of how P must function in a sustainable documentary framework: “It is important that the document being discussed is a coherent P narrative with but few gaps” (R. Rendtorff, The Problem of the Process of Transmission in the Pentateuch [JSOTSup 89; Sheffield: Sheffield Academic, 1977], 137). His discussion continues, in great detail, exclusively along the lines of discontinuity within the P narrative, for the next twenty pages.

284

Joel S. Baden

One of the places where that sort of perfectionism comes into play is in the assessment of the reconstructed sources. Unquestionably with regard to E, but also to a degree with regard to J – and, it must be admitted, in a few places with P as well, though the implications of this are rarely considered anymore2 – when the reconstructed sources of the Documentary Hypothesis are set forth, they have gaps, often significant ones. And these gaps are taken, and have been taken for generations now, as evidence against the existence of the source in question. What is E without a beginning?3 How can J lack any account of Mosesʼs birth or origin?4 Lack of narrative continuity – gaps – becomes the defining feature of nontexts. If something expected or required is missing, then we have not a text unto itself but a fragment, or a layer, or an insertion, or anything, really, other than a self-standing composition. Although we are thankfully past this in the present discussion, one could point, as an example, to Frank Crossʼs famous opinion regarding Pʼs redactional, rather than independent, nature: since P has no covenant at Sinai, and Cross could simply not imagine a narrative without such a covenant, P must have a hole in its narrative and therefore must be redactional.5 It is now recognized that it is actually quite acceptable for P not to have a covenant at Sinai, even as an independent text, and that Cross was reading the canonical text back into his reconstructed source, which is methodologically problematic.6 (Those who long for a formal introduction for Moses in J, it might be suggested, are potentially falling into a similar trap.) Part of the issue, then, is the question of what constitutes a real gap. It cannot be merely the lack of something that I, as a reader, expect to find – whether on the basis of some other similar story, or on the basis of the canonical text, or on the basis of my own aesthetic judgment. I might desperately want to know where Moses comes from in J – who his parents are, what his birth story is, etc. – but I cannot fault the author of J for not providing that information for me. At least, I cannot fault him to the point of denying his existence because I happen not to care for his literary choices.7  See the comments in the previous note, which are directed at P.  Already a sticking point for Green, Higher Criticism (see n. 1), 84. 4  A question leveled more than once against P, in fact; see Rendtorff, Problem (see n. 1), 157; E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 231; J. L. Ska, Introduction to Reading the Pentateuch (Winona Lake, IN: Eisenbrauns, 2006), 147. 5  F. M. Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge: Harvard University Press, 1973), 318–20. 6  See J. S. B aden , The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven: Yale University Press, 2012), 181–82. 7  This does happen, however, with dismaying frequency in scholarship: the aesthetic literary judgments of the scholar are used as evidence for or against the very possibility that anyone could ever have written such a thing in such a way. See, for example, the comments of Marc 2 3

Continuity between the Gaps

285

A true gap, rather, is the absence of information that is absolutely necessary for the narrative to get from point A to point C. There are certainly such gaps in the reconstructed sources of the documentary theory. The J source is missing the beginning of the spies story; it jumps straight from the conclusion of the Kibroth-hatta‍ʾavah episode to Moses telling “them” to go up to the Negev.8 P is missing the account of the births of Jacob and Esau; it jumps straight from the notice that Isaac was forty when he married Rebekah to the notice that Isaac was sixty when “they” were born.9 These sorts of relatively minor gaps are not usually too much of a problem, even for critics of the documentary approach. Under almost any theory such missing pieces can be accounted for by recourse to minor redactional intervention. Still, they are gaps. More substantial, of course, are the gaps in what we call E. The E source, for example, is missing the entire account of Mosesʼs encounter with Pharaoh; it jumps from Godʼs revelation to Moses and instructions regarding his mission straight to the release of the Israelites.10 E has no beginning to speak of: whether one thinks that E begins in Gen 20 or Gen 15, neither of these could be the true beginning of the story.11 These gaps are serious gaps. They are so serious, in fact, that it appears highly likely – indeed, there is almost no other way to explain it Vervenne with regard to the Priestly narrative of Exod 14: “it is not thrilling literature”; “the vocabulary is very stereotypic”; “this strand is without the effects of surprise nor does it move towards a real dénouement” (M. Vervenne, “The ‘P’ Tradition in the Pentateuch: Document and/or Redaction? The ‘Sea Narrative’ [Ex 13,17–14,31] as a Test Case,” in Pentateuchal and Deuteronomistic Studies [ed. C. Brekelmans and Johan Lust; BETL 94; Leuven: University Press, 1990], 67–90 [here 79]). 8  The J story in Num 11 comprises vv. 1–10, the phrase “Moses said to Yahweh” in v. 11, v. 13, the phrase “Yahweh said to Moses” in v. 16, vv. 18–24a, and vv. 31–34 (see B aden , Composition [see n. 6], 82–102). The final verse of the chapter, v. 35, is a redactional insertion linking the J story of Kibroth-hatta‍ʾavah in Num 11 with the E story of Hazeroth in Num 12. The J spies story, for its part, comprises Num 13:17b–20, 22–24, the words “they went to Moses” in v. 26a, 26b–31, 33; 14:1b, 11–25, 39–45. Note that the J spies story is not necessarily lacking mention of any movement from Kibroth-hatta‍ʾavah, as that could well be the staging area for the spiesʼ departure in J. What is lacking is the antecedent of “them” – which means that no more is necessarily missing than a single clause such as “Moses gathered a group of men” or some such. 9  In Gen 25, the verses attributable to P are vv. 7–11a, 12–17, 19–20, 26b. 10  To E belongs Exod 4:18 and then nothing more until 11:1–3. The fundamental scholarly work that restricted the plagues cycle to two sources, rather than the three so often posited in earlier scholarhip, was M. Greenberg, Understanding Exodus (New York: Behrman House, 1969), 183–92. 11  In classical documentary scholarship, the first passage from E was largely thought to be the opening verses of Gen 15, vv. 1–6, or some section thereof. In recent years, this position has essentially been abandoned in favor of two theories that share one significant common feature but are otherwise in opposition. It is almost universally agreed now that Gen 15 is not to be divided into sources but is the work of a single hand. Many scholars believe that hand to be very late; documentary scholars tend to ascribe the chapter in its entirety to E.

286

Joel S. Baden

within the confines of the documentary theory – that the E source was already materially discontinuous before it was combined with the other pentateuchal sources. That is, the compiler of the Pentateuch included in his work an E text that already had substantial gaps when it came into his hands. Though this may appear to be something of a bold claim, I would suggest that objecting to the existence of E on the basis of its discontinuous nature would be a case of putting the cart before the horse. It is no oneʼs a priori assumption that discontinuous texts were necessarily taken up into the Pentateuch at the time of its compilation, just as (though it is often forgotten) no one ever decided in advance that there would be four documents that appear to have been interwoven. It is only after analyzing the text, and coming to the conclusion (whether one accepts it) that the evidence leads to the existence of four sources, that we begin to be able to identify the various states of each of those sources. In other words, the question of how a text could have so many gaps is not an argument against the identification of various passages as belonging together so as to constitute that text in the first place.12 It is a reasonable question to ask of the text once it has been identified – why is this text so gap-filled? – but it is not the basis on which the source was first identified.13 The question of interest for this essay is how or whether we are able to describe as a self-standing text a group of passages that do not necessarily connect to each other perfectly. It is not, to be clear, whether we should be more tolerant of ostensible gaps in the narrative (although indeed we probably should), for instance, the manner in which the episodes in the wilderness seem at times to run into each other without any clear marker of the passage of time or space. There is comparative work to be done that would likely demonstrate that where we see a gap an ancient author or reader would not encounter any difficulty. Rather, if we start from the admission that these passages were not originally continuous, can we still speak of them as constituting an independent literary work? We tend, as a rule, to be less tolerant of gaps in some circumstances than we are in others. Or, to put it another way, there are times when we think that gaps are perfectly acceptable, even perfectly understandable, and we have no difficulty whatsoever in speaking about a single, unified, self-standing composition  The comments of two scholars are worth repeating here. “The gapped connection of the elohistic texts is therefore not a serious argument against the existence of an E source, but merely a statement about its condition” (A. G raupner , Der Elohist: Gegenwart und Wirksamkeit des transzendenten Gottes in der Geschichte [WMANT 97; Neurkirchen-Vluyn: Neukirchener Verlag, 2002], 11; translation mine); “Fragmentary preservation of a work cannot be used as an argument about its original scope” (E. A. Speiser, Genesis [AB 1; New York: Doubleday, 1964], xxxiv). 13  It is worth pointing out here that, among the narrative problems that the documentary theory – or indeed any literary-historical theory – purports to solve, gaps are not one of them. Gaps are a feature of the canonical text that no amount of subdivision, be it into sources or layers, can ameliorate. 12

Continuity between the Gaps

287

that is, for all intents and purposes, continuous despite its patent discontinuities, despite its gaps. The Ugaritic epic of Kirta is preserved in three tablets, none of which is complete. The first column of the first tablet is missing approximately six lines of text. In other words, we do not have the beginning of the Kirta story. What is more, we cannot be sure at all where Kirta started. Although it is widely assumed that those first six missing lines contained some basic exposition of who Kirta is, for all we know there was an entire tablet before this one.14 And, in fact, although the three tablets we now have seem to span a reasonable narrative arc from the loss of Kirtaʼs family to the attainment of a new one, there is nothing to mitigate against the possibility that we have here an episode of a larger original whole, the extension of which, on both ends, has been lost to us. Between the end of the first column and the beginning of the second we are missing text: the majority of Elʼs speech to Kirta. As Kirta restates the content of Elʼs speech, in typical Ugaritic style, no one worries too much about what might have been in this gap.15 Still – it is a gap. In the middle of the fifth column of the first tablet, approximately twelve to fifteen lines are missing, about a third of the column. We know that the missing text must begin with King Pabilʼs speech to his wife and that somewhere in there he stops talking to her and starts instructing the messengers he will send to Kirta. But beyond that nothing more can be said. The transition from the first tablet to the second is deeply convoluted. The last thing we read on the first tablet is the return of the messengers to Pabil, beginning the repetition of Kirtaʼs reply. The first thing we read on the second tablet is apparently the departure of Pabilʼs daughter Hurraya, Kirtaʼs new queen. This first column of the second tablet is missing around forty lines, or a full five-sixths of its content. Some of what was there can be reconstructed: the report of the messengers, for example, again following typical Ugaritic style. But there is much that remains unknown. In fact, some have conjectured that we could be missing an entire tablet or more here.16 Who knows what may have happened between the delivery of Kirtaʼs message and the final delivery of Hurraya to her new husband? The transition from the first to the second column of the second tablet is also sticky. About twenty lines are missing, in which the scene seems to have 14  This was the claim made in the seminal analysis of Kirta by H. L. Ginsberg, who posited the existence of at least one tablet preceding what we now consider the beginning of Kirta. See H. L. Ginsberg, The Legend of King Keret: A Canaanite Epic of the Bronze Age (BASORSup 2–3; New Haven: American Schools of Oriental Research, 1946), 8. 15  Indeed many scholars simply fill in the lines completely; see, e.g., N. Wyatt, Religious Texts from Ugarit: The Words of Ilimilku and His Colleagues (BibSem 53; Sheffield: Sheffield Academic Press, 1998), 184. 16  See, e.g., K.-H. Bernhardt, “Anmerkungen zur Interpretation des KRT-Textes von Ras Schamra-Ugarit,” Wissenschaftliche Zeitschrift der Ernst Moritz Arndt-Universität Greifswald: Gesellschaftltiche und Sprachwissenschaftliche Reihe 5 (1956): 102–21 (here 111).

288

Joel S. Baden

shifted from Pabilʼs kingdom back to Kirtaʼs, and invitations are going out for the wedding feast. Another fifteen-line gap exists between the second and third columns, between the announcement of the future birth of Kirtaʼs eldest son Yasib and the listing of his six future daughters; the gap presumably contained the list of Kirtaʼs six remaining future sons. About twelve lines are missing between columns three and four. We leave off with Athirat beginning to castigate Kirta for breaking his vow and pick up again with Kirta calling to Hurraya. What precisely did the goddess say? Another fifteen lines are missing between columns four and five, which must, based on the context, contain Kirtaʼs dictation to Hurraya of her forthcoming speech to Kirtaʼs nobles, and another twenty-five lines between columns five and six in which she presumably delivers that speech. We lack forty lines at the end of the second tablet, after the arrival of Kirtaʼs nobles and before the beginning of the third tablet, in which Kirtaʼs son Ilhu is (perhaps) being told what he should say upon entering his fatherʼs presence.17 Is it possible that we are missing a tablet or more between the second and third tablets? It is possible, even if it might not be likely. What did the nobles say to Kirta? What else took place? Is this where Kirta falls ill? How much time passes here? At the end of the first column of the third tablet, Ilhu is telling his sister Thitmanit that Kirta is not ill but is holding a banquet; after another fifteen-line gap or so, she is returning to Ilhu and challenging him with the fact of Kirtaʼs illness, which, in that gap, she must have witnessed for herself. Almost forty lines of missing text later, we are in what seems a completely different world. Someone is speaking to someone else – who either might be is entirely unclear – about a famine. The famine is not anticipated, nor is it ever mentioned again.18 Thirty-five missing lines later, a divine herald is being summoned, perhaps by El, perhaps not. Another thirty-five-plus missing lines later, El is trying to find a deity of healing. It is fair to presume that in that gap the gods have been summoned, though it is quite possible that something more was also related there. The end of the fifth column is a shambles, with only a few words preserved, 17  For the suggestion, see E. L. G reenstein , “Kirta,” in Ugaritic Narrative Poetry (ed. S. B. Parker; SBLWAW 9; Atlanta: Scholars Press, 1997), 9–48 (here 30). Wyatt suggests perhaps it is a deity who is providing Ilhu with the instructions (Wyatt, Religious Texts [see n. 15], 221). 18  The common assumption has been that expressed by B. M argalit , “The Legend of Keret,” in Handbook of Ugaritic Studies (ed. W. G. E. Watson and N. Wyatt; Handbuch der Orientalistik; Leiden: Brill, 1999), 203–33 (here 229): “The illness of the king induces a paralysis of Nature.” See similarly G reenstein , “Kirta” (see n. 17), 35; W yatt , Religious Texts (see n. 15), 232. But it is admitted also that “the text of this scene is very fragmentary, and its contents consequently are obscure” (Margalit , “Legend of Keret,” 228). In every case, however, the direct line between Kirtaʼs illness and the drought is simply stated as fact, with no explicit grounding in the text.

Continuity between the Gaps

289

though it is clear from the context that El has created a clay goddess of healing, Shatiqat. The sixth column is, blessedly, preserved almost entirely and even concludes with a colophon from our friendly neighborhood scribe, Ilimilku. Nevertheless, many scholars have argued that the story must have continued on – perhaps in an earlier recension, shortened by Ilimilku, or perhaps somehow even beyond the colophon.19 To summarize, these are the phenomena we see in the Kirta epic: It has no beginning. It contains a number of small gaps that, by virtue of cross-reference or repetition, we can fill in almost entirely. It contains a larger number of substantial gaps where we can guess as to at least part of what must have been there but cannot reconstruct it with any certainty. It quite possibly contains one or more gaps of unknown length, potentially entire tablets. The scenes shift precipitously over space and time, with the shift identifiable only through contextual clues rather than any explicit markers of such. There is at least one passage that has absolutely no literary connection with any other part of the text. There are thematic elements that appear only in one of the tablets – notably, for instance, the idea of Kirtaʼs illness, which is absolutely central to the third tablet but which is unmentioned in the first two; conversely, the vow to and curse of Athirat, which is the major plot element in the first two columns, is entirely absent in the third, where Kirtaʼs survival is chalked up to the triumph of Shatiqat, Elʼs creation, over Mot – though one would think that Kirtaʼs death could be avoided only by appeal to Athiratʼs leniency. Yet she is never mentioned. Despite the gaps, shifts, discontinuities, and unexpected changes, we still speak of a Kirta epic. Indeed, no one suggests that what we have here may in fact be literarily unrelated fragments – at most it has been suggested that the present Kirta epic is a combination of previously independent Kirta traditions, an ostensibly reasonable suggestion that has not found much favor.20 In fact, those who argue for a full-blown Kirta epic use these gaps almost as support for such a claim: by filling in the gaps, sometimes quite expansively, we can see how Kirta must have originally had a nice full form. Gaps are not a detriment – they are simply a part of the story. The basic thrust of the comparison between the Kirta epic and the sources of the Pentateuch is surely clear enough. Of course one is not unconscious of the significant material differences between Kirta and any given pentateuchal source. Kirta is written on tablets, and exists – however one thinks of it – independently from any other texts. The pentateuchal sources exist only as the result  See, e.g., S. B. Parker , The Prebiblical Narrative Tradition: Essays in the Ugaritic Poems Keret and Aqhat (SBL Resources for Biblical Study 24; Atlanta: Scholars, 1989), 145; Ginsberg, Legend of King Keret (see n. 14), 8–9. 20  This view is associated with A. M errill , “The House of Keret: A Study of the Keret Legend,” SEÅ 33 (1968): 5–17, and S. B. Parker, “The Historical Composition of KRT and the Cult of El,” ZAW 89 (1977): 161–75. 19

290

Joel S. Baden

of scholarly inquiry, are written on our minds and hearts, and were interwoven with other texts. Yet the driving question here is not whether E is to be equated with Kirta in any ham-handed sort of way, but is rather what we consider evidence of, or evidence against, continuity. Given all of the discontinuities in Kirta, there remain ostensibly good reasons to talk about it as a composition. There are the superscriptions, to begin with: the first and third tablets open with the title l-kirta, which would seem to connect them. Then again, there are the seventy-three psalms that open with l-David, and we do not consider those all to be a single composition. There is the fact that Kirta is found on tablets in the first place, so that we are forced on material grounds to read it through, at least within each tablet. Then again, there is nothing to say that multiple originally independent compositions could not be collected onto a single tablet, just as, for example, KTU 1.114 has the narrative of Elʼs drunkenness and a related, though surely originally independent, recipe for a hangover. We are all perfectly familiar, in the biblical realm, with multiple compositions being collected onto a single scroll. What convinces scholars of the continuity of Kirta is not necessarily its material features but rather the internal coherence of the narrative. Across the gaps we find the same characters at work: Kirta and Pabil, Hurraya and Yasib, El and Athirat. We find clear references to the same events, though we pick the story up after a large break: the embassies between Kirta and Pabil, or Kirtaʼs banquet, or the fear of Kirtaʼs imminent death. We recognize that there are common threads winding through these tablets and the bits and pieces written on them. Even when we know we are missing texts we can see how what we do have is internally coherent – or, sometimes, we create rationales to make what we have internally coherent. We accept the continuity of Kirta, and the existence of a broad Kirta epic, despite the fact that we do not have a beginning for it; despite the fact that we cannot directly connect one piece to the next; despite the fact that, in some places, we can barely understand what certain passages are doing, indisputably, in the middle of the story, and despite the fact that the story as a whole is less than perfectly clear. Scholars are uncertain what to make of the Kirta epic: Is it royal propaganda? Is it historical, somehow? Is it satire?21 The “truth” may not really matter here; what matters is that, again, despite its poorly preserved state, which perhaps makes it impossible to know what the text “means,” the “meaning” of Kirta is still debated. Interpretive worlds are built on the basic assumption that we have a text here that can be interpreted, that has enough existence of its own, enough integrity as a composition, to be treated as such.

21  For a survey of the varied history of interpretation, see Margalit, “Legend of Keret” (see n. 18), 204–18.

Continuity between the Gaps

291

Now to be slightly ham-handed about it: I would posit that the reconstructed E document of Neo-Documentary source criticism is far, far less problematic than the Kirta epic. It too is lacking a beginning, and, again, is of unknown length. It too has gaps, many of which can be filled in, at least in general outlines, with little difficulty. It too has one or two substantial gaps, in which we can suggest some events that may have taken place, but perhaps not all. But, unlike Kirta, it also has lengthy stretches of ostensibly uninterrupted narrative; unlike Kirta, it contains no passages that seem to have absolutely zero relationship with anything else in the story; unlike Kirta, there are always close narrative, thematic, and linguistic links even across the gaps. If you gave an impartial observer the E narrative and the Kirta narrative and asked which one was more likely to be an originally continuous whole, despite its present tattered shape, I think there would be no question that the answer would be E. To return to the objection that I already anticipated above, Kirta and E are not comparable texts, because we have Kirta, we can hold it in our hands, and E does not exist except in the minds of a handful of scholars. To this, I offer three responses. First: we should not overstate the material unity of the Kirta epic. It exists on three separate tablets, only two of which are marked with the superscription l-krt. That the three belong together seems obvious now, but it remains to some extent a common scholarly decision rather than a “fact” apart. Indeed, in its initial publications, the tablets were presented in what we would now consider the wrong order, with what is now the second tablet coming at the end.22 The material unity of Kirta is predicated on the literary judgment that these three materially separate tablets belong together and always belonged together. Second: if the existence of a text as nothing other than theoretical reconstruction means that we cannot talk about its potential original shape, its potential gaps and what may have occupied those spaces, or its themes and function despite our uncertainty about its beginning and ending, then scholars probably have to stop talking about P; frankly the entire enterprise, which consists of nothing other than reconstructing literary texts and holding them up for consideration, would be thrown into disarray. What documentary scholars say about E is different from what virtually everyone says about P only in degree, not in kind. Some scholars, such as the present author, may be willing to tolerate more imperfection than are most. But almost everyone tolerates a little of it. And every pentateuchal scholar talks about texts that exist only in our minds. Third: this essay is not an attempt to convince anyone that E exists. It is after a methodological point: not that E exists but that the argument from gaps – the argument that if the reconstructed text is not perfectly continuous then we cannot talk about it as a self-standing composition – is a faulty one. Why must 22  Margalit, “Legend of Keret” (see n. 18), 203. There is similar scholarly disagreement, it may be noted, about the order of the tablets in the Baal Cycle.

292

Joel S. Baden

a reconstructed text of E be held to a higher standard than the actually existing text of Kirta, again, not as to whether it exists but as to whether its literary shape is worthy even of consideration as a text in its own right? If Kirta is assumed to be a self-standing unified composition, despite its abundant flaws, then the flaws undeniably present in E should not be a necessary impediment to its receiving the same status. The process of identifying E in the Pentateuch is not dependent on the existence of gaps or the existence of continuity. As with the other sources, E is identified by recognizing narrative contradictions between some texts and narrative agreements between others. As the passages that belong exclusively with each other and stand in opposition to the rest of the text are collected and a profile begins to emerge, then the question can be asked as to why these passages seem not to be perfectly continuous at points. It should be noted that the first step, of identifying the passages that belong together, is often accomplished along similar lines by both documentary and nondocumentary scholars.23 But nondocumentary scholars then take these disparate passages and attribute them to a redactional layer rather than considering the possibility that they could have constituted a self-standing text unto themselves. Whether it is explicitly stated or merely implicit, it seems likely that this is due to a suspicion that in the absence of observable narrative continuity there can be no consideration given to the possibility of narrative independence. Continuity is a sliding scale. It depends on what sort of text we are dealing with, what degree of preservation we expect, what the age of the text is, and, most important, who is doing the judging and what sorts of results said judge wants to find. There is no denying the subjectivity inherent in determining what is continuous and what is not, what is literarily tolerable and what is not. I may find my reconstructed J text far more readable than another scholar might. That is perfectly acceptable. But within our own constructed systems of continuity, we should at least try to be consistent from text to text. And I would again contend that – everything else being equal – in terms of continuity, at least, the pentateuchal sources of the documentary theory all, even E, are far more continuous than a text like Kirta. The existence of the pentateuchal sources may be denied if it must – but not on these grounds.

23  One may note how regularly groups of passages that documentary scholars assign to E are isolated and linked together by scholars who do not work in a documentary framework. See, e.g., A. H. J. Gunneweg, “Das Gesetz und die Propheten: Eine Auslegung von Ex 33,7–11; Num 11,4–12,8; Dtn 31,14f.; 34,10,” ZAW 102 (1990): 169–80; more strikingly, E. Blum, “The Decalogue and the Composition History of the Pentateuch,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 289–301.

Part Three The Role of Historical Linguistics in the Dating of Biblical Texts

Introduction The Strengths and Weaknesses of Linguistic Dating Shimon Gesundheit Since Geseniusʼs pioneering diachronic study of Biblical Hebrew in Geschichte der hebräischen Sprache und Schrift,1 it has become commonplace to broadly divide the language reflected in the prose parts of the Hebrew Bible into two chronolects: preexilic and postexilic. The underlying rationale for this distinction is that the exilesʼ forced immersion in the culture and language(s) of Babylonia would have dealt a heavy blow to the Hebrew language of the time, influencing its syntax, reshaping its grammar, and suffusing it with numerous loanwords. Needless to say, the field of linguistic dating has advanced a great deal in the ensuing years, though the point of departure remains that of Gesenius. The methodology has evolved, and new data has been considered. Archaeological discoveries, including epigraphic material, have enriched our understanding of the biblical era, and advances in historical-critical analysis have led to a rethinking of certain conclusions. In the last two years alone, several important works relating to this topic have appeared, including Diachrony in Biblical Hebrew, edited by Cynthia Miller-Naudé and Ziony Zevit; Early Biblical Hebrew, Late Biblical Hebrew, and Linguistic Variability: A Sociolinguistic Evaluation of the Linguistic Dating of Biblical Texts, by Dong-Hyuk Kim; Historical Linguistics and Biblical Hebrew: Steps Toward an Integrated Approach, by Robert Rezetko and Ian Young; and A Concise Lexicon of Late Biblical Hebrew: Linguistic Innovations in the Second Temple Period, by Avi Hurvitz, in collaboration with Leeor Gottlieb, Aaron Hornkohl, and Emmanuel Mastéy.2 I would like to thank Idan Dershowitz for his invaluable assistance in the preparation of this article. 1  W. G esenius , Geschichte der hebräischen Sprache und Schrift: Eine philologischhistorische Einleitung in die Sprachlehren und Wörterbücher der hebräischen Sprache (Leipzig: Friedrich Christian Wilhelm Vogel, 1815). 2  C. Miller-Naudé and Z. Zevit (eds.), Diachrony in Biblical Hebrew (Winona Lake, IN: Eisenbrauns, 2012); D.-H. Kim, Early Biblical Hebrew, Late Biblical Hebrew, and Linguistic Variability: A Sociolinguistic Evaluation of the Linguistic Dating of Biblical Texts (Leiden: Brill, 2013); R. Rezetko and I. Y oung , Historical Linguistics and Biblical Hebrew: Steps Toward an Integrated Approach (Atlanta: Society of Biblical Literature, 2014); A. Hurvitz et al., A Concise Lexicon of Late Biblical Hebrew: Linguistic Innovations in the Second Temple Period (VTSup 160; Leiden: Brill, 2014).

296

Shimon Gesundheit

Alas, despite this flurry of activity across the globe, it cannot be said that the discipline has moved toward a consensus with regard to the dating of biblical texts, linguistically or otherwise. Rather, we find widely divergent approaches and conclusions and little agreement regarding the dating of parts of the Hebrew Bible whose content does not explicitly address postexilic events. Even when a work is indisputably postexilic, there is inevitably a significant range of opinions about the date of its composition that vary from the Persian era to the Hellenistic. What we do find, perhaps unfortunately, is “camps” or “schools.” Over the past twenty-odd years in particular, we have witnessed a separation of the field into so-called minimalists and maximalists (the terms are of course problematic), and, as the terms themselves suggest, there exists a broad continuum between the two extremes. Furthermore, it should be noted that this particular charting of scholarly opinion is based on the types of conclusions scholars draw. One could just as well plot the map according to methodology or according to various other considerations. Nevertheless, this perspective is current in todayʼs discourse, making it a natural point of departure. I will begin my survey with Avi Hurvitz. Hurvitzʼs approach is among those most frequently cited in the present day, and naturally it has a large number of supporters and – especially in recent years – critics. The former group includes, to varying degrees (listed alphabetically), Jan Joosten, Noam Mizrahi, Frank Polak, Gary Rendsburg, Mark Rooker, Choon-Leong Seow, and Richard Wright, among others.3 According to Hurvitz and most of his followers, the great majority of compositions that are not explicitly exilic or later are in fact preexilic.4 Thus, the various sources of the Pentateuch and Former Prophets were composed in the monarchic period.5 On the other end of the spectrum we find Joseph Blenkinsopp, Frederick Cryer, Philip Davies, Ernst Axel Knauf, Niels Lemche, Jacobus Naudé, Sverrir Ólafsson, Robert Rezetko, Thomas Thompson, Ian Young, and 3  E.g., J. J oosten , “The Distinction between Classical and Late Biblical Hebrew as Reflected in Syntax,” HS 46 (2005), 327–339; N. Mizrahi, “The History and Linguistic Background of Two Hebrew Titles for the High Priest,” JBL 130 (2011), 687–705; F. Polak, “The Oral and the Written: Syntax, Stylistics and the Development of Biblical Prose Narrative,” JANES 26 (1998), 59–105; G. A. Rendsburg, “Some False Leads in the Identification of Late Biblical Hebrew Texts: The Cases of Genesis 24 and 1 Samuel 2:27–36,” JBL 121 (2002), 23–46; M. F. Rooker, Biblical Hebrew in Transition: The Language of the Book of Ezekiel (JSOTSup 90; Sheffield: JSOT Press, 1990); C.-L. Seow, “Linguistic Evidence and the Dating of Qohelet,” JBL 115 (1996), 643–666.; R. M. Wright, Linguistic Evidence for the Pre-Exilic Date of the Yahwistic Source (London: T&T Clark, 2005). These scholarsʼ research domains span a broad spectrum including syntax, comparative linguistics, sociolinguistics, and dialect. 4  Hurvitz himself was strongly influenced by his teacher, Yechezkel Kutscher, who covered related ground; see E. Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll: IQIsaa (Leiden: Brill, 1974) and other writings. 5  See Hurvitz, Concise Lexicon (see n. 2), 3.

Introduction

297

others, most of whom date the bulk of the biblical corpus to the Persian and/or Hellenistic periods.6 These respective conclusions are, of course, tied up with methodological considerations, and I would therefore like to address some of the strengths and weaknesses of each campʼs approach, specifically with regard to linguistic arguments. Hurvitz is frequently credited with greatly advancing the fieldʼs methodology. Rooker, for instance, has said that Hurvitz has single-handedly contributed “the most detailed treatment of the methodology for diachronic research,”7 and many others simply treat Hurvitzʼs methodology as the benchmark for good scholarship in the field.8 Hurvitz himself has called out his detractors for a lack of methodological rigor, suggesting that their theories are therefore “castles in the air.”9 6  J. Blenkinsopp, “An Assessment of the Alleged Pre-Exilic Date of the Priestly Material in the Pentateuch,” ZAW 108 (1996), 495–518; F. H. Cryer, “The Problem of Dating Biblical Hebrew and the Hebrew of Daniel,” in In the Last Days: On Jewish and Christian Apocalyptic and Its Period (ed. K. Jeppesen et al.; Aarhus: Aarhus University Press, 1994), 185–198; P. R. D avies , In Search of “Ancient Israel” (JSOTSup 148; Sheffield: Sheffield Academic, 1992); idem, “Biblical Hebrew and the History of Ancient Judah: Typology, Chronology and Common Sense,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 150–163; E. A. Knauf, “War ‘Biblisch-Hebräisch’ eine Sprache? Empirische Gesichtspunkte zur linguistischen Annäherung an die Sprache der althebräischen Literatur,” ZAH 3 (1990), 11–23; N. P. L emche , “The Old Testament – A Hellenistic Book? (Revised Version),” in Did Moses Speak Attic? Jewish Historiography and Scripture in the Hellenistic Period (ed. L. L. Grabbe; JSOTSup 317; Sheffield: Sheffield Academic, 2001), 287–318; J. A. N audé , “The Transitions of Biblical Hebrew in the Perspective of Language Change and Diffusion,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 189–214; idem, “A Perspective on the Chronological Framework of Biblical Hebrew,” JNSL 30 (2004), 87–102; S. Ólafsson, “Late Biblical Hebrew: Fact or Fiction?” in Intertestamental Essays in Honour of Józef Tadeusz Milik (ed. Z. J. Kapera; Qumranica Mogilanensia 6; Krakow: Enigma, 1992), 135–147; R. R ezetko , “The Spelling of ‘Damascus’ and the Linguistic Dating of Biblical Texts,” SJOT 24 (2010), 110–128; T. L. Thompson, “The Intellectual Matrix of Early Biblical Narrative: Inclusive Monotheism in Persian Period Palestine,” in The Triumph of Elohim: From Yahwism to Judaism (ed. D. V. Edelman; CBET 13; Kampen: Kok Pharos, 1995), 107–124; I. Y oung and R. R ezetko , Linguistic Dating of Biblical Texts (2 vols.; London: Equinox, 2008). 7  M. F. Rooker, “Diachronic Analysis and the Features of Late Biblical Hebrew,” BBR 4 (1994), 135–144, here 137. 8  See, e.g., W. M. Schniedewind, “Steps and Missteps in the Linguistic Dating of Biblical Hebrew,” HS 46 (2005), 377–384, here 381–382: “For the diachronic study of the Hebrew language to go forward, it must pay careful – even pedantic – attention to primary sources using the methodology that Hurvitz has long employed.” 9  A. H urvitz , “The Historical Quest for ‘Ancient Israel’ and the Linguistic Evidence of the Hebrew Bible: Some Methodological Observations,” VT 47 (1997), 301. Quote attributed to Wilhelm von Humboldt.

298

Shimon Gesundheit

There is no question that the fieldʼs movement toward more objective criteria is a positive one. However, some have identified what appear to be fundamental biases in Hurvitzʼs methodology.10 Hurvitz takes pride in being exceedingly rigorous in his demand for a confluence of evidence in order to conclude that a given text is late. He demands that three strict criteria be met in order to abandon that default and concede that a book is in fact postexilic: Fulfillment of the first condition [late distribution in the Bible] is necessary in order to even propose the thesis that there is room to doubt the antiquity of the linguistic item in question. Fulfillment of the second condition [prevalence in late external sources] suggests that the late distribution in the Bible is not due to chance and is not devoid of chronological significance, but, on the contrary, faithfully reflects a broad, authentic linguistic reality that existed in the late period – even beyond the limited boundaries of the biblical corpus that we possess today. Finally, fulfillment of the third condition [presence of alternative linguistic usages in the classical stratum] indicates that one should not explain the absence of the suspect item in the classical stratum as due to a lack of appropriate context, or suitable opportunity, to use it in texts formulated in Classical Biblical Hebrew. As noted above, only compound evidence supported by all three of the aforementioned criteria would be sufficient to provide a solid foundation for the conclusion that a given linguistic phenomenon faithfully reflects “Late Biblical Hebrew.”11

It is worth dwelling on Hurvitzʼs suggestion that a linguistic item must be found in incontrovertibly late contexts to permit “room to doubt [its] antiquity.” It is difficult to escape the impression that we are dealing here with a philosophy according to which all biblical texts are early until proven late. This and other similar statements indicate that, according to Hurvitz, every work has a default status, and that status is preexilic. Every ostensible stringency is a leniency, as well. To be stringent about determining that a text is late is to be lenient in assuming the textʼs antiquity. Hurvitz acknowledges the existence of several categories of linguistic variance, such as register and dialect, but he nevertheless interprets almost all instances of variation diachronically – typology is almost never separated from chronology in Hurvitzʼs writings. This seems problematic, given that there is precedent for ancient linguistic varieties surviving intact for several centuries. For instance, Eleanor Dickey wrote in her recent book, Ancient Greek Scholarship:

10  Davies was among the first to do so, in 1992 (Davies, In Search [see n. 6], 98), though the critical trend has picked up steam in the past fifteen years. More recently, see Young and Rezetko, Linguistic Dating of Biblical Texts (see n. 6), 1:55–56, 92–93. 11  A. H urvitz , “The Archaeological-Historical Debate on the Antiquity of the Hebrew Bible in Light of Linguistic Research of the Hebrew Language,” in The Controversy over the Historicity of the Bible (ed. L. I. Levin and A. Mazar; Jerusalem: Yad Ben-Zvi and Merkaz Dinur, 2001), 38 (Hebrew); my translation and emphasis.

Introduction

299

Second-century authors like Lucian learned to produce literary works in nearly flawless imitations of fifth-century Attic, and even in other classical dialects. The perfection of these imitations is especially impressive considering that non-literary Greek (as seen for example in documentary papyri) had undergone considerable evolution in the intervening five or six centuries, becoming a language markedly different from that of Plato or Herodotus.12

Most of Hurvitzʼs analyses over the years have been of lexical items.13 This decision has its shortcomings, however, as lexemes, though prime examples of rapidly evolving linguistic features,14 are also particularly prone to editorial updating, as apparent from even a superficial comparison of the various textual witnesses of the Bible.15 Lastly, methodology cannot be considered rigorous if it is not applied consistently. For example, Noam Mizrahi has a paper in this volume in which he shows that ‫ אחד עשר‬vs. ‫ עשתי עשר‬has all the markers of a “Hurvitzian” pair, and yet Hurvitz did not include it in his own dataset. The fact is that ‫ עשתי עשר‬looks very much like an Aramaism and nevertheless appears in books that Hurvitz believes to be preexilic. Ultimately, Mizrahi reaches the conclusion that ‫עשתי‬ ‫ עשר‬made its way into Hebrew quite early on, but be that as it may, it is hard  E. D ickey , Ancient Greek Scholarship (London: Oxford University Press, 2007), 9. Some may emphasize the word “nearly” in Dickeyʼs statement, suggesting that it is precisely the scattered minor errors that disclose a late work and that these are absent in disputed biblical works. However, one cannot have oneʼs cake and eat it too. After all, Hurvitzʼs approach calls for a concentration of late features to be present, and not just scattered occurrences, for a text to be deemed late. See, e.g., A. Hurvitz, “Linguistic Criteria for Dating Problematic Biblical Texts,” Hebrew Abstracts 14 (1973), 74–79: “The text will not be considered late unless it manifests numerous late elements – one or two isolated examples can always be interpreted as a coincidence” (76). Similarly, in A. Hurvitz, “The Recent Debate on Late Biblical Hebrew: Solid Data, Expertsʼ Opinions, and Inconclusive Arguments,” HS 47 (2006), 191–210, he writes: “A text will not be declared late unless it contains a heavy accumulation of neologisms” (194). See also H. M. G ass , review of Lucian, Satirist and Artist, by Francis G. Allinson, Sewanee Review 35 (1927), 372–374: “In the days when Greek had fallen as far as possible from the classical heights, in the second century of the Christian era when the world was under Roman sway, a Syrian, taking Plato and Aristophanes as his patterns [. . .] wrote a pure and flawless Attic Greek” (372; my emphasis). The continued use of principally archaic chronolects in specialized contexts is not unusual. Cf. Late Middle Egyptian, also known as égyptien de tradition, and Classical (Pāṇinian) Sanskrit, inter alia. 13  Several other scholars have worked to fill in these gaps. See, for instance, R. P olzin , Late Biblical Hebrew: Toward an Historical Typology of Biblical Hebrew Prose (Missoula, MT: Scholars Press, 1976); Joosten, “Distinction” (see n. 3), 327–339; Rooker, “Diachronic” (see n. 7), 135–144. 14  J. A itchison , Language Change: Progress or Decay? (3rd ed.; Cambridge, Engl.: Cambridge University Press, 2001), 16–17. 15  M. E skhult , “The Importance of Loanwords for Dating Biblical Hebrew Texts,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 8–23. 12

300

Shimon Gesundheit

to understand why Hurvitz did not address such a salient lexical pair.16 In light of this issue, it is perhaps prudent to reconsider some of Hurvitzʼs conclusions, such as, “P in its final – extent [sic] – stage of crystallization, does not betray distinctive linguistic elements which post-date the classical phase of BH.”17

***** I now return to the second camp in the field of linguistic dating, namely, those scholars who attribute the vast majority – if not all – of the biblical corpus to the postexilic or even late postexilic period.18 One difficulty with their approach may be summed up by a paraphrase of Franz Delitzschʼs famous statement: “If the book of Koheleth were of ancient Solomonic origin, then there is no history of the Hebrew language.”19 To wit: “If the book of Samuel were of late Hellenistic origin, then there is no history of the Hebrew language.”20 A common response to this argument is that the variation present in the biblical corpus results from synchronic phenomena rather than a long history of linguistic evolution.21 Some, such as Young et al., accept the existence of Classical Biblical Hebrew (CBH)  For a similar example, see Blenkinsopp, “Assessment” (see n. 6), 506.  A. Hurvitz, “Once Again: the Linguistic Profile of the Priestly Material in the Pentateuch and Its Historical Age: A Response to J. Blenkinsopp,” ZAW 112 (2000), 180–191, here 191. 18  For a list of some of this campʼs representatives, see note 6. 19  F. Delitzsch, Commentary on the Song of Songs and Ecclesiastes (trans. M. G. Easton; Edinburgh: T&T Clark, 1891), 190. 20  This argument is the linguistic analog of Brettlerʼs critique of Lemcheʼs literary characterization of the Bible: “This is too flat a description of the Bible, ignoring its richness and variegation [. . .]. A counter-hypothesis might see the Bible as a more complex work, from various periods, reflecting various degrees of historicity, and having a multiplicity of purposes”; M. Z. Brettler, “The Copenhagen School: The Historiographical Issues,” AJSR 27 (2003), 1–21, here 15. Also, as a counterpoint, one must consider whether there is enough variation to allow for a millennium of linguistic development, as many believe is the case. 21  It is worthwhile quoting Lemche at length: “The remarkable qualitative distance between Chronicles and the Deuteronomistic History is not only a distance in time [. . .] which may be short or long; it may just as easily bear witness to the fact that the two histories were composed in very different environments [. . .]. This discussion sounds like a new version of the debate concerning the respective date of J, E and P that has lasted for more than a century. It is well known that much of this discussion was based on arguments like differences of religious or political outlook, on linguistic matters, etc. It was also assumed that the Yahwist was a more simple-minded fellow (although by all means a great narrator) than his Elohist colleague, and that the persons behind P display a view of religion (if not theology) that is very different from the one found in J and E. Such differences were explained a long time ago by Johannes Pedersen as not necessarily the consequence of a difference in time; they could just as well be the outcome of different milieus and/or abilities and preferences of their authors. To deny that Pedersenʼs argument is valid would be the same as maintaining that Plato could not be a contemporary of Xenophon!” (Lemche, “Old Testament” [see n. 5], 314). 16 17

Introduction

301

and Late Biblical Hebrew (LBH) as meaningful categories, even if they are not chronolects per se.22 This is in contrast to Cryer, who rejects the dichotomy altogether.23 These scholars appear sometimes to take an all-or-nothing approach. If a feature appears in both CBH and LBH, then it is utterly useless for dating, even if that feature appears predominantly in just one of the corpora. Young et al. sum up the counterargument as such: “Even when trying to write ‘pure Classical Hebrew’, authors in the postexilic period inevitably failed, leaving tell-tale traces of their actual date in the form of LBH linguistic items.” They then conclude: “We have seen that this argument cannot work except insofar as it leads to the conclusion that all the biblical texts were written in the postexilic period, since all BH texts contain LBH linguistic features.”24 Again, there seems to be no good reason to treat a case of 50:50 as identical to a case of 99:1. A related problem with these scholarsʼ proposal is the difficulty in explaining the coincidence of obviously late prose books almost invariably being written in the LBH “style,” if it is not in fact a late chronolect. Why should Chronicles, Ezra, Nehemiah, Daniel, and Esther – all of which present themselves as late compositions – be written in the same style? Why arenʼt any of these books written in CBH? Young suggested in the past that these books were composed in the eastern diaspora, meaning that LBH and CBH are dialects.25 Today he and his colleagues argue that these books were written as a “conscious attempt to distance the style of literature from literature produced in the EBH style.”26 In other words, Young now believes that CBH and LBH are registers. Either way, it remains difficult to understand why not one single explicitly late book was written outside of the eastern diaspora or, alternatively, without an attempt to distance its style from the norm. On a final note, it is undoubtedly convenient for those who attribute the CBH/ LBH dichotomy to synchronic sociolinguistic phenomena that their system is immune to discrepant data. Should an allegedly late work exhibit signs of antiquity, it would pose no problem, since late writers could have chosen a quasi-early register for their compositions, even if their reasons for doing so may be unknown. And should a CBH work be found to contain late features, this too would be taken in stride, as its presumed postexilic date of composition

 Young and Rezetko, Linguistic Dating of Biblical Texts (see n. 6), 2:72–105, et passim.  Cryer, “Problem of Dating Biblical Hebrew” (see n. 6), 186–187. 24  Young and Rezetko, Linguistic Dating of Biblical Texts (see n. 6), 2:97. 25  For a different perspective on the interplay between dialect and chronolect, see the numerous publications of Rendsburg and his students. This line of inquiry has its roots in the work of Rendsburgʼs own teacher, Cyrus Gordon (see especially C. Gordon, “North Israelite Influence on Post-Exilic Hebrew,” IEJ 5 [1955], 85–88). 26  Young and Rezetko, Linguistic Dating of Biblical Texts (see n. 6), 2:99. 22 23

302

Shimon Gesundheit

allows for late influences. Unlike some other approaches, this thesis is perhaps insufficiently falsifiable.27

***** I turn now to more general issues. Revision – both literary and textual – is a major concern for both camps, as several scholars have noted.28 Textual adaptation can pollute the well by making early works look late, and conversely, it can make late features seem early. Similarly, dialect, register, and bilingualism, to name but a few examples, seriously complicate matters for any scholar, irrespective of school of thought. To conclude, we have seen that the field of historical dating is exceptionally thorny and utterly lacking in consensus.29 There are numerous methodological pitfalls, and there are problems surrounding various conclusions as well. We can only hope that increasing dialogue between scholars worldwide, as exemplified in the research group whose first fruits we are now seeing in this volume, will pave the way toward some clarity in this most important field of research.

27  For a detailed examination of the arguments of Young et al., see B. E. Dresher, “Methodological Issues in the Dating of Linguistic Forms: Considerations from the Perspective of Contemporary Linguistic Theory,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Winona Lake, IN: Eisenbrauns, 2012), 19–38. 28  E.g., F. I. Andersen and A. D. Forbes, Spelling in the Hebrew Bible: Dahood Memorial Lecture (BibOr 41; Rome: Biblical Institute, 1986), 312; Y oung and R ezetko , Linguistic Dating of Biblical Texts (see n. 6), 1:347; T. L. Thompson, “The Intellectual Matrix of Early Biblical Narrative: Inclusive Monotheism in Persian Period Palestine,” in The Triumph of Elohim: From Yahwism to Judaism (ed. D. V. Edelman; CBET 13; Kampen: Kok Pharos, 1995), 110. Pace A. Hurvitz, “The Relevance of Biblical Hebrew Linguistics for the Historical Study of Ancient Israel,” in Proceedings of the Twelfth World Congress of Jewish Studies, Jerusalem, July 29–August 5, 1997, Division A: The Bible and Its World (ed. Ron Margolin; Jerusalem: World Union of Jewish Studies, 1999), 31*. 29  Some, indeed, have given up all hope, as expressed most starkly by M. Ehrensvärd , “Why Biblical Texts Cannot Be Dated Linguistically,” HS (2006), 177–189.

The Linguistic Dating of Biblical Texts An Approach with Methodological Limitations Erhard Blum This paper will defend the following thesis concerning the methodological status of “linguistic dating” in historical exegesis of the Hebrew Bible: Linguistic analyses are of fundamental significance for biblical exegesis. Inter alia they provide an indispensable tool for the dating of biblical texts. At the same time, it is important to be aware of the limitations of purely linguistic dating: it does not surpass other methodological approaches in terms of “objectivity”; moreover, in most cases linguistic criteria alone would not allow one to determine that a controversial text originated in the postexilic period or, vice versa, in preexilic times – let alone require a more specific dating. Only a comprehensive analysis that includes all data available to historical philology has the potential to provide solid results. This position will be generally outlined in the first part of this article (I). In the second part (II), some examples will be discussed.

I 1 The Task of Dating Biblical Texts: Preliminary Considerations Dating biblical texts remains an essential and indispensable task for biblical studies, if only for the reason that there can be no comprehensive understanding of these texts without sufficient knowledge of the authorsʼ and recipientsʼ worlds. The quest for the historical context of ancient texts does not arise from some historicist obsession with explaining everything as a link in a process of historical development. Instead, this quest is imperative first of all if one is endeavoring to understand an ancient text in the way it was (or at least could be) understood by its primary author and/or addressees.1 Since biblical literature (including its  I refer here to a “principle” more metaphorically expressed by H. Gese: “Es muß als hermeneutische Grundregel gelten, daß der Text so zu verstehen ist, wie er verstanden sein will, d.h. wie er sich selbst versteht”; see H. Gese, “Der auszulegende Text,” ThQ 167 (1987), 1

304

Erhard Blum

poetic prose and poems) did not know the modern notion of fictional literature2 but generally served communicative purposes (Mitteilungsliteratur), one must reckon with the possibility or even probability that essential aspects of a textʼs meaning (in terms of semantics and pragmatics) will be missed without a more or less specific knowledge of the world and the context of the addressees for whom the author or tradent articulated his text. To be sure, there are biblical texts that treat timeless questions, but in the main their messages were addressed to people in particular spaces and times with particular identities, certainties, doubts, fear, and hopes.3 After all, it is most likely that such pragmatic aspects would in some way or other leave their mark in the individual text, in its profile, or in its rhetoric. Moreover, it seems plausible that a close analytical reading in modern times would be able to perceive those marks and to connect them with the known fractures of a past world. This endeavor should be undertaken by pursuing every reasonable methodological approach and without a priori restrictions. It may be possible to date a given biblical text (whether as an originally independent unit or as part of a diachronically defined layer) if some of the following data are available: (a) Textual features that significantly point to or exclude specific historical contexts. Such features are to be expected on the level of language, genre, and content. Focusing on the realm of semantics, one might look for clear reflections of particular historical situations in texts. Such references do exist4 but are rather exceptional. More often, texts allow dating to a rather wide period by means of an obvious terminus a quo or ad quem based on the narrative plot.5 The etiological elements widespread in traditional narratives imply basic termini a quo/ad quem as well (e.g., the hieros logos of the main sanctuary of Israel in Gen 28:11–22).6 Less obvious but nevertheless important is dating based on specific traditions in terms of conceptions, ideas, or theologumena.7 252–265, here 254, reprinted as idem , Alttestamentliche Studien (Tübingen: Mohr Siebeck, 1991), 266–282, here 269. 2  See in more detail E. Blum, “Historiography or Poetry? The Nature of the Hebrew Bible Prose Tradition,” in Memory in the Bible and Antiquity: The Fifth Durham-Tübingen Research Symposium, Durham, September 2004 (ed. S. C. Barton et al.; WUNT 212; Tübingen: Mohr Siebeck, 2007), 25–45. 3  There is, of course, a wide range of quasi-universal sapiential sentences on the one hand and traditions in which the topics are related to specific matters (as in the case of etiological material) or to the address of a peculiar situation (as in some prophetic texts) on the other hand. 4  Like Isa 1:7–9 and Sennacheribʼs campaign (701 BCE). 5  For instance, the change of King Jehoiachinʼs status by Amel-Marduk functions of course as the terminus post quem for the literary work that ended with 2 Kgs 25:27–30. Gen 22:14b presupposes the temple mount in Jerusalem. Jerusalemʼs devastation and an enduring exile of Judeans in Babylonia are clearly presupposed in the prophecies beginning with Isa 40. 6  Unless the formula “until this day” is being used as a deliberate means for suggesting a fictitious narratorʼs time (as for instance in Judg 1:21). 7  In German-speaking exegesis, the history of conceptions, ideas, etc. is called Tradi­tions­

The Linguistic Dating of Biblical Texts

305

(b) Compelling evidence showing that the text as a whole or in its very substance was addressing its audience by referring to or having an impact on specific constellations existing in the audienceʼs own historical context. This evidence pertains to the pragmatics of an individual unit, of a complex composition, or of a collection of individual texts that determined the overall profile of the text in question (examples will be given in part II). (c) Significant stratigraphical and/or typological relations to other reliably dated textual units/components. Stratigraphy refers in this context to textual interrelations that might show that the textual elements in question belonged to the same (primary) composition or that they represented certain diachronic intratextual dependencies or a demonstrable intertextual relationship/dependence between independent units. Typology refers to a possible concurrence of textual features concerning language, style, and conception that points to an affiliation of texts in terms of schools of tradition (Dtr, Priestly, etc.), of specific spatiotemporal, social, or diachronic settings (for instance, Israelite vs. Judean; royal court vs. temple; pre- vs. postexilic; land vs. diaspora). The soundest results are to be expected if several data out of constellations a–c coincide in one case or text. This rule applies also to linguistic criteria for dating texts in the Hebrew Bible (as part of the data according to category a or c). Moreover, for methodological reasons, linguistic arguments should be part of a much more comprehensive historical-philological endeavor. 2 Basic Limitations of Linguistic Dating Naturally, scholars are inclined to claim methodological superiority for their own disciplines. In this vein, one might read in language studies concerning the Hebrew Bible that exegetical arguments are based on interpretations and conjectures and therefore are merely subjective in comparison with purely linguistic data and judgments. At the same time, it is not unusual to see other scholars focusing exclusively on their own exegetical analyses, disregarding established insights into the history of Biblical Hebrew. I fear that all such approaches, relying exclusively on a preferred profession, will prove shortsighted. 2.1 Systemic Limitations There is hardly any “purely linguistic judgment” regarding the dating of ancient Hebrew texts, if only because diachronic linguistic distinctions have to be based on the analysis of sufficient comparative corpora. Defining those corpora exclusively on linguistic grounds, however, would result in a vicious circle. It seems convincing, therefore, to rely on a Late Biblical Hebrew (LBH) corpus consisting primarily of books whose postexilic dating is undisputed simply on grounds ge­schichte, whereas the genesis of texts in transmission processes is called Überlie­fe­rungs­ ge­schichte.

306

Erhard Blum

of their narrative plot, such as Esther, Ezra–Nehemiah, and Daniel. Chronicles is grouped here as well, and rightly so. In this case, however, the narrative plot alone would not compel a much later dating then DtrH or Deutero-Isaiah. The general consensus concerning Chronicles is instead based on differentiated literary-historical analysis, the linguistic argument being just part of it. Arguing for a core corpus of Classical Biblical Hebrew (CBH) exclusively on linguistic grounds would be even more difficult, if not impossible, since we do not have sufficient “external” (i.e., epigraphic) data for Hebrew prior to the Hellenistic period. Indeed, in Persian times such sources are lacking almost completely.8 Thus, if we are to reckon with linguistic data only, why not locate CBH and LBH texts, for example, on a time scale extending from the sixth century BCE until the late Persian period? There would of course be many arguments against such a proposal, but they would have to be rooted in general historical and exegetical assumptions. In short, linguistic reasoning cannot escape the hermeneutical circle that characterizes every interpretive effort. The critical impact of this insight for the whole discussion becomes even clearer when we consider the rather generalized dating of books or canonical units preferred in linguistic studies. This contrasts with much more diversified debates on individual texts in the realm of historical exegesis. The supposedly high diversity of authors and tradents, of literary genres (beyond the dichotomy of poetry and prose), and of levels of literary caliber calls into question not only many statistics concerning the frequency of lexemes or (morpho)syntactical structures in “books” (quite apart from units like “the Pentateuch” or “the Former Prophets”9) enjoying much popularity in recent debate10 but also more 8  Except for a few stamps, bullae, and coins with personal names (and some titles). Significantly, the only major exception so far seems to be the famous silver amulets from Ketef Hinnom (based on their orthography). 9  Not to mention DtrH being wrongly equated with Former Prophets, as in D.-H. K im , Early Biblical Hebrew, Late Biblical Hebrew, and Linguistic Variability: A Sociolinguistic Evaluation of the Linguistic Dating of Biblical Texts (VTSup 156; Leiden: Brill, 2013), 70 and passim. At the same time, Kimʼs drawing a distinction between three corpora in Isa (103) is a (first) step in the right direction. 10  Such statistics measuring the absolute quantity or relative frequency (per signs, etc.) of linguistic features in books can be found in investigations by critics challenging the linguistic approach (e.g., Young, Rezetko [see below, n. 13]) and by its defenders (e.g., Rendsburg [see below, nn. 36, 46], Hornkohl [see below, n. 24]). Although language usage depends very often on Gattungsstil, authorial style, Tendenz, content, on the intellectual milieu, etc., such parameters are disregarded in referring to formally/traditionally defined biblical corpora. A simple example concerning the usage of the pronouns ‫ אנכי‬vs. ‫ אני‬should suffice in this context. Checking the occurrences in the Pentateuch as a whole would not reveal that Priestly texts (in the broadest sense) show an almost total preference for ‫( אני‬120 times) over ‫( אנכי‬once, Gen 23:4). The evidence is more impressive than in recognized LBH books (for instance in Chronicles, with ‫ אני‬and ‫ אנכי‬in a ratio of 30:1). Accordingly, this seems to be strong evidence for a postexilic date for P, which would perfectly fit my own understanding of P. There is,

The Linguistic Dating of Biblical Texts

307

subtle comparative approaches. In the present context, I will confine myself to the latter, discussing a few examples based on classical studies by A. Hurvitz. His extensive investigation of the language of Ezekiel, in particular, represents a major contribution to the debate on the date of P. Through comparison with the language of P and a supposed preexilic (CBH) corpus, he attempts to show that Ezekiel represents a “transitional” state of language which is later than preexilic P and earlier than LBH.11 But is it self-evident that the book of Ezekiel represents (early) exilic Hebrew in toto? What would be the consequences if major parts of the book are pseudepigraphic?12 Moreover, Hurvitzʼs comparative corpus representing CBH comprises inter alia Deuteronomy and the Former Prophets, from which undisputed Deuteronomistic examples are quoted quite often, among them such recognized late-Deuteronomistic pericopes as Deut 4; 30; Josh 13; and 1 Kgs 8:46–53,13 not to mention however, evidence for ‫ אנכי‬in post-P compositions as well (see below, section II). Therefore, the exclusive usage in P might be explained not by diachrony alone but by a kind of Schulstil that was maintained in P circles throughout the generations. Interestingly, A. H urvitz , A Linguistic Study of the Relationship between the Priestly Source and the Book of Ezekiel: A New Approach to an Old Problem (CRB 20; Paris: Gabalda, 1982), 169, n. 35, falls back on nonlinguistic dating in this case: “the use of this word [scil., ‫ ]אני‬becomes extremely widespread in later texts, and its employment is clearly characteristic of LBH. [. . .] However, ‫ אני‬is dominant in H as well, and the antiquity of H is acknowledged by [. . .] late-daters of P.” But the last argument hardly fits current research (at least since K. Elliger and I. Knohl) and the view “that the very same element may in certain texts reflect LBH usage, while in others it represents simply a stylistic peculiarity that does not necessarily have any distinctive chronological implications” (Hurvitz, Linguistic Study, 169, n. 35) might conversely be true of data put forward in favor of early dating for P and other biblical traditions as well. 11  Hurvitz, Linguistic Study (see n. 10). 12  True, Hurvitz accepts “major subgroups” with respect to P (“H”) and Ezek (“Deutero Ezekiel” [= Ezek 40–48]) in chapter 1 of his study, declaring that “their implication should be examined only after the linguistic analysis of the extant texts has been completed” (Hurvitz, Linguistic Study [see n. 10], 20–21). As far as I can see, however, even those rather rough distinctions remain without any crosscheck in his investigation. 13  Cf. Hurvitz, Linguistic Study (see n. 10), item 1 (26, n. 11), item 18, and items 31–33. The example of item 32 (123–125) deserves special attention. Hurvitz argues there that the verb ‫ כנס‬used in Ezek 39:28 is typical of LBH, following famous forerunners like F. Delitzsch, S. R. Driver, and others (124, n. 198). Although Hurvitz relies mainly on late- or postexilic texts, like Deut 30:3–4; Isa 11:12; and Mic 2:12, as alleged evidence for the usage of ‫קבץ‬ or ‫ אסף‬as CBH equivalents (123, 125), the theory nevertheless seemed to be confirmed not only by the LBH corpus and by the evidence of Rabbinic Hebrew and Aramaic but also by Ezek 39:28, which is regarded as a relatively late redactional Fortschreibung with reason. However, an ostracon (seventh century BCE) found some years ago in the City of David twice lists inter alia the designation )]‫“( הכנס (כס[ף‬the collector of cash”); cf. J. N aveh , “Hebrew and Aramaic Inscriptions,” in Excavations at the City of David, 1978–1985, Vol. VI: Inscriptions (ed. T. D. Ariel; Qedem 41; Jerusalem: Institute of Archaeology, The Hebrew University, 2000), 2–4; S. Aḥituv, HaKetav VeHamiktav: Handbook of Ancient Inscriptions from the Land of Israel and the Kingdoms beyond the Jordan from the Period of the First

308

Erhard Blum

later post-DtrH insertions like Josh 22:9–3414 or Judg 17–21. Such a proceeding, however, has implications not only for the persuasiveness of particular arguments. If one assumes – for the sake of argument – the prevalent dating (exilic/early postexilic), the comprehensive editions and redactions in Deuteronomy, “H,” the Former Prophets, and Jeremiah would be more or less contemporaneous with the book of Ezekiel. Thus, assigning (like Hurvitz) the books/traditions first mentioned to CBH and the book of Ezekiel to a transitional linguistic status between CBH and LBH would require drawing at least two conclusions of major importance: (1) There is no clear correlation between linguistic typology and historical chronology regarding texts of the Hebrew Bible. (2) The diachronic relation between texts that are different in regard to linguistic typology cannot be determined by linguistics alone. In terms of analytical procedure, the conclusion is even more fundamental: it is not possible to assign a biblical text to CBH just because it belongs to the Former Prophets or to the Pentateuch, nor can such a procedure be justified by claiming that one should “avoid” basing linguistic analyses on exegetical hypotheses a priori.15 There is simply no “neutral/objective position” in such issues, because refraining from diachronic analysis would itself inevitably involve a diachronic assessment. Moreover, relying on traditional dating cannot replace historical-critical judgments made by reasons; otherwise, one would have to give up Deutero-Isaiah and to date every “Psalm of David” to the tenth century. In sum: all linguistic dating of Hebrew Bible texts is inevitably based on premises that imply the acceptance or denial of literary-historical judgments.

Commonwealth (Biblical Encyclopaedia Library 21; Jerusalem: Mosad Bialik, 2005), 41–43. It might be for this reason that the cited verb is not included in A. Hurvitz, A Concise Lexicon of Late Biblical Hebrew: Linguistic Innovations in the Writings of the Second Temple Period (VTSup 160; Leiden: Brill, 2014). See also the examples listed in I. Young and R. Rezetko, Linguistic Dating of Biblical Texts I: An Introduction to Approaches and Problems (London: Equinox, 2008), 163–168, with alleged LBH features in the Arad Ostraca. (Whereas the instances of weqatal can be explained according to the standard verbal system, at least items 6.5.2, 6.5.6–10 seem to be significant.) At any rate, the examples should be taken as a warning concerning the reliability of linguistic judgments, even in cases with apparently full and decisive evidence. Thus they call into question “that it is possible to use linguistic alternatives in comparable contexts, thereby removing any suspicion [sic] that a featureʼs absence from CBH stems from accidental distribution or from the limited scope of the Hebrew Bible” (Hurvitz, Concise Lexicon, 10). 14  This story plays a role as alleged evidence for a CBH use of ‫( תבנית‬cf. Hurvitz, Linguistic Study [see n. 10], 84, n. 97) and of ‫( עדה‬in the sense used in P). R. Goldstein, however, has convincingly substantiated the dating of Josh 22:9–34 in the Persian era; see R. Goldstein, “Joshua 22:9–34: A Priestly Narrative from the Second Temple Period,” Shnaton 13 (2002), 43–81 (Hebrew). 15  Cf. Hurvitz, Linguistic Study (see n. 10), 20.

The Linguistic Dating of Biblical Texts

309

2.2 Limitations of Identifying Late Texts on Linguistic Grounds The interdependence of linguistic and exegetical analyses just outlined might be mitigated to some degree if linguistic dating focused exclusively upon identifying late (postexilic) compositions, thereby relying on a core corpus of LBH texts in the Hebrew Bible that is almost unquestioned (see above) and on a clearly defined set of criteria for identifying LBH elements.16 Nevertheless, even in this instance, the claim of purely linguistic investigation tends to fall victim to positivistic illusion(s). Thus linguistic arguments are largely premised on the assumption that the biblical texts have preserved the Textgestalt they received when they were composed or redacted. At the same time, we know, of course, that the received texts went through hundreds of years of scribal transmission and that there are plenty of examples of linguistic changes and adjustments by later scribes, as evidenced by “empirical” manuscripts or by double transmission of the same text.17 Such scribal “adaptations” can involve lexemes or grammatical phenomena. An example of the latter would be the variant forms of the third-person masculine plural pronominal suffix after the feminine plural nominal ending ‫־ות‬, i.e., ‫־ם‬ versus ‫ אבותם( ־יהם‬vs. ‫)אבותיהם‬. A. Hurvitz persuasively points to the following examples in parallel traditions or textual witnesses:18 2 Samuel 22:46 Foreigners [. . .] came trembling out of their fastnesses (‫)ממסגרותם‬ Psalm 18:46 Foreigners [. . .] came trembling out of their fastnesses (‫)ממסגרותיהם‬ 1 Kings 8:34 the land which Thou gavest to their fathers (‫)לאבותם‬ 2 Chronicles 6:25 the land which Thou gavest to them and to their fathers (‫)לאבתיהם‬ 16  I am referring, of course, to the widely accepted set of four criteria established by A. Hurvitz in several publications: (1) “Biblical distribution” (the element being exclusively or predominantly attested in the core corpus of LBH), (2) “Extra-Biblical sources” (the element being current in external postexilic sources like Aramaic and/or Mishnaic Hebrew), (3) “linguistic opposition” (the element in question having a CBH equivalent), and (4) “accumulation” (“a heavy concentration of late linguistic elements” in a given text); see, for instance, A. Hurvitz, “Can Biblical Texts Be Dated Linguistically? Chronological Perspectives in the Historical Study of the Old Testament,” in Congress Volume: Oslo, 1998 (ed. A. Lemaire and M. Sæbø; VTSup 80; Leiden: Brill, 2000), 143–160, here 148–149, 153. 17  See I. Young, “Biblical Texts Cannot Be Dated Linguistically,” HS 46 (2005), 341–351, here 349–351 (with bibliography) and especially the excellent discussion in D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011), 125–132. 18  Hurvitz, Linguistic Study (see n. 10), 26. For a convenient presentation of the debate about the evidence and Hurvitzʼs position, see Kim, Early Biblical Hebrew (see n. 9), 99–107.

310

Erhard Blum

Isaiah 59:8 MT: there is no justice on their paths (‫)במעגלותם‬ 1QIsaa: ‫במעגלותיהמה‬ Exodus 36:34 MT: And he made their rings (‫ )טבעתם‬of gold Sam.: ‫טבעתיהם‬

In view of such evidence, however, one might ask how we can be sure that the probably late19 form ‫מושב(ו)תיהם‬, which occurs twice in Ezek (6:14; 37:23),20 did not come from the pen of some copyist(s) as well. The option of scribal phenomena appears almost certain in a case like Ezek 18:9, the only instance of late language with ‫ הלך‬in the piel (‫)יהלך‬, in contrast to fifteen (!) incidences of the “classical” ‫ הלך‬qal in semantically corresponding contexts of Ezekiel. The text-critical evidence points to a secondary reading caused by dittography (of the yod).21 Similar (text-critical) considerations are significant with regard to isolated phenomena in Ezek 18:7 and 18:12.22 These few examples should suffice to indicate a difficult constellation. On the one hand, any text-critical judgment appears to be problematic without support from textual witnesses. On the other hand, testimony of textual variants in the received manuscripts remains more or less accidental. Nevertheless, the question remains serious if certain phenomena do reveal the linguistic profile of a specific literary unit or, rather, the textual nature of the received manuscript tradition – the great Isaiah scroll (1QIsaa) probably being the most telling evidence for what might be conceivable in this respect.23 Unfortunately, the textual testimony is not the only matter of complexity regarding the evidence and the unpredictable absence of evidence that one faces in this context and that cannot be solved systematically. Two further lines of reasoning should at least be mentioned, for they are quite widespread in the exegetical realm. First is the argument of “Aramaisms,” which almost automatically results 19  Although the evidence is not clear regarding LBH (Qumran) and MH, see E. Qimron, The Hebrew of the Dead Sea Scrolls (HSS 29; Atlanta: Scholars Press, 1986), 63, with n. 81. 20  Cf. Hurvitz, Linguistic Study (see n. 10), 24 (item 1). 21  Cf. the main arguments in W. Z immerli , Ezechiel I (2nd ed.; BKAT 13/1; Neukirchen-Vluyn: Neukirchener Verlag, 1979), 383, n. to v. 9b. Hurvitz, Linguistic Study (see n. 10), 49, n. 72, refers inter alia to Zimmerli. Nevertheless, he argues for counting the MT reading in 18:9 as a late feature of Ezek (item 9) by defending the possibility of an imperfect form besides the perfect in v. 9 (which is not the main issue) and by passing over the text-critical observations. 22  Cf. Hurvitz, Linguistic Study (see n. 10), 128–131 (item 34), 43–46 (item 7), and the critical commentaries ad loc. 23  As should be well-known, the groundbreaking investigation of E. Y. K utscher , The Language and Linguistic Background of the Isaiah Scroll (Jerusalem: Magnes Press, 1959) (Hebrew), has shown that the scroll does witness many textual changes in terms of orthography, morphology, lexicon, etc., which strongly reflect influences of LBH, MH, Aramaic, et al.

The Linguistic Dating of Biblical Texts

311

in a postexilic dating of the text in question, though the need for a differentiated approach was emphasized long ago (Kautzsch, Hurvitz24), and rightly so. For this context, it should suffice to note the fact that Aramaic and Hebrew stood in close contact at least from the ninth century BCE onward, with Aramaic being for the most part the dominant language. Thus Aramaic loanwords and calques would be expected in preexilic Hebrew, especially in the northern Israelite dialect. Judging from the evidence for “Canaanisms” in Damascene Aramaic at the end of the ninth century BCE, the influence could even be quite substantial.25 The other problematic line of argumentation is the use of statistics. The discussion above already hinted at some problems with arguing on the basis of frequency counts.26 Beyond Hebraistic research proper it has become quite popular to “prove” the “lateness” of a specific lexeme through quoting a handful (or less) of instances of the given lexeme in supposedly late biblical books. Judging in terms of statistics, such a procedure is insignificant. It can be taken seriously only if additional linguistic data of importance is brought to bear on the case. But even in such instances, the implied argumentum e silentio might prove to be deceptive.27 In general, such deliberations show how important it is that “statistical” data in the realm of Biblical Hebrew be screened by informed statisticians.

Since the outlined possible fallacies of identifying “late” linguistic features cannot be completely controlled, the overall data have to be weighted carefully in judgments of higher or lower probability. Taken together, these deliberations strengthen the doubt whether linguistic data alone can prove the lateness of a specific biblical text decisively. 2.3 The Impossibility of Identifying Preexilic Texts on Linguistic Grounds Alone The doubt, however, that linguistic analyses can prove that a biblical text was composed in preexilic times is even stronger. Naturally, such a proof would be based on the absence of LBH features in the respective text or corpus. But the “absence of evidence” could be decisive only if a late author would unavoidably have produced sufficient LBH features, even if he tried to write in classical 24  E. Kautzsch, Die Aramaismen im Alten Testament (Halle a.S.: Max Niemeyer, 1902); A. H urvitz , “The Chronological Significance of ‘Aramaisms’ in Biblical Hebrew,” IEJ 18 (1968), 234–240; recently also A. H ornkohl , Ancient Hebrew Periodization and the Language of the Book of Jeremiah: The Case for a Sixth-Century Date of Composition (Studies in Semitic Languages and Linguistics 74; Leiden: Brill, 2014), 12. 25  As is well-known, the Damascene inscriptions from Tel Dan and Tell Deir ʿAlla provide excellent evidence of substantial “South-Canaanisms” concerning lexicon (for example ‫לחם‬ in the t-stem [“to fight”], ‫ראה‬, ‫משא‬, ‫“[ לכו‬go!”], etc.) and grammar (the verbal-tense system, etc.). Concerning the identification of the language, see E. Blum, “Die aramäischen Wand­ inschriften von Tell Deir ʿAlla,” in Weisheitstexte, Mythen und Epen (ed. B. Janowski and D. Schwemer; TUAT 8; Gütersloh: Gütersloher Verlagshaus, 2015), 459–474, here 459–460, and in more detail idem , “A ‘Neglected’ Aramaic Idiom: The Language of the Kingdom of Aram-Damascus,” forthcoming (Hebrew). 26  See above, n. 10. 27  Cf. above, n. 13.

312

Erhard Blum

language. In fact, that is what proponents of this approach in the field of Hebrew linguistics do assert. The problem of such an allegation, however, lies in its a priori nature, for no one will ever be able to prove or to disprove such a general statement. Though repeated “slip-ups” in comparison to the classical language were possible,28 they were by no means necessary. Nevertheless, it seems useful to discuss conceivable options regarding the competence and the intentions of later scribes to write “classical” literary Hebrew. Three possible factors should be considered: (a) the textʼs (implied) claim regarding the time of its origin, (b) the social status and institutional context of writing Hebrew in postexilic times, and (c) the time factor. Factor a, which deserves special attention, concerns the difference between avowedly “contemporary” texts and allegedly “ancient” texts in postexilic contexts. Prominent examples for the first group are prophetic texts like Deutero-Isaiah,29 Haggai, and Zechariah.30 The most important example for the second type is of course the epitome of Hebrew pseudepigraphy, the Torah of Moses, which started with the Deuteronomic Torah and reached a peak in the Priestly composition. Plausibly, an author of pseudepigraphic literature with high reputation would have had an interest in carefully writing according to classical standards. In this respect, one would expect a significant difference between, letʼs say, Ezra–Nehemiah or Esther and pentateuchal texts. The prospects of success in maintaining traditional language standards were increased the more a given scribal tradition was anchored in a school context that trained students through memorization and

28  Occasional deviations would not suffice according to the accepted criteria for identifying LBH texts. 29  S. M. Paul, “Signs of Late Biblical Hebrew in Isaiah 40–66,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; LSAWS 8; Winona Lake, IN: Eisenbrauns, 2012), 293–299, presents quite a lot of material, especially concerning the lexicon. He reaches the following conclusion: “Though these prophecies are still well anchored in Classical Hebrew, the beginnings of what would become common in Late Biblical Hebrew and even later forms of Hebrew have left their imprint in the text.” (294) It should be noted, however, that a major part of his material comes from texts in Isa 56–66 whose dating is a controversial matter. 30  The book of Esther represents another interesting instance. According to J. Naveh and J. C. Greenfield, “[t]he author of Esther displays great narrative skill and stylistic ability. He was a master of Classical Hebrew style and used such typical elements as the ‘waw-conversive’ [. . .] with finesse. However, even though he carefully writes hammelek ˀăḥašvērōš [. . .] rather than ˀaḥašvērōš [sic] hammelek [. . .], the later Aramaizing form, he cannot avoid using the late malkūt [. . .] rather than mamlākā for kingdom. In addition, the numerous loanwords from Aramaic [. . .] and from Persian [. . .], the many neologisms and the use of new nominal patterns also reveal the date of the composition”; J. Naveh and J. C. Greenfield, “Hebrew and Aramaic in the Persian Period,” in The Persian Period, vol. 1, The Cambridge History of Judaism (ed. W. D. Davies and L. Finkelstein; Cambridge: Cambridge University Press, 1984), 115–129, here 121. But why should a narrative about the Persian court and empire “avoid” loanwords from Aramaic and Persian, etc.? The author had nothing to hide in this respect.

The Linguistic Dating of Biblical Texts

313

reproduction of classical literature31 (factor b). The orientation toward ancient Hebrew traditions probably32 did gain additional importance at the time when Judean/Jewish writing was split into two separate sectors: (1) writing for everyday communication and administration in dominant Aramaic on the one hand and (2) writing a literary Hebrew idiom that existed almost exclusively in sacral and “ethnic” literature, that is, mostly in the realm of religion and cult or in sectoral school contexts, on the other.33 Nevertheless, it is highly plausible that written Hebrew in the Persian period gradually developed its own inner dynamics and lost close continuity with the classical language. Some later books representing “LBH” and the postbiblical idioms confirm this assumption (factor c).34 3 Issues Concerning “Postexilic Hebrew” The last considerations have implicitly led to the question how the change from Standard/Classical Biblical Hebrew to Late Biblical Hebrew might be dated, assuming that the presumably gradual process does allow such a determination. In any case, it would make a considerable difference whether such a change is placed “shortly after the beginning of the Babylonian exile” (J. Joosten) or instead toward the end of the sixth century BCE (A. Hurvitz, R. Polzin). The first position seems to me difficult to maintain. I do not see any historical probability that the language of the Judeans should have undergone fundamental changes one or two generations after the fall of Jerusalem. Moreover, why should one presume that Judean scribes, be they priests or laymen, would (unconsciously?) change their own Literatursprache shortly after the fall of Jerusalem? What would be the indications of this in Deuteronomistic prose35 or even in the prophecies of Haggai36 and Proto-Zechariah?37 31  See especially D. M. Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (New York: Oxford University Press, 2005). 32  Regarding the unbroken continuity of literary Hebrew after 587 BCE, it seems rather inappropriate to speak about an “archaizing style” of postexilic scribes referring to authors, let us say, of the sixth century BCE. 33  Nota bene, again, that epigraphic material in Hebrew ceases abruptly after the capture of Jerusalem by the Babylonians in the sixth century BCE; see above, n. 8. 34  In theory one might not exclude the possibility that a learned and talented author, even in Hellenistic times, succeeded in writing a text in more or less perfect classical Hebrew. In fact, however, this possibility seems improbable. 35  In terms of method, such assumptions would bring back some aporiai already discussed above in section 2.1. 36  G. A. R endsburg , “Late Biblical Hebrew in the Book of Haggai,” in Language and Nature: Papers Presented to John Huehnergard on the Occasion of His 60th Birthday (ed. R. Hasselbach and N. Pat-El; SAOC 67; Chicago, IL: The Oriental Institute, 2012), 329–344, lists twenty features; for several reasons only four examples seem to me more or less reliable (items 1, 4, 11, and 12). 37  Cf. M. E hrensvärd , “Why Biblical Texts Cannot be Dated Linguistically?,” HS 47

314

Erhard Blum

The second position, which reckons with the persistence of CBH into the sixth century BCE, comes very close to an earlier understanding summarized by S. R. Driver. A much-quoted assertion in his introduction to the Hebrew Bible says: “The change [sc. in Hebrew style] is beginning (ca. 450) in the memoirs of Nehemiah and in Malachi; but Aramaisms and other marks of lateness (esp. in syntax) are abundant only in works written after this date––Esther, Chr., Eccl. &c.”38 Given such a perspective, the whole debate about P and its relation to LBH appears in a different light.39 According to traditional dating in German scholarship, namely, Wellhausen, Noth, Elliger, Zimmerli, and others, the Grundschrift of P was exilic. Moreover, according to Wellhausen many of the cultic laws and rituals of P were nothing other than old professional traditions written down after the great fall. In this line of reasoning, P would be expected to be more or less CBH. In my own view, the basic P composition was probably initially formed around 500 BCE. But this basic composition integrated earlier Priestly material and was probably composed over an extended period of time. Moreover, it would be misleading in my judgment to think in terms of one author. We have to reckon with a group of tradents. That means that the P composition represents one work but was created by several “hands.” These authors were apparently erudite priests and professional, well-trained scribes who knew traditional Hebrew literature by heart, and they no doubt intended to give an impression of “Mosaic epigraphy.” Accordingly, we would expect a rather conservative idiom in P. It surely would not have abundant neologisms. At best it would include some moderate adjustments to the actual Literatursprache. And “hey, presto,” this is what we more or less have in the language of “P” according to S. R. Driver40 or R. Polzin.41 (2006), 177–189; Y oung and R ezetko , Linguistic Dating (see n. 13), 136, with n. 84. See also the next note. 38  S. R. Driver, An Introduction to the Literature of the Old Testament (9th ed.; International Theological Library; Edinburgh: T&T Clark, 1913), 156, and see 504–505, where he states even more definitely, “it must be borne in mind that the great turning-point in Hebrew style falls in the age of Nehemiah.” He continues in a note: “And not, as is sometimes supposed, the Captivity. This appears with especial clearness from Zech. the style of which [. . .] is singularly pure” (505, n. *). 39  This would apply also to a position that proposes to replace the binary CBH-LBH opposition by a tripartite structure: CBH – Transitional Biblical Hebrew (sixth century) – LBH; see, e.g., Hornkohl, Ancient Hebrew Periodization (see n. 24). “Die Gleichzeitigkeit des Ungleichzeitigen” is widely evidenced in synchronic research on empirical languages. Deliberate preference of classical language standards beside texts with “transitional” features would therefore be quite conceivable. 40  Driver, Introduction (see n. 38), 155–157. 41  R. P olzin , Late Biblical Hebrew: Toward an Historical Typology of Biblical Hebrew Prose (HSM 12; Missoula, MT: Scholars Press, 1976), 85–122. Significantly, prominent LBH features like prospective suffixes or singular collective nouns with plural verbal forms prevail in later P pieces (38–42, 98, 103).

The Linguistic Dating of Biblical Texts

315

II In the following, we shall deal with some interconnected texts in Genesis in order to indicate what is meant by a “comprehensive approach” that includes linguistic data as part of exegetical dating. Let us begin with the fine analysis of Gen 24 by A. Rofé which was published first in Hebrew in 1976 and then in an English version in 1990 in the Festschrift in honor of the late Rolf Rendtorff.42 According to Rofé, “Genesis 24 is a very late composition, written in the fifth century BCE, about the time of Ezra and Nehemiah. The dating can be borne out by examining the story from various aspects: its language, legal institutions, religious concepts, literary connections, literary character, and message.”43 There is no need to enlarge here upon Roféʼs very substantial argumentation. Some crucial aspects, however, must be highlighted. Concerning the language, he points inter alia to the divine epithet )‫( אלהי השמים (ואלהי הארץ‬vv. 3, 7), which occurs otherwise seven times in Jonah, Ezra, Nehemiah, and Chronicles, and its Aramaic equivalent, which appears twelve times in Daniel and Ezra and in Elephantine documents;44 and to hapax legomena with cognates in Rabbinic Hebrew, like ‫“( הגמיא‬give drink”) in v. 17 and ‫“( משתאה‬watch,” “wait”) in v. 21 and ‫( אם לא‬v. 38) with the meaning of rabbinic ‫“( ֶא ָּלא‬but”) instead of the classical ‫כי אם‬. A feature of late syntax is the use of ‫ אשר‬instead of ‫ כי‬introducing an object sentence (v. 3).45 G. A. Rendsburg has discussed the linguistic arguments for Roféʼs dating of Gen 24 in detail.46 On the one hand, he recognizes almost all of the late linguistic features that Rofé identified. On the other hand, he attempts to explain them as deliberate stylistic devices: “the Aramean setting [scil. of the plot] serves as the catalyst for the inclusion of Aramaisms in the narrative.”47 Thus he denies the possibility of using those features for dating the story to a late period. In his opinion, Gen 24 belongs rather “to the classical stratum of biblical narrative,” which 42  A. Rofé, “The Betrothal of Rebekah (Genesis 24),” Eshel Beer-Sheva 1 (1976), 42–67 (Hebrew); idem, “An Enquiry into the Betrothal of Rebekah,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag (ed. E. Blum et al.; Neukirchen-Vluyn: Neukirchener Verlag, 1990), 27–39. 43  Ibid., 27. See already B. Diebner and H. Schult, “Alter und geschichtlicher Hinter­grund von Gen 24,” Dielheimer Blätter zum Alten Testament 10 (1975), 10–17, here 14: “Für die nachexilische Entstehung des vorliegenden Textes sprechen literaturgeschichtliche, sprach­ liche und historische Überlegungen. Das Kapitel läßt sich am besten auf dem Hintergrund der Heiratsbestimmungen des qhal hagola (ab Mitte des 5. Jh.s vC) verstehen.” 44  See Rofé, “Enquiry” (see n. 42), 28, nn. 7–8, and n. 9 with evidence in Jdt and Tob. 45  Rofé, “Enquiry” (see n. 42), 30. 46  G. A. R endsburg , “Some False Leads in the Identification of Late Biblical Hebrew Texts: The Cases of Genesis 24 and 1 Samuel 2:27–36,” JBL 121 (2002), 23–46. Rendsburg justifies his selective discussion as follows: “I will treat only [. . .] the language issue, for it is on this pillar, I think one may state safely, that the remaining ones rest” (24). This assessment, however, does not do justice either to Roféʼs understanding or to the profile of the story. 47  Rendsburg, “Some False Leads” (see n. 46), 24.

316

Erhard Blum

“antedates the Mesha Stele, written in the mid-ninth century B.C.E. We are dealing here with texts composed in the early monarchic period.”48 A close inspection, however, reveals that his explanation is not consistent, for several reasons. (a) The story as a whole is not “set in Aram”;49 it begins and ends in the household of Abraham. (b) Putting some Aramean color in the mouth of the Mesopotamian protagonists could be an interesting narrative device indeed, but several Aramaisms appear in formulations of the narratorʼs voice (items 5, 6, 8, 9 in Rendburgʼs presentation) and even in the mouth of Abraham speaking to the servant (items 1 and 2).50 (c) According to Rendsburg, the Aramaisms in Gen 24 would be Old Aramaic though the features in question are admittedly “attested only in later Aramaic sources (Persian period onward).”51 In principle, this could be due to the shortage of ancient Aramaic sources. But this explanation falls short in at least one case. The occurrences of the phrase ‫ אלהי השמים‬in Jonah, Ezekiel, Nehemiah, and 2 Chronicles are said to be “all under the influence of Aramaic ‫אלה‬ ‫ׁשמיא‬.”52 All attestations, however, for this Aramaic phrase (“the God of heaven”) are limited to Ezra and Daniel and to Jewish sources from Elephantine. In the Old Aramaic Zakkur Stele we find the plural formulation ‫“( אלהי שמי[ן ואלה]י ארק‬the gods of heave[n and the god]s of earth”) (KAI 202 B 25–26) instead. With regard to the singular phrase, its absence in Old Aramaic texts is not accidental at all, for assuming ‫ אלה שמיא‬in pre-Persian times (and a non-Jewish context?) would be an anachronism.53 In other words, ‫ אלה שמיא‬constitutes not only a linguistic feature but also a conceptual one that is definitively “late.”

Late conceptual elements demonstrated by Rofé include the role of the prayer in the story (with the triplet structure “prayer – Godʼs answer – thanksgiving”), which recalls similar patterns in the books of Daniel, Judith, and Tobit or the angelology in Gen 24.54 Two tenets, however, marking the central point of the narrative in terms of its pragmatics,55 constitute (in my view) the most striking  Rendsburg, “Some False Leads” (see n. 46), 35.  Rendsburg, “Some False Leads” (see n. 46), 34. 50  Rendsburg attempts to explain this by “completing” the story with information the biblical narrator failed to supply: “We know that Abraham brought servants with him from Haran (see Gen 12:5), and no doubt he obtained more servants later. [. . .] Most likely the servant that he sent to Haran was from the former group, someone who knew the environs and could speak the language” (Rendsburg, “Some False Leads” [see n. 46], 32). Furthermore, he refers to the midrashic tradition about ‫דמשק אליעזר‬, stating, “Eliezerʼs name may indicate that he hailed from Damascus, that is, Aram. Did Abraham obtain him already in Haran? Did he obtain him en route from Haran to Canaan, when he passed through Damascus? Naturally, we do not know the answers to these questions” (32, n. 45). Whose questions are those? The following note (n. 46) seems to clarify: “I am not judging the historicity of the patriarchal narratives. Rather, I simply am judging the story on its own terms, that is, the story as a literary creation” – in sum, the ancient narrator apparently did not care to fill the gaps created by Rendsburgʼs theory. 51  Rendsburg, “Some False Leads” (see n. 46), 31, n. 44. 52  Rendsburg, “Some False Leads” (see n. 46), 25. 53  In this respect there is a difference also between ‫ אלה שמין‬and ‫בעל שמין‬. 54  R ofé , “Enquiry” (see n. 42), 31–33; and see J. Van S eters , Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975), 243. 55  Working out the specific pragmatics of an individual text should not be confused with the quest for more or less general ideas as means of dating. For the latter see the somewhat polemical criticism of B. D. Sommer, “Dating Pentateuchal Texts and the Perils of Pseudo48 49

The Linguistic Dating of Biblical Texts

317

clue for the historical context of the story.56 Both tenets are explicated at the beginning, in Abrahamʼs instructions to the steward. The first one is the interdiction of marriage with “Canaanite” women and the instruction to take a woman from the kin of Abraham instead (vv. 3, 4):57 I will make you swear by the Lord, the God of heaven and earth, that you will not get a wife for my son from the daughters of the Canaanites, [. . .] but will go to my country and to my kindred and get a wife for my son Isaac.

As the servant says, this aim could be realized by bringing Isaac to the woman in Mesopotamia. Abraham, however, fiercely rejects this option (i.e., leaving the promised land). And this point constitutes the second tenet (v. 6): Abraham said to him, “See to it that you do not take my son back there.”

Since the orders could be mutually exclusive, Abraham immediately adds that God, who had sworn to give him the land, will bring the servantʼs mission to a good end (v. 7, 8). The Lord, the God of heaven, who took me from my fatherʼs house and from the land of my birth, and who spoke to me and swore to me, “To your offspring I will give this land,” he will send his angel before you, and you shall take a wife for my son from there. But if the woman is not willing to follow you, then you will be free from this oath of mine; only you must not take my son back there.

And that is what the story is all about: the importance of endogamy-based kinship and of staying in the land – the land having the priority in case of a conflict between both goals. The lengthy explication of these issues in the exposition of the story (24:3–8) suggests that the explanation for the servant is actually directed at the readers/hearers of the story. One detail is even more revealing in this respect: though the servant verbally repeats most of 24:3–8 in the presence of Rebekahʼs family (24:37–41), including the servantʼs objection that the woman might not be willing to come with him and Abrahamʼs freeing him from the oath in such a case, one item is omitted in this repetition – the option that Isaac could be brought to Aram-naharaim. True, mentioning this possibility could complicate the servantʼs mission considerably, thus this issue does not recur in the narrative plot at all. In terms of the narration, this option represents a “surplus,” a kind of blind motive. It is therefore all the more remarkable that the issue is introduced in the exposition, not in passing at all but in triplicate, with clear emphasis:

Historicism,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 85–108. 56  E. B lum , Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 386–387. 57  The biblical quotations in English are taken from the New Revised Standard Version.

318 Genesis 24:5b Genesis 24:6b Genesis 24:8b

Erhard Blum

‫׃‬ ‫את ִמ ָ ּֽׁשם‬ ָ ‫ל־ה ָא ֶרץ ֲא ֶׁשר־יָ ָצ‬ ָ ‫ת־ּבנְ ָך ֶא‬ ִ ‫ֶה ָה ֵׁשב ָא ִׁשיב ֶא‬ ‫׃‬ ‫ת־ּבנִ י ָ ֽׁש ָּמה‬ ְ ‫ן־ּת ִׁשיב ֶא‬ ָ ‫ִה ָּׁש ֶמר ְלָך ֶּפ‬ ְ ‫ַרק ֶא‬ ‫׃‬ ‫ת־ּבנִ י לֹא ָת ֵׁשב ָ ֽׁש ָּמה‬

Obviously, this emphatic instruction by Abraham (including its foundation, the divine oath, according to v. 7) is addressed to the intended recipients of the narrative.58 It apparently represented a significant issue outside the text in their social world. Suffice it to imagine the constellation of “kinship outside the land – marriage – question of staying in/leaving the land” in order to understand that it points strongly to the context of Persian Judah (or later). Earlier periods are out of the question because this issue does not relate to an (always conceivable) idea of deportation or exile. Rather, the issue of leaving the land to marry kin pertains to the existence of some sort of diaspora59 alongside the “homeland” and proper behavior in this constellation. Last but not least, the proposed dating of Gen 24 is corroborated stratigraphically by intratextual connections of this chapter to both non-Priestly and Priestly texts in the ancestral stories. All in all, one gets the impression that the narrative was designed as a mosaic piece fit to the present context. The narrative, naturally, presupposes the basic plot of the Abraham stories, especially his “exodus” from Mesopotamia (24:4, 7 → Gen 12) and Isaacʼs late birth (v. 36 → Gen 18; 21). At the same time, our text resumes terms and phrases of the older tradition in an accumulative manner, such as the terms for Abrahamʼs former “home” in vv. 4a and 7a* (compare 12:1 and 31:13b) and the description of his wealth in v. 35 (compare 12:16 and 13:2).60 H. Gunkel long ago remarked that the narrator had the older Jacob story in mind regarding Labanʼs role (in contrast  If it is merely the implementation of divine Heilsgeschichte that underlies the narrative logic (J. Joosten, personal communication), why should the possibility that human behavior might endanger Godʼs plan to keep Israel in its land be conceived and why should it be introduced so emphatically? The whole issue is apparently based on collective experience in which most fundamental matters seem to be called into question. Against this “historical” backdrop, Gen 24 nevertheless also implies an intratextual reference to Jacobʼs route to Mesopotamia; see below, n. 62. 59  The proposed reading does not imply an allegorical interpretation of Gen 24. That is, the Aramean protagonists are not understood as symbols or metaphors for the Jewish diaspora in the Persian period. Rather, Abraham and his family (through Isaac) represent Israel at its beginning. The family of Nahor, Abrahamʼs brother, in the narrated world of Gen 24 represents the lineage group that allows Abrahamʼs son to marry legitimately. It is this analogy to the relationship between Judeans and Jewish diaspora in Mesopotamia that would no doubt be well understood by the addressees. 60  Compare also v. 35a with 26:12b, 13; cf. Van Seters, Abraham, 247, who rightly points out that Gen 24 (probably deliberately) underlines that Abrahamʼs wealth has come from Godʼs blessing. 58

The Linguistic Dating of Biblical Texts

319

to Bethuelʼs factual absence) in the plot.61 In v. 59, finally, the story seizes the opportunity to introduce Rebekahʼs nurse, who will play a role in Gen 35:8. Nevertheless, as A. Rofé has noted, the story contains densely woven references to the surrounding Priestly texts as well. Right at the beginning, Abrahamʼs words, ‫( אשר לא תקח אשה לבני מבנות הכנעני אשר אנכי יושב בקרבו‬v. 3b), call strongly to mind Isaac sending his son Jacob “to Paddan-aram to the house of Bethuel,” in order to “take as wife from there one of the daughters of Laban” (!) (Gen 28:2), for ‫( לא תקח אשה מבנות כנען‬28:1, 6).62 At its end, the narrative resumes the preceding episode of Sarahʼs death in Gen 23 (P): “Then Isaac brought her into his mother Sarahʼs tent.63 He took Rebekah, and she became his wife; and he loved her. So Isaac was comforted after his motherʼs death” (24:67). Given these prominent references to P texts, the verbal connections between 24:36b and Gen 25:5 (P); 24:40 and 17:1;64 and 24:62 (Beer-lahai-roi) and 25:11b (P) prove intratextual as well. At the same time, the overall profile of Gen 24 as well as differences in detail (Aram-naharaim in contrast to Paddan-aram; ‫ בנות הכנעני‬in contrast to ‫ )בנות כנען‬leave no doubt about the non-Priestly character of the story. It is post-Priestly, a mélange with Deuteronomistically marked features as well. The most important element of the Deuteronomistic legacy, the promise of land as divine oath in 24:7,65 provides the thread that leads to a closely related group of Fortschreibungen in Genesis and beyond.66 In the narrow context of Gen 24, this group includes the following divine promise segments: (a) the second speech of the angel in Gen 22 (15–18), (b) an extension of the divine speech in 26:3–5, and finally (c) Gen 15.  H. Gunkel, Genesis (3rd ed.; HK 1/1; Göttingen: Vandenhoeck & Ruprecht, 1910), 256.  Given the textual pragmatics outlined above, it stands to reason that the emphasis on the interdiction against bringing Isaac to Aram-naharaim aims also to exclude drawing the “wrong conclusions” from Isaacʼs/Jacobʼs example in Gen 28 (P). 63  The syntax of ‫ האהלה שרה אמו‬seems somewhat awkward. D. Kimhi, however, points to ‫( הארץ כנען‬Num 34:20) and ‫( הארון הברית‬Josh 3:14) (see Rofé, “Betrothal” [see n. 42], 51), and Ibn Ezra explains it as an ellipsis for ‫ האהלה אהל שרה אמו‬through pointing inter alia to ‫הנבואה‬ ‫( עודד הנביא‬2 Chr 15:8) and ‫( יי׳ אלהים צבאות‬Ps 59:6). Thus the alternative reading “into the tent to his mother Sarah,” at the cost of an emendation of “his mother” in the last sentence to “his father,” seems unnecessary, pace Blum, Komposition (see n. 56), 384. 64  See also Gen 48:15(–16), words of blessing that interrupt the narrative thread of 48:13–14+17–20 (n.b. the “angel who has redeemed me from all harm” in v. 16). 65  For its dependence on Deuteronomistic tradition, see B lum , Komposition (see n. 56), 370–377, and more concisely Rofé, “Enquiry” (see n. 42), 32–33. 66  For the following, see the detailed discussion in Blum, Komposition (see n. 56), 362–399. Unfortunately, I had not paid enough attention then to the post-Priestly nature of the main texts in question. Therefore (and for additional reasons), I partially revised my KD-hypothesis in E. Blum, “Die literarische Verbindung von Erzvätern und Exodus: Ein Gespräch mit neueren Endredaktionshypothesen,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 119–156, here 151–155. 61 62

320

Erhard Blum

Genesis 24 ‫ ואשר נשבע לי לאמר‬7 . . . ‫וְ יִ ַירׁש זרעְך את שער שנאיו‬ . . .60

Genesis 22 . . . ‫בי נשבעתי נאם יְ הוָ ה‬ . . . 16 . . . ‫והרבה ארבה את זרעך ככוכבי השמים‬ . . . 17 ‫וְ יִ ַרׁש זרעך את שער איביו‬ ‫ והתברכו בזרעך כל גויי הארץ‬18 ‫ֵע ֶקב אשר שמעת בקלי‬ Genesis 26

‫ ואשר נשבע לי לאמר‬7 . . . ‫לזרעך אתן את הארץ הזאת‬

‫כי לך ולזרעך אתן את כל הארצת האל‬ . . . 3  ‫ת־ה ְּׁש ֻב ָעה‬ ַ ‫  והקמתי ֶא‬ ‫  אשר נשבעתי לאברהם אביך‬ ‫ והרביתי את זרעך ככוכבי השמים‬4  ‫  ונתתי לזרעך את כל הארצת האל‬ ‫  והתברכו בזרעך כל גויי הארץ‬ ‫ ֵע ֶקב אשר שמע אברהם בקלי‬5  ‫  וישמר משמרתי מצותי חקותי ותורתי‬

In Genesis we find divine promises to the patriarchs as oath (‫שבע‬, niphal) only in these passages.67 According to Gen 26, the promise sworn to Abraham comprised the land, the multiplied offspring, and the blessing. The other occurrences focus on different themes according to the relevant main issue in the narrative context: Gen 24 refers only to the land; the promise in 22:16–18 concerns offspring and blessing. In the latter, however, the actual performance of the divine oath is reported too, introduced by the “performative” perfect: ‫בי נשבעתי‬. Thus the question remains: In which text was the oath concerning the land referred to in 24:7 and 26:3 sworn to Abraham? There is only one option: Gen 15. The chapter ends with Godʼs “covenant” (‫)ברית‬, which consists of transferring “the land” to Abrahamʼs offspring: ‫לזרעך נתתי את הארץ הזאת מנהר מצרים עד‬ ‫( הנהר הגדל נהר פרת‬15:18b).68 The solemn self-commitment called ‫ ברית‬is equivalent to an oath. That is why ‫ ברית‬and ‫( שבועה‬or its verbal derivatives) function as synonyms in human treaty contexts (e.g., Gen 26:28, 31 and 31:44, 52) and in theological language, especially in Dtn/Dtr contexts:

67  Cf. also Gen 50:24 without quoting the promise. The synopsis in the main text highlights the tight connection, especially between the divine speeches in 22:16–18 and 26:3*–5. For a more detailed description, see B lum, Komposition (see n. 56), 363–364, with reference to the 1972 dissertation of R. Kessler , later published as Die Querverweise im Pentateuch: Überlieferungsgeschichtliche Untersuchung der expliziten Querverbindungen innerhalb des vorpriesterlichen Pentateuchs (BEATAJ 59; Frankfurt am Main: Lang, 2015), 98–99. 68  The qatal form ‫ נתתי‬functions again as a “performative perfect.”

The Linguistic Dating of Biblical Texts

Deuteronomy 7:8 Deuteronomy 8:18 Deuteronomy 4:31

321

‫משמרו את השבעה אשר נשבע לאבתיכם‬ ‫למען הקים את בריתו אשר נשבע לאבתיך‬ ‫ולא ישכח את ברית אבתיך אשר נשבע להם‬

That Gen 15 is indeed part of the discussed web of promise passages can be confirmed by several additional links. (a) In contrast to the promise of land, the promise of multiplied offspring as oath is purposefully connected with the context of the Akedah (Gen 22). Nevertheless, not only the issue but also the idiomatic imagery of 22:17, “like the stars of heaven,” are solemnly introduced in Gen 15 in a special scene (v. 5): “He brought him outside and said, ‘Look toward heaven and count the stars, if you are able to count them.’ Then he said to him, ‘So shall your descendants be.’ ” (b) The promises in Gen 22 and 26 are explicitly founded on Abrahamʼs piety and observance (“because you have/he has obeyed my voice,” see above); in Gen 15 the reference to Abrahamʼs credit (“your reward shall be very great”) precedes the promises. This theological profiling is found only here 69 in the ancestral stories. (c) The singularly expansive territorial concept of the land according to 15:18–19 (from the Nile to the Euphrates) corresponds to the plural designation in 26:3, 4 (‫)כל הארצת האל‬, which is also unique and corresponds to Isaac being in the Philistine area. (d) The steward of Gen 24 (‫ )זקן ביתי‬is introduced in 15:3 as ‫בן ביתי‬. Moreover, such a figure does not appear elsewhere in Abrahamʼs household.70 The programmatic discourse of Gen 15 apparently represents the basic text for the outlined group of Verheißungsreden linked with Gen 24. Therefore, it is no surprise that once again we find affinities to “D” traditions together with links to Priestly texts,71 including a controversial reference. Godʼs self-predication in 15:7 in particular shows a mingling of Deuteronomistic and Priestly phraseology recalling the Decalogue (Exod 20:2; Deut 5:6) and a common Deuteronomistic phrase concerning the land, ‫ לרשתה‬. . . ‫ הארץ‬. . . ‫( נתן‬Deut 3:18; 12:1, etc.; see 69  Including the scribal expansion in 26:24 (especially the last sentence), which is obviously connected to 26:4, 5 (and 15:1); cf. Blum, Komposition (see n. 56), 365. 70  The text in 15:2b is awkward. It seems partly corrupt (‫ דמשק‬. . . ‫)בן משק‬. But even without ‫דמשק‬, vv. 2–3 remain strangely repetitive; vv. 2b, 3a could be an insertion with ‫אליעזר‬ being induced by the connection with Gen 14 (gematria with 14:14). A conclusive analysis, however, is out of reach. We shall also refrain from the other diachronic problems discussed in this chapter. 71  See J. Ha, Genesis 15: A Theological Compendium of Pentateuchal History (BZAW 181; Berlin: de Gruyter, 1989), 94–103.

322

Erhard Blum

also Gen 28:4b), but even more Lev 25:38 together with the P-tradition concerning Ur-kasdim (Gen 11:31). Moreover, one cannot fail to notice the contrasting correspondence between Abrahamʼs faith in 15:6 and his laughing in 17:17, both times reacting to the promise of offspring. As for direction of dependence, it is easier to suppose that the patriarchʼs praised faith in 15:6 was introduced to correct his behavior in Gen 17:17,72 rather than vice versa.73 The intratextual connections and dependencies just noted imply that Gen 15 must be located at an advanced stage in the formation of Genesis/the Pentateuch. This is in accordance with its genre and individual Gestalt: the text can scarcely be called a “narrative.” Rather, it represents a programmatic divine-promise speech “set on stage.” Given this overall profile, it is no surprise that there is a growing consensus in recent European scholarship dating the chapter to the exilic/postexilic periods.74 In my understanding, the Persian-period dating for Gen 15 should be preferred, not only on the basis of the tight linkages between Gen 15 and the Gen 24 group of promise texts but especially for the very structure of the pericope and its significance for the readers/hearers. From the beginning to the end it is about the inheritance of the land: at first God promises his help and a great present, “wrapped,” so-to-speak, into the indefinite “Your reward is very great” (v. 1). Surprisingly, Abrahamʼs first question focuses not on the rewardʼs nature but on the possibility of gaining an heir, due to his own childlessness. The answer he gets is comprehensive (vv. 4–7): first he receives the promise of offspring that will be as countless as the stars. Abrahamʼs acceptance of this promise with unreserved trust (‫)והאמן‬75 is “reckoned to him” as fidelity/loyalty (‫)צדקה‬. So  See also the “correcting” reference to P from Gen 24 noted above in n. 62.  See recently M. Köckert, “ ‘Glaube’ und ‘Gerechtigkeit’ in Gen 15,6,” ZTK 109 (2012), 415–444, here 418 (with literature). 74  J. Hoftijzer, Die Verheissungen an die drei Erzväter (Leiden: Brill, 1956), being “der Vogel, der vor dem Morgen singt” with his dating Gen 15 to the exilic period. For orientation concerning the actual research see the recent contributions of Köckert, “Glaube” (see n. 73); idem , “Gen 15: Vom ‘Urgestein’ der Väterüberlieferung zum ‘theologischen Programmtext’ der späten Perserzeit,” ZAW 125 (2013), 25–48. 75  The widespread interpretation of ‫ והאמן‬in v. 6 as a unique copulative perfect in a carefully designed prose text does not seem plausible unless one conjectures a technical mistake (Old Hebrew yod misread as heh). But there is no need for that. It can be explained within the framework of the standard Hebrew verbal system as a subtle syntactical device for expressing relative simultaneity (without durative Aktionsart) in the past context; see E. B lum , “Das althebräische Verbalsystem: Eine synchrone Analyse,” in Sprachliche Tiefe: Theologische Weite (ed. O. Dyma and A. Michel; Biblisch-Theologische Studien 91; Neukirchen-Vluyn: Neukirchener Verlag 2008), 91–142, here 124–132. This use of yiqtol or weqatal respectively is known from so-called circumstantial sentences and from the use of long yiqtol with ‫ אז‬or ‫ ;טרם‬but compare also the semantically equivalent syntagm in 1 Kgs 13:3: ‫ונתן ביום ההוא‬ (= ‫)אז יתן‬. In the case of Gen 15:6, ‫ והאמן‬denotes that Abraham trusted/believed in God while God was speaking to him. In English or German it is difficult to translate this meaning and 72 73

The Linguistic Dating of Biblical Texts

323

God, for his part, can answer, introducing himself solemnly with his name and announcing his ‫צדקה‬: he brought Abraham out of Ur-kasdim ‫לתת לך את הארץ‬ ‫הזאת לרשתה‬. Finally, the rewardʼs “content” is also revealed: it is the inheritance of the land! All has come to a good end: the questions are answered, the trustful relationship is ensured. Nevertheless, this is precisely where the greater, second part of the “story” begins. Abraham picks up the Stichwort ‫ ירש‬in Godʼs answer in order to ask a new question: ‫במה אדע כי אירשנה‬. What happened to the faithful patriarch? What more does he need? This seeming contradiction in Abrahamʼs behavior, of course, has not escaped the attention of several literary critics, who conclude that the chapter must be composite. In fact, however, the inconsistency indicates that the “story” goes beyond the scope of the protagonistʼs character and that the “I” in his question is vicarious: it is the “I” of the descendants, that is, of the intended readers. The breathtaking covenant ritual, implying horribile dictu a conditioned self-condemnation of God, is meant for them. Last but not least, that is why the patriarchʼs question, “How am I to know that I shall possess it?” can finally be answered by the oath “To your offspring I give this land.” The old ancestor narratives explained the possession of the land by telling etiological stories: Why are Ammon and Moab not living in this land? Because their ancestor Lot went away from Abraham and had to flee to the mountains in the east during the catastrophe of Sodom. Why Edom? Because Jacob, not Esau, became the true heir. And so on. These are the stories of people for whom the possession of the land is as natural as the air they are breathing. The issue of the divine promises and especially the cause of Gen 15 are different altogether. This programmatic text stems not merely “from some time during which the relationship between the nation of Israel and the land of Israel was a concern of some thinkers.”76 It is different even from the experience of people after 720 BCE who might fear the fate of exile. The rhetoric of our chapter, which proves to be as lucid as it is emphatic, can be understood best as an answer to the situation after the exile, when major parts of the land were de facto in possession of other people – not least the region of Hebron, where we must locate Abraham according to the narrative context (Gen 13:18 [14:13]). Jan Joosten has presented Gen 15 as a case study for linguistic dating and has reached the conclusion that the chapter is written in CBH. Since he identifies two linguistic features as typical for LBH but nevertheless already conceivable in CBH, he tends to a dating of Gen 15 toward the end of the seventh or in the keep the syntactical structure of the Hebrew. In any case, the main intention seems to be an expression of Abrahamʼs spontaneous and unreserved trust. See also the interpretation by Köckert, “Glaube” (see n. 73), 422–426, which is along the same lines. 76  Sommer, “Dating” (see n. 55), 98. The cited sentence continues: “in other words, these features of the chapter suggest that it was written (1) sometime after the nation Israel or its tribal or familiar forebears first emerged and (2) sometime before this morning.”

324

Erhard Blum

early sixth century BCE. An important line in his argumentation is to show that Gen 15 has several usages that are “idiomatic in CBH and unattested in LBH.” In my view, Joosten has good cases for a typologically classical BH in Gen 15 indeed, especially in the realm of syntax, and I would even add to these cases the purported “copulative perfect” in v. 6.77 Nevertheless, not all of his interpretations seem convincing to me, especially his proposal to read ‫ מגן‬in v. 1 as a participle;78 in other words, I would adhere to the (presumably late) use of ‫ הרבה‬as an adjective in v. 1. Moreover, there are additional features that do not represent (evidenced) preexilic usage; one is the formula “the word of YHWH came to X” (Wortereignisformel) in v. 1. As is known, this formula occurs very frequently especially in Jeremiah and Ezekiel. All its occurrences in the Former Prophets are Dtr or post-DtrH; in Jeremiah, there is no clear evidence for it among authentic prophetic sayings. The other example is ‫( רכש‬v. 14). This lexeme occurs sixteen times in Chronicles, Ezra, and Daniel; seven times in P; four times in Gen 14; and once in Gen 15. Although it does not play any significant role in postbiblical Hebrew, it cannot be assigned to CBH; it belongs rather to later style. Either way, the main question is which conclusion can be drawn from the linguistic evidence in a text like Gen 15, even if we presumed, for the sake of argument, a typologically “pure” CBH. As argued above (see 2.3), the absence of LBH features in a given text does not per se prove its preexilic origin, because denying the ability of postexilic scribes to write CBH from the outset remains a prejudice, and a generalizing prejudice cannot be proven by either a few or many positive examples. The same is true for denying the possibility of having simultaneously written literary texts with typologically different usages (i.e., “Gleichzeitigkeit des Ungleichzeitigen”). In other words, we must distinguish methodologically (and terminologically) between typologically defined CBH and chronologically defined preexilic Hebrew. In Gen 15, a small-sized piece with presumably pseudepigraphic claims, the plausibility of finding a “purely CBH” text would not be low. Nevertheless, the linguistic approach also proves its worth in this case, for Gen 15 can be shown to belong to the same literary stratum as Gen 22:16–18; 26:3*–5 and the large narrative of Gen 24, which manifests decisive LBH features.79  See n. 75.  Since ‫ מגן‬in the meaning “give as a present” occurs otherwise only in the D-stem, the reading as participle would require a “haplology” of the mem (i.e., regular ‫)ממגן‬, which would be unique in the given constellation. Regarding the semantics, the proposed reading does not go well with the preceding “do not fear” on the one hand and with “reward” on the other (because ‫ מגן‬has the connotation “present”). With regard to ‫ הן‬in 15:3, I do not understand why its usage here should be different from the use in (e.g.) Isa 42:1 (emphasizing the following word; cf. HAL, s.v.). 79  See also the long chain of nomina in 26:5b. 77

78

The Linguistic Dating of Biblical Texts

325

In sum, this marked group of passages in the Abraham story was in all probability not written before the fifth century BCE. This can be demonstrated on the basis of the textsʼ pragmatics, their narrative and rhetorical structure, their profile in terms of Traditionsgeschichte, their stratigraphical dependencies and interrelations, and (partially) on the basis of linguistic features. Beyond the ongoing dispute about methodological questions and textual details, both disciplines – historical exegesis and Hebrew linguistics – need to work in constant and tight cooperation. Such cooperation would be of immense benefit for both, especially in terms of method.

Diachronic Linguistics and the Date of the Pentateuch Jan Joosten Dating biblical texts, if it is to be done by reasoned argument, is very difficult. The extant manuscripts of the Hebrew Bible are certainly much later than the texts they transmit, but how much later is hard to determine. Although many of the texts relate to historical situations, few present themselves as contemporary accounts. The circumstances recounted or alluded to, even if they can be dated, provide a terminus post quem, nothing more. The attribution of books to figures whose time can be approximately determined says little about the date of the texts, for such traditional attributions may be wholly or partly fallacious. The process of composition introduces further complexity. Many biblical texts are held to be the fruit of a complicated redaction history: an early source may have been complemented in later periods; distinct writings may have been combined into one single text; or a now lost original may have been completely rewritten, once or several times. Giving pentateuchal texts a single well-defined date would not be satisfactory in any of these scenarios. The issue of dating cannot simply be abandoned, however. The whole historical-critical enterprise stands or falls with the idea that the writings can be restored in one way or another to the historical background that saw them emerge. If one can find out under what circumstances a given text came into being, one will be able to interpret it better. Biblical scholars have therefore engaged in the process of dating biblical texts with gusto. But the results have not always been satisfactory. Wildly divergent dates are proposed for many of the biblical writings, even relatively large and homogeneous ones such as the Deuteronomic law code or the Joseph story. In spite of over two hundred years of research and discussion, no consensus is in view. This situation invites reflection on the methodological side of the question, on the criteria to be used, and on the hierarchy among them. A variety of methods are used in dating biblical texts, all of them imprecise and involving a measure of subjectivity. The present paper will focus on one of them, namely, the dating of texts on the basis of their language. In doing so, the intention is not to suggest that this method is a sort of shortcut allowing one to do without the other approaches. What is contended is that the linguistic criterion has fallen somewhat into disuse among biblical scholars. Specialists of the Hebrew Bible are on the whole more attracted to the development of ideas

328

Jan Joosten

than to diachronic linguistics. The central idea of this paper is to rehabilitate the linguistic approach as a viable method in biblical studies.

Classical Biblical Hebrew, Late Biblical Hebrew, and Their Historical Background* The linguistic approach to dating biblical texts is based on diachronic study of the Hebrew language. Languages change over time, and so does Hebrew over the biblical period – or more precisely, over the period during which the biblical writings were composed. A Brief History of the Historical Approach to BH The history of the Hebrew language was first conceptualized in European universities during the seventeenth century.1 An authoritative synthesis, adding much new material, was presented by Wilhelm Gesenius in his monograph of 1815, Geschichte der hebräischen Sprache und Schrift.2 Gesenius argued that an historical approach to Biblical Hebrew needed to distinguish two style forms, prose and poetry, and two epochs, preexilic and postexilic. Poetry submits less easily to the diachronic approach, notably because of the use of archaisms. After Gesenius, Samuel R. Driver added many important contributions, collecting most of them in his Introduction to the Literature of the Old Testament (1891). Arno Kropat and Rebecca Corwin independently compared the language of Chronicles to that of Samuel–Kings in monographic treatments.3 The discovery of Hebrew inscriptions from the monarchic period has had enormous impact on this type of research, as has the discovery of the Qumran scrolls. Scholars such as Yehezkel Kutscher and Avi Hurvitz fully exploited these new resources. Hurvitz also designed a methodological framework that makes * Abbreviations: BH = Biblical Hebrew, CBH = Classical Biblical Hebrew, LBH = Late Biblical Hebrew. 1  One of the earliest applications of this type of research was Hugo Grotiusʼs discussion of the date of Qoheleth, in which he argued that the Hebrew of this book makes it nearly impossible to attribute it to Solomon. See H. Grotius, Annotata ad Vetus Testamentum (3 vols.; Paris: Cramoisy, 1644), 1:521. 2  W. Gesenius, Geschichte der hebräischen Sprache und Schrift (Leipzig: Vogel, 1915). See J. Joosten, “Wilhelm Gesenius and the History of Hebrew in the Biblical Period,” in Biblische Exegese und hebräische Lexikographie: Das “Hebräisch-deutsche Handwörterbuch” von Wilhelm Gesenius als Spiegel und Quelle alttestamentlicher und hebräischer Forschung, 200 Jahre nach seiner ersten Auflage (ed. S. Schorch and E.-J. Waschke; BZAW 427; Berlin: de Gruyter, 2013), 94–106. 3  See A. Kropat, Die Syntax des Autors der Chronik (Gießen: Alfred Töpelmann, 1909); R. Corwin, The Verb and the Sentence in Chronicles, Ezra and Nehemiah (Borna: R. Noske, 1909).

Diachronic Linguistics and the Date of the Pentateuch

329

explicit what many others practiced.4 Many other researchers also contributed to the issue.5 Frank Polak showed that the CBH corpus is not homogeneous but divides into at least two subcorpora, one closely following the cast of oral literature and the other reflecting more openly the workshop of scribes.6 David Talshir argued on good grounds that the historical background of LBH in its fullblown form is to be looked for among the exiles in Babylonia. Elisha Qimron identified many features attesting continuity from LBH to Qumran Hebrew.7 He and some other Hebraists contributed to the understanding of the syntax of LBH, hitherto a rather neglected field.8 Nevertheless, the basic parameters established in Geseniusʼs monograph continue to be valid today.9 This is not because Hebraists are a traditionalist and unimaginative bunch but because the data on which the approach is based speak out with particular clarity. Illustrations Some of the evidence for the diachronic approach is well-known and has often been rehearsed.10 This should not numb our perception of its effectiveness. Let us consider one example. In the books of Samuel, Kings, Isaiah, and Jeremiah, 4  E. Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll (I Q Isaa) (STDJ 6; Leiden: Brill, 1974); A. Hurvitz, The Transition Period in Biblical Hebrew: A Study in Post-Exilic Hebrew and Its Implications for the Dating of Psalms (Jerusalem: Mosad Bialik, 1972) (Hebrew). 5  See, e.g., the studies in I. Y oung (ed.), Biblical Hebrew: Studies in Chronology and Typology (JSOTSup 369; London: T&T Clark, 2003); C. Miller-Naudé and Z. Zevit (eds.), Diachrony in Biblical Hebrew (Winona Lake, IN: Eisenbrauns, 2012). 6  See, e.g., F. Polak, “The Oral and the Written: Syntax, Stylistics and the Development of Biblical Prose Narrative,” JANES 26 (1998), 59–105. 7  See, e.g., E. Q imron , The Hebrew of the Dead Sea Scrolls (HSS 29; Atlanta: Scholars Press, 1986), 88–95. 8  See, e.g., E. Qimron, “Consecutive and Conjunctive Imperfect: The Form of the Imperfect with Waw in Biblical Hebrew,” JQR 77 (1987), 151–153; M. Eskhult, “Verbal Syntax in Late Biblical Hebrew,” in Diggers at the Well: Proceedings of a Third International Symposium on the Hebrew of the Dead Sea Scrolls and Ben Sira (ed. T. Muraoka and J. F. Elwolde; STDJ 36; Leiden: Brill, 2000), 84–93; J. J oosten , The Verbal System of Biblical Hebrew: A New Synthesis Elaborated on the Basis of Classical Prose (Jerusalem Biblical Studies 10; Jerusalem: Simor, 2012); 377–410; O. C ohen , The Verbal System in Late Biblical Hebrew Prose (HSS 63; Winona Lake, IN: Eisenbrauns, 2013). 9  Joosten, “Wilhelm Gesenius” (see n. 2). 10  Meaningful evidence of evolution can take many forms, but what is required in every instance is the attestation of at least two phenomena, one of which will be claimed as early and the other as late. The fact that a linguistic item is found only in “late” writings does not show that the word is late, since it may be unattested in early texts by chance. For some examples, see R. H endel , “Is the ‘J’ Primeval Narrative an Independent Composition?: A Critique of Crüsemannʼs ‘Die Eigenständigkeit der Urgeschichte,’ ” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; Tübingen: Mohr Siebeck, 2011), 181–205, in particular 201.

330

Jan Joosten

a “letter” – a written missive, usually sent by the king but occasionally also by a private person (Jer 29:1, 25, 29) – is invariably called ‫ס ֶפר‬, ֵ twenty-one times in all. In the books of Esther, Nehemiah, and Chronicles, the same meaning is expressed, ten times, by the word ‫;אּגֶ ֶרת‬ ִ alongside this word, ‫ ֵס ֶפר‬is used six times in Esther and once in Chronicles (rephrasing a parallel in Kings; see 2 Chr 32:17, and compare 2 Kgs 19:14) but never in Nehemiah. Both words can be dated approximately. The noun ‫ ֵס ֶפר‬is a good Hebrew word well integrated into the language as a whole.11 The meaning “letter” is a specialization of the more general meaning “writing”; the word also means “document, inscription, book.” As a designation of a letter, the word is frequently attested in epigraphic Hebrew from the monarchic period, notably with about a dozen occurrences ִ for its part, is an Aramaic loanword.12 in the Lachish letters. The noun ‫אּגֶ ֶרת‬, There is no epigraphic Hebrew of the Persian period to speak of, but Aramaic ‫ ִאּגֶ ֶרת‬is well attested in Achaemenid documents, notably in the Elephantine letters. It is not excessively speculative to submit that the native word for letter, ‫ס ֶפר‬, ֵ was gradually replaced with an Aramaic loanword, ‫אּגֶ ֶרת‬, ִ in the Hebrew of the Persian period. Most letters written during this period would have been in Aramaic, thus making it natural to refer to them with an Aramaic word. The replacement had the added advantage of eliminating the ambiguity arising from the use of ‫ס ֶפר‬, ֵ which could refer to all kinds of written documents, in particular books. The conclusion must be that ‫ ֵס ֶפר‬in the meaning “letter” is an earlier word and ‫ ִאּגֶ ֶרת‬a later one. In regard to this one item, Samuel–Kings reflects an earlier type of Hebrew than do Chronicles, Nehemiah, and Esther. One might object that the earlier word continued in use during the later period: in Esther ‫ ֵס ֶפר‬is found six times, ‫ ִאּגֶ ֶרת‬only twice. The use of ‫ ֵס ֶפר‬in the meaning “letter” is therefore not by itself indicative of earlier Hebrew. It remains remarkable, however, that Samuel–Kings never uses ‫אּגֶ ֶרת‬. ִ Also, the use of both the earlier and the later lexeme in Esther and Chronicles is only natural. Languages evolve gradually, as is known from many studies in general linguistics. Words and constructions do not all of a sudden disappear when a more modern equivalent turns up; there is a period of overlap.13 Of course, one single pair like this does not suffice to establish a relative chronology. But there are many analogous cases: ‫שׁש‬, ֵ “fine linen,” in the Pen11  According to HALOT, ‫ ֵס ֶפר‬is a loanword from Akkadian, but this is very unlikely. See P. V. M ankowski , Akkadian Loanwords in Biblical Hebrew (HSS 47; Winona Lake, IN: Eisenbrauns, 2000), 121–122. 12  The origin of the word is debated, but it certainly came to Hebrew from Aramaic. 13  See, e.g., the following studies in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Winona Lake, IN: Eisenbrauns, 2012): B. E. Dresher, “Methodological Issues in the Dating of Linguistic Forms: Considerations from the Perspective of Contemporary Linguistic Theory,” 19–38; J. A. C ook , “Detecting Development in Biblical Hebrew Using Diachronic Typology,” 83–96; and R. D. H olmstedt , “Historical Linguistics and Biblical Hebrew,” 97–124.

Diachronic Linguistics and the Date of the Pentateuch

331

tateuch and Proverbs corresponds to ‫ּבּוץ‬, “fine linen,” in Esther and Chronicles (in Ezekiel both words are found); ‫ֹמועד‬, ֵ “appointed time,” in the Pentateuch and historical books corresponds to ‫ זְ ָמן‬in Qoheleth, Esther, and Nehemiah (‫ֹמועד‬, ֵ “appointed time,” is found in Daniel, but in Ezra, Nehemiah, and Chronicles the word invariably means “festival”); ‫ יָ ם‬and ‫מבֹוא ַה ֶּשׁ ֶמשׁ‬, ְ “west,” in the Pentateuch, Joshua, and Ezekiel correspond to ‫ ַמ ֲע ָרב‬in Second Isaiah and Chronicles (Daniel has both ‫ יָ ם‬and ‫)מ ֲע ָרב‬. ַ Another phenomenon that may be used to get a handle on the evolution of the Hebrew language in the biblical period is the change in meaning of individual words and expressions. The verb ‫ עמד‬means “to take oneʼs stand, to be in a standing position” in most of the biblical books, but in Ezra–Nehemiah, Chronicles, Qoheleth, Esther, and Daniel it also expresses the ingressive meaning “to stand up,” which is more typical of CBH ‫( קום‬the change also affects the hiphil of ‫)עמד‬. In this case, the change in meaning seems to have come about for internal reasons. With other words, it happens under influence from Aramaic. In the Pentateuch and Former Prophets, the root ‫ בהל‬expresses notions of terror and dismay. In Qoheleth, Esther, and Chronicles, the root more often means “to hasten,” a meaning familiar from Aramaic (see also Prov 20:21 according to the qərê). There are many more examples like this. Hurvitzʼs recent Lexicon of Late Biblical Hebrew lists eighty items characteristic of LBH that find precise counterparts in CBH.14 The Two Main Corpora and the Nature of the Difference between Them On the basis of a great number of case studies, linguists have proposed the division of the biblical books into two main linguistic corpora, one written in Classical Biblical Hebrew, or CBH (also called Standard BH or Early BH), and one in Late Biblical Hebrew, or LBH. The former corpus consists of the Pentateuch and Former Prophets and the latter of the books of Qoheleth, Esther, Ezra, Nehemiah, Chronicles, and Daniel. Books that do not figure in either corpus are mostly written in poetry and are thus more difficult to categorize. Nevertheless, at least some of these books can be meaningfully related to the diachronic division into CBH and LBH. Notably, Jeremiah, Ezekiel, Isaiah 40–66, Job, and Lamentations appear to represent a type of transitional Hebrew standing between the two main corpora.15 As was  A. Hurvitz, A Concise Lexicon of Late Biblical Hebrew: Linguistic Innovations in the Writings of the Second Temple Period (SVT 160; Leiden: Brill, 2014). 15  On Jeremiah, see now A. Hornkohl, Ancient Hebrew Periodization and the Language of the Book of Jeremiah: The Case for a Sixth-Century Date of Composition (Studies in Semitic Languages and Linguistics 74; Leiden: Brill, 2013); on Ezekiel, see M. R ooker , Biblical Hebrew in Transition: The Language of the Book of Ezekiel (Sheffield: Sheffield Academic, 1990); on Second Isaiah, see S. M. Paul, “Signs of Late Hebrew in Isaiah 40–66,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Winona Lake, IN: Eisen14

332

Jan Joosten

stated already, languages evolve gradually, and the division into just two corpora is certainly artificial to a certain extent. With regard to the two main corpora, there should be no doubt that they are indeed systematically different in their language use. Most of the arguments fielded in the discussion have come from the realm of the lexicon. Consequently, some critics of the diachronic approach have expressed the idea that CBH and LBH do not reflect different ages in the writing of Hebrew but different styles – the one more conservative, the other more open to innovations.16 This suggestion would have weight if the differences between the two pertained to vocabulary only. One can indeed imagine an author who, when writing about the monarchic period (as in Samuel–Kings), would try to avoid Aramaic loanwords such as ‫אּגֶ ֶרת‬, ִ “letter.” But the difference between the two corpora goes much further. Notably, there are many contrasted usages in the area of syntax.17 This is not the place to explore this type of evidence at length, but one example may illustrate. In CBH, the weqatal form (perfect consecutive) is regularly used, side by side with yiqtol, to express repetition in the past: Genesis 2:6

‫י־ה ֲא ָד ָמה‬ ָ ֵ‫ל־ּפנ‬ ְ ‫ת־ּכ‬ ָ ‫ן־ה ָא ֶרץ וְ ִה ְׁש ָקה ֶא‬ ָ ‫וְ ֵאד יַ ֲע ֶלה ִמ‬ A mist would go up from the earth and would water the whole face of the ground.

There are over 150 examples of this usage in CBH. In LBH, in contrast, yiqtol is still used in this way, but weqatal almost never.18 Instead, LBH uses the participle, or weyiqtol (weyiqtol is never found in this function in CBH): Nehemiah 9:28

‫וְ ַא ָּתה ִמ ָּׁש ַמיִ ם ִּת ְׁש ַמע וְ ַת ִּצ ֵילם ְּכ ַר ֲח ֶמיָך ַרּבֹות ִע ִּתים‬ In your compassion you would hear from heaven and would rescue them time and again.

The difference coheres with a well-known development of the Hebrew verbal system over the longer term, namely, the obsolescence of the so-called converted tenses. In Mishnaic Hebrew weqatal will completely disappear as a specific verbal form. brauns, 2012), 293–300; on Job, see J. Joosten, “Linguistic Clues as to the Date of the Book of Job: A Mediating Position,” in Interested Readers: Essays on the Hebrew Bible in Honor of David J. A. Clines (ed. J. K. Aitken et al.; Atlanta: Society of Biblical Literature, 2013), 347–357; on Lamentations, see F. W. Dobbs-Allsopp, “Linguistic Evidence for the Date of Lamentations,” JANES 26 (1998), 1–36. 16  This is one of the claims in I. Y oung et al., Linguistic Dating of Biblical Texts: An Introduction to Approaches and Problems (2 vols.; London: Equinox, 2009). 17  For bibliographical references, see above in notes 3 and 8. 18  See J. Joosten, “The Disappearance of Iterative WEQATAL in the Biblical Hebrew Verbal System,” in Biblical Hebrew in Its Northwest Semitic Setting: Typological and Historical Perspectives (ed. S. E. Fassberg and A. Hurvitz; Jerusalem: Magnes / Winona Lake, IN: Eisenbrauns, 2006), 135–147.

Diachronic Linguistics and the Date of the Pentateuch

333

The syntactic evidence speaks against the notion that CBH and LBH are to be regarded as style forms. Though one can imagine ancient authors consciously selecting certain types of vocabulary, it would have been much harder for them systematically to distinguish two sets of syntactic constructions. Rather than plead the improbable, one should admit that the linguistic data indeed indicate that the two corpora, CBH and LBH, reflect distinct periods in the history of the Hebrew language. The Historical Background of Classical Biblical Hebrew and Late Biblical Hebrew Distinguishing the two main corpora is rather straightforward. Determining the relative priority of CBH over LBH also raises few problems. Relating each type of Hebrew to a definite period in the history of Israel is much harder. Since Gesenius, most linguists have associated CBH with the period of the monarchy and LBH with the postexilic period. The lateness of LBH is still agreed upon by all. The LBH corpus includes books that openly refer to events and circumstances of the Persian period (Esther, Ezra, Nehemiah, and Chronicles). In addition, there are at least sixteen different Persian loanwords in the LBH corpus, distributed over all the books in this corpus.19 The time frame of CBH has come to be debated, however. It has been remarked that the whole scheme could fit into the postexilic period: CBH could reflect the sixth and fifth centuries BCE and LBH the late Persian and early Hellenistic periods.20 There are good linguistic grounds, however, to resist this suggestion and to keep to the earlier position. (a) Firstly, a comparison with Hebrew inscriptions makes it possible to anchor CBH in the monarchic period. Above we saw that whereas ‫ ִאּגֶ ֶרת‬designates Persian-period letters (written in Aramaic), ‫ ֵס ֶפר‬is the auto-referential term used in preexilic Hebrew letters. Linguistic features common to CBH and epigraphic Hebrew of the seventh and sixth centuries are numerous.21 Not all of them are significant, but some are. The inscriptions attest many words and constructions ַ “first,” is that in the Bible are exclusive to CBH. For instance, the adverb ‫ּכּיֹום‬, found in Gen 25:31, 33; 1 Sam 2:16; 9:27; 1 Kgs 1:51; 22:5 (and // 2 Chr 18:4):

 See now the study of A. Wilson-Wright, “From Persepolis to Jerusalem: A Reevaluation of Old Persian-Hebrew Contact in the Achaemenid Period,” VT 65 (2015), 152–167. One of the words also occurs in Ps 144:13 (‫ )זַ ן‬and one in Song 4:13 (‫)ּפ ְר ֵּדס‬. ַ 20  See C. L evin , review of Linguistic Evidence for the Pre-exilic Date of the Yahwistic Source, by R. M. Wright, RBL, Jan. 7, 2006 http://www.bookreviews.org/bookdetail.asp? TitleId=4860&CodePage=4860 (accessed 8/10/15). 21  See, e.g., S. Landis Gogel, A Grammar of Epigraphic Hebrew (Atlanta: Scholars Press, 1998). 19

334 1 Kings 22:5 First seek an oracle from the LORD.

Jan Joosten

‫ת־ּד ַבר יְ הוָ ה‬ ְ ‫ְּד ָרׁש־נָ א ַכּיֹום ֶא‬

The adverb is also attested six times in the Lachish letters (Lak 1.2.3; 1.5.3; 1.4.2; 1.8.2). In later Hebrew, whether in the Bible or elsewhere, the particle is no longer found. Its meaning was forgotten, as is shown, inter alia, by the fact that the Septuagint renders it with σήμερον, “today,” in most passages.22 Even more significant similarities have been noted in the realm of verbal syntax. Several usages characteristic of CBH but unusual, or unattested, in LBH do occur in the inscriptions:23 1. the performative use of qatal (‫ברכתך‬, “I hereby bless you”; Arad 16:2–3, etc.);24 2. use of the infinitive absolute in peremptory commands (‫נתן‬, “give [said by a superior]”; Arad 2:1, etc.);25 3. the paronomastic use of the infinitive absolute (‫שלח שלחת‬, “sending I have sent”; Mur 1:2);26 4. the use of weqatal continuing imperatives (‫ ולקחת‬. . . ‫בא‬, “go in . . . and take”; Arad 17:1–4; see also Arad 2:1–5).27 “Isoglosses” like this show that something very much like CBH was written in Judah before the exile. Admittedly, the inscriptions cannot testify as to when CBH began or when it came to an end. There is very little epigraphic Hebrew before the eighth century and practically none between the latest Arad and Lachish letters and the Qumran scrolls. (b) Secondly, as was mentioned before, a number of biblical writings appear to come between CBH and LBH in the development of the language. Some of these can with some confidence be tied to the sixth century. The book of Ezekiel, which exhibits a mix of CBH and LBH traits, also contains a striking number of loanwords from Babylonian.28 This feature confirms the connection of the book  Some Septuagint translators seem to have a correct idea of what the particle means; see, e.g., 1 Sam 2:16. 23  See J. J oosten , “The Distinction between Classical and Late Biblical Hebrew as Reflected in Syntax,” Hebrew Studies 46 (2005), 327–339. 24  See Eskhult, “Verbal Syntax” (see n. 8), 86. 25  See Kropat, Syntax (see n. 3), 23; Eskhult, “Verbal Syntax” (see n. 8), 90. 26  See Eskhult, “Verbal Syntax” (see n. 8), 90. The usage is attested seven times in LBH (1 Chr 4:10; 21:17, 24; 2 Chr 18:27; 32:13; Esth 4:14; 6:13), but it is much more frequent in CBH (almost three hundred examples). 27  See Eskhult, “Verbal Syntax” (see n. 8), 87. 28  See, e.g., A. W initzer , “Assyriology and Jewish Studies in Tel Aviv: Ezekiel among the Babylonian Literati,” in Encounters by the Rivers of Babylon: Scholarly Conversations between Jews, Iranians, and Babylonians in Antiquity (ed. U. Gabbay and S. Secunda; Tübingen: Mohr Siebeck, 2014), 163–216. 22

Diachronic Linguistics and the Date of the Pentateuch

335

to the historical figure whose name it bears. If exilic writings stand between CBH and LBH, this strongly suggests that the former is preexilic and the latter postexilic. Admittedly, the notion of a transitional phase presents some complications. Isaiah 40–66, Haggai, and Zechariah, although they appear to come from the same general period, have some distinctly late features, but not as many as Ezekiel.29 (c) Thirdly, there are signs of a break between CBH and LBH. The differences between the two phases cannot all be attributed to natural language development. Some of the authors of LBH books visibly try to link up with the language of earlier literature but fail do so. More tellingly, later authors at times take over words and expressions from CBH in a way that indicates that they did not understand them correctly.30 The phenomenon shows that the Hebrew language of the LBH corpus is partially artificial. Many words and phrases did not come to the LBH authors from the natural Hebrew with which they grew up but from the study of ancient texts that were only partially understood. Of course, those “ancient texts” may not have been exactly the ones we read in the Bible today. But the whole phenomenon does indicate a kind of gap between CBH and LBH. The one did not organically grow into the other; there was a disruption. This fits the traditional chronology. Between the monarchic period and the time of Ezra and Nehemiah, the Judean elite, who would have been the prime “carriers” of any type of national literature, were forcibly removed from their original setting. This is precisely the type of event that could account for the break between the two main corpora. An interesting example illustrating the break is the use of the word ‫ֹיומם‬. ָ In CBH, this word is found fifteen times, always meaning “by day.” Indeed, ‫ֹיומם‬ ָ is to be analyzed as the noun ‫ֹיום‬, “day,” with the adverbial ending -am known also from ‫חּנָ ם‬, ִ “by grace,” and ‫ּדּומם‬, ָ “silently.”31 The usage is found also in poetic texts. In Nehemiah, however, the word is used as a noun meaning “daytime”: Nehemiah 9:19

‫יֹומם‬ ָ ‫יהם ְּב‬ ֶ ‫א־סר ֵמ ֲע ֵל‬ ָ ֹ ‫ת־עּמּוד ֶה ָענָ ן ל‬ ַ ‫ֶא‬ The pillar of cloud did not stop guiding them in the path during the day.  On Haggai, see G. A. R endsburg , “Late Biblical Hebrew in the Book of Haggai,” in Language and Nature: Papers Presented to John Huehnergard on the Occasion of His 60th Birthday (ed. R. Hasselbach and N. Pat-El; Chicago: Oriental Institute, 2012), 329–344. 30  See J. Joosten, “Pseudo-Classicisms in Late Biblical Hebrew, in Ben Sira, and in Qumran Hebrew,” in Sirach, Scrolls and Sages: Proceedings of a Second International Symposium on the Hebrew of the Dead Sea Scrolls, Ben Sira, and the Mishnah, Held at Leiden University, 15–17 December 1997 (ed. T. Muraoka and J. F. Elwolde; Leiden, Brill, 1999), 146–159; idem, “The Evolution of Literary Hebrew in Biblical Times: The Evidence of Pseudo-Classicisms,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Winona Lake, IN: Eisenbrauns, 2012), 281–292. 31  See GKC 100g. 29

336

Jan Joosten

The addition of the preposition ‫ ב‬shows that ‫ֹיומם‬ ָ was analyzed as a noun meaning “daytime.” A similar usage is found in the Qumran scrolls and in some other sources.32 The archaic adverbial ending -am seems no longer to have been understood. This paved the way for an association of ‫ֹיומם‬ ָ with the Aramaic noun ‫ימם‬, “day (as opposed to night).” Instead of an adverb derived from a noun, Hebrew ‫ֹיומם‬ ָ was understood as a noun distinct from ‫ֹיום‬, the former designating only the daytime, the latter the whole day. Differences like this originate not in natural developments but in the desire of later generations to use older forms of the language, even if they are no longer exactly understood. In light of these considerations, the conclusion is warranted that biblical books written in CBH were created over a period coeval with the time of the monarchy. Exactly when the period of writing in CBH came to an end is hard to say. Judeans who had mastered CBH during the last years of the monarchy would not suddenly have forgotten it when the state came to an end. Whether they succeeded in passing on this skill to the next generation is doubtful. The available data do not make it likely that they did.

Implications for the Dating of Pentateuchal Texts The diachronic approach to Biblical Hebrew was not designed primarily with a view to dating biblical texts. Once there is a diachronic framework, however, it becomes possible to relate individual writings – books or shorter passages – to the scheme. Texts written in CBH are to be dated to the monarchic period or shortly thereafter and texts written in LBH to the Persian period or later. Three case studies will illustrate the relevance of this approach to pentateuchal studies. The Pentateuch: A Writing of the Monarchic Period A first inference to be drawn from the diachronic framework is that the Pentateuch is to be regarded substantially as preexilic. Ascribing large parts of the Pentateuch to the Persian period, as is done routinely by many OT scholars, is impossible to reconcile with the linguistic data. The prose parts of the Pentateuch are for the most part written in CBH: Aramaic loanwords are relatively scarce and Persian words are lacking, and the syntax is regular and quite distinct from that of LBH. This is true even for relatively late layers of pentateuchal composition. Genesis 12:10–20, the story of Abram and Sarai in Egypt, provides a nice illustration. As has increasingly been recognized among critical scholars, the  See P. P iovanelli , “JrB 33,14–26 ou la continuité des institutions à lʼépoque maccabéenne,” in The Book of Jeremiah and Its Reception (BETL 128; ed. A. H. W. Curtis and T. Römer; Leuven: Peeters, 1997), 255–276. 32

Diachronic Linguistics and the Date of the Pentateuch

337

passage relates the patriarchal narratives to the story of the exodus.33 A whole network of verbal allusions presents Abram and his wife as prototypes of the Israelites in Egypt. One important effect of this is to tie the books of Genesis and Exodus closer together. As many exegetes have observed in recent years, these two books are rather different in their themes and theological outlooks.34 They reflect different human experiences, and the underlying traditions may go back to different milieus. Our story and very few other passages serve as literary clamps between these books, thus underscoring the essential unity of salvation history. Although the passage cannot be qualified properly as redactional, it does fulfill a function similar to that of redactional passages. In recent writing, Gen 12:10–20 has therefore been dated very late, after the Priestly layer of the Pentateuch, itself regarded as late exilic at the earliest.35 But a date in the Persian period, say around the middle of the fifth century or later, is very difficult to accept in view of the language of the passage. The text has no marked features of LBH. It exemplifies a brilliant CBH style found also in other patriarchal stories, as well as in the “succession narrative” in Samuel– Kings. Although the passage is short, it exhibits several expressions typical of CBH that are never found in LBH: Genesis 12:11

‫ת־מ ְר ֶאה ָא ְּת‬ ַ ‫ל־ׂש ַרי ִא ְׁשּתֹו ִהּנֵ ה־נָ א יָ ַד ְע ִּתי ִּכי ִא ָּׁשה יְ ַפ‬ ָ ‫אמר ֶא‬ ֶ ֹ ‫וַ ּי‬ He said to his wife Sarai, “Look, I know that you are a beautiful woman.”

The syntagm ‫ ִהּנֵ ה־נָ א‬is attested twenty-two times in the CBH corpus and four times in the poetry of Job, but not even once in LBH. These statistics are not an accident. The modal particle ‫ נָ א‬is vastly more frequent in CBH than in LBH. Moreover, whereas in CBH it is syntactically very flexible, being attached to verbs and particles and even once to a noun in the vocative, in LBH its use is severely restricted: it only occurs immediately following a modal verbal form (imperative, jussive, or cohortative). The limited distribution of the particle in LBH shows that it is, so to speak, “on its way out.” In Mishnaic Hebrew it is used only in petrified expressions such as ‫אנא‬, “please.” Genesis 12:12

‫וְ ָהיָ ה ִּכי־יִ ְראּו א ָֹתְך ַה ִּמ ְצ ִרים וְ ָא ְמרּו ִא ְׁשּתֹו זֹאת‬ When the Egyptians see you they will say, “This is his wife.”

 See, e.g., T. Römer, “The Exodus in the Book of Genesis,” SEÅ 75 (2010), 1–20.  See, e.g., K. Schmid, “Genesis and Exodus as Two Formerly Independent Traditions of Origins for Ancient Israel,” Bib 93 (2012), 187–208. 35  See, e.g., J. C. Gertz, “The Transition between the Books of Genesis and Exodus,” in A Farewell to the Yahwist (SBL Symposium Series 34; ed. T. B. Dozeman and K. Schmid; Atlanta: Society of Biblical Literature, 2006), 73–87, in particular 75. 33 34

338

Jan Joosten

Temporal ‫ ִּכי‬is frequent in CBH but virtually absent from LBH.36 Genesis 12:15 So the woman was taken into the household of Pharaoh.

‫וַ ֻּת ַ ּ֥קח ָה ִא ָ ּׁ֖שה ֵ ּ֥בית ַּפ ְר ֽעֹה‬

The passive qal is a regular feature in CBH, but it is almost completely lacking in LBH. The passive qal of ‫ לקח‬is attested fifteen times: seven times in the CBH corpus, seven times in prophetic books (Jeremiah, Ezekiel, and Second Isaiah), and once in Job, but never in LBH. The parallel to Gen 12:15 in the Esther story is ‫ל־ּבית ַה ֶּמ ֶלְך‬ ֵ ‫ – וַ ִּת ָּל ַקח ֶא ְס ֵּתר ֶא‬with a niphal instead of the passive qal (Esth 2:8). The three features enumerated show that Gen 12:10–20 is written in CBH. The nature of the usages virtually excludes the possibility that a later author would have been able to imitate them. In light of what was said before, this would suggest that the story was created, more or less in the form we have it, during the time of the monarchy. In a pinch, it might be dated soon after the fall of Jerusalem. Taking it down to 450 BCE, or later, seems completely impossible. Late Language in the Pentateuch? The Pentateuch is, from a linguistic point of view, remarkably unified. It is hard to detect developments from book to book or from one stratum to another. The Priestly code stands out for its religious cast and its literary presentation, but it reflects roughly the same type of Hebrew as the non-P sections.37 The same is true for the “Deuteronomistic” parts of Deuteronomy. One finds unusual features here and there, among them features that will gain currency in LBH. But generally they do not accumulate in any meaningful way.38 There are a few exceptions, however. In some passages, the presence of late features suggests that an addition was made to the original text, or that 36  See E. J. Revell, “Conditional Sentences in Biblical Hebrew Prose,” in Semitic Studies in Honor of Wolf Leslau (ed. Alan S. Kaye; Wiesbaden: Harrassowitz, 1991), 1278–1290. 37  See A. Hurvitz, “The Evidence of Language in Dating the Priestly Code – A Linguistic Study in Technical Idioms and Terminology,” RB 81 (1974), 24–56; idem, A Linguistic Study of the Relationship between the Priestly Source and the Book of Ezekiel: A New Approach to an Old Problem (CahRB 20; Paris: Gabalda, 1982). Linguistic evidence for a later date of P was presented by R. Polzin, Late Biblical Hebrew: Toward an Historical Typology of Biblical Hebrew Prose (HSM 12; Missoula: Scholars Press, 1976), but most of it is inconclusive; see G. A. Rendsburg, “Late Biblical Hebrew and the Date of ‘P,’ ” JANES 12 (1980), 65–80. 38  This is true even for Gen 15 (as I hope to show in a future publication) and for Gen 24. The linguistic arguments brought forward by Alexander Rofé are interesting, but they do not show that Gen 24 is written in LBH. See A. Rofé, “An Enquiry into the Betrothal of Rebekah,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag (ed. E. Blum et al.; Neukirchen-Vluyn: Neukirchener Verlag, 1990), 27–39; and the critical review of G. A. Rendsburg, “Some False Leads in the Identification of Late Biblical Hebrew Texts: The Cases Of Genesis 24 and 1 Samuel 2:27–36,” JBL 121 (2002), 23–46.

Diachronic Linguistics and the Date of the Pentateuch

339

an originally CBH text was rewritten at a later period. A possible example is found in Num 36:1–12. The story on the rectification of the legislation regarding heiresses is dependent on, and secondary to, the earlier story of the daughters of Zelophehad in Num 27.39 The style is priestly, but some of the vocabulary is not (note ‫שׁ ֶבט‬, ֵ “tribe,” instead of the expected ‫ ַמ ֶּטה‬in v. 3). The pericope is widely regarded as a late addition. The language is very wordy. A few turns of phrase are frankly bizarre (notably ‫ וַ אד ֹנִ י ֻצּוָ ה ַביהוָ ה‬in v. 2). But verbosity and stylistic unwieldiness are not the prerogative of any single age. There is, however, one linguistic feature that definitely points to a late date: Numbers 36:1 ‫יֹוסף וַ יְ ַד ְּברּו ִל ְפנֵ י מ ֶֹׁשה‬ ֵ ‫ן־מנַ ֶּׁשה ִמ ִּמ ְׁש ְּפחֹת ְּבנֵ י‬ ְ ‫ן־מ ִכיר ֶּב‬ ָ ‫אׁשי ָה ָאבֹות ְל ִמ ְׁש ַּפ ַחת ְּבנֵ י־גִ ְל ָעד ֶּב‬ ֵ ‫וַ ּיִ ְק ְרבּו ָר‬ ‫אׁשי ָאבֹות ִל ְבנֵ י יִ ְׂש ָר ֵאל‬ ֵ ‫וְ ִל ְפנֵ י ַהּנְ ִׂש ִאים ָר‬ The heads of the ancestral houses of the clans of the descendants of Gilead, son of Machir son of Manasseh, of the Josephite clans, came forward and spoke in the presence of Moses and the leaders, the heads of the ancestral houses of the Israelites.

The implication of the verse is simply that the heads of the clan of Gilead spoke to Moses and the other leaders. But the formulation is peculiar: instead of “to speak to,” with the preposition ‫ אל‬or ‫ל‬, as is normal in BH, the text has the expression “to speak before.” The usage finds a parallel only in the book of ֶ ַ‫ו‬, “Esther spoke again before the Esther (Esth 8:3, ‫ּתֹוסף ֶא ְס ֵּתר וַ ְּת ַד ֵּבר ִל ְפנֵ י ַה ֶּמ ֶלְך‬ king”).40 The background of this LBH usage is well understood. It reflects court language of the Persian period and is well attested in Official Aramaic.41 One did not speak to the king or his high officers but before them. The convention has left many traces in Biblical Aramaic. In Jewish Aramaic texts, it was projected into the spiritual realm: one prays before God, as commonly in the Targums. In our passage, the usage is not theological. Moses and the leaders are high officials, like the king in Daniel and the Sheikh el Fadl inscription, or like Arsham in the Elephantine letters. “To speak before” instead of “to speak to” is a deferential turn of phrase borrowed from Official Aramaic. One does not normally date a text on the basis of a single linguistic feature. An accumulation of data is necessary.42 One feature might be just a blip; if a text 39  See S. C havel , Oracular Law and Priestly Historiography in the Torah (FAT 2/71; Tübingen: Mohr Siebeck, 2014), 243–250. 40  The ostensibly similar instances in Exod 6:12 and 1 Kgs 3:22 are to be explained differently: Moses and the prostitutes speak not to God and the king, respectively, but before him in a spatial sense. 41  See J. J oosten , “Lʼaraméen de Qumran entre lʼaraméen dʼempire et les targumim: Lʼemploi de la préposition ‘devant’ pour exprimer le respect dû au roi et à Dieu,” in Aramaica Qumranica: Proceedings of the Conference on the Aramaic Texts from Qumran in Aix-enProvence, 30 June–2 July 2008 (ed. K. Berthelot and D. Stökl Ben Ezra; Leiden: Brill, 2010), 83–97. 42  This is one of the methodological points on which Avi Hurvitz has insisted in many of his publications.

340

Jan Joosten

is late, this should show up in several usages. In the present case, however, the late character of the expression is particularly clear. Moreover, the language of the rest of the passage is amenable to a late date, even though it contains no other features demanding such a date. The paucity of marked features may reflect the fact that much of the passage is faithful to the phraseology of Num 27:1–11. In aggregate, the linguistic data may be taken to indicate that Num 36:1–12 was written during the Persian period or later.

Objections to the Linguistic Approach The linguistic approach leads one to date the bulk of the Pentateuch to the preexilic period. Only some very few and obviously secondary passages may have been added after the Babylonian exile. Such a view of the time frame of the composition of pentateuchal texts would have raised few eyebrows forty years ago. But in contemporary pentateuchal studies, particularly in Europe, the tendency is to give a much later date to the texts. In response to the divergence between dates arrived at on the basis of language and dates proposed on other grounds, a number of objections to the linguistic approach have been formulated. The principal objections merit brief discussion. Archaizing? The most common response to the linguistic approach as applied to the Pentateuch is to invoke the possibility of archaizing. If the Pentateuch is written in a language that aligns with Hebrew inscriptions of the monarchic period, does this necessarily mean that the biblical texts were composed in that period or shortly thereafter? Can one not imagine that later authors, in the fifth, fourth, or third century BCE, were able to write Classical Hebrew in a way that could fool present-day linguists? Later authors, particularly authors supplementing earlier pentateuchal texts, may have attempted to write early Hebrew, thus creating the impression that the text is old, whereas in fact it is not. Perhaps Gen 12:10–20 was composed at the end of the fifth century by an author who had a good grasp of pentateuchal style. Archaizing did certainly happen in later stages of Hebrew composition. The phenomenon is well illustrated by some Qumran writings, such as the Damascus Document or the Temple Scroll. Although these writings are widely recognized as Hellenistic compositions, they are written in a style that comes close to that of the Pentateuch. When medieval manuscripts of the Damascus Document were discovered in the Cairo Geniza, at the end of the nineteenth century, Solomon Schechter, the editor of the text, qualified the language as follows: The language of the MS. is for the most part pure Biblical Hebrew. The first three pages rise even to the dignity of Scriptural poetry, though a good deal of it is obscured by

Diachronic Linguistics and the Date of the Pentateuch

341

the unfortunate condition in which the text is at present. But there are in it terms and expressions which occur only in the Mishna or even only in the Rabbinic literature dating from the first centuries of the Middle Ages.43

Schechterʼs astute judgment shows why the phenomenon of archaizing, at least in the way it manifests itself in the Qumran scrolls, cannot be used to give CBH passages in the Pentateuch a postexilic date. The author of the Damascus Document took great pains to adhere to the classical language he knew from close study of scriptural texts. Nevertheless, he could not refrain from using a number of expressions reflecting the Hebrew of his time. The later elements give away the date of his writing.44 The same principle operates in earlier samples of archaizing. As was argued above in regard to Num 36, a few sections of the Pentateuch consist of later supplements imitating the Hebrew style of the earlier text on which they elaborate. As in the Dead Sea Scrolls, the later Fortschreiber tripped up and betrayed himself by the use of a later expression. An interesting additional example is the account of the building of the tabernacle, in Exod 35–39. As is well known, large parts of this account are missing in the Septuagint. Many scholars hold that this indicates the existence of a Hebrew text, older than the MT, in which these chapters were absent. In this earlier version, the divine instructions on the building of the tabernacle, related in Exod 25–31, were simply presupposed to have been carried out. The language of the later chapters bears this out: usually the wording follows that of the earlier chapters to the letter, but on occasion, a later expression creeps in: Exodus 36:10

‫ל־א ָחת‬ ֶ ‫ל־א ָחת וְ ָח ֵמׁש יְ ִריעֹת ִח ַּבר ַא ַחת ֶא‬ ֶ ‫ת־ח ֵמׁש ַהיְ ִריעֹת ַא ַחת ֶא‬ ֲ ‫וַ יְ ַח ֵּבר ֶא‬ And he coupled the five curtains one to another: and the other five curtains he coupled one to another. Exodus 26:3

‫ל־אח ָֹתּה‬ ֲ ‫ל־אח ָֹתּה וְ ָח ֵמׁש יְ ִריעֹת ח ְֹבר ֹת ִא ָּׁשה ֶא‬ ֲ ‫ֲח ֵמׁש ַהיְ ִריעֹת ִּת ְהיֶ ין ָ ח ְֹבר ֹת ִא ָּׁשה ֶא‬ The five curtains shall be coupled together one to another; and the other five curtains shall be coupled one to another.

Whereas Exod 26:3 uses the CBH expression ‫ל־אח ָֹתּה‬ ֲ ‫א ָּשׁה ֶא‬, ִ literally “a woman to her sister,” for “the one to the other,” Exod 36:10 uses ‫ל־א ָחת‬ ֶ ‫א ַחת ֶא‬, ַ “one to one,” with the same meaning. The construction ʾeḥād–preposition–ʾeḥād is not found elsewhere in the CBH corpus, but it does occur in Ezekiel, Job, and Qoheleth (Ezek 37:17; Job 41:8; Qoh 7:27). In the Samaritan Pentateuch, it is substituted for the CBH expression wherever it occurs (see Exod 26:3). The late scribe who supplemented the report on the building of the tabernacle  S. Schechter, Fragments of a Zadokite Work (Cambridge: Cambridge University Press, 1910), xi. 44  See also Joosten, “Evolution of Literary Hebrew” (see n. 30). 43

342

Jan Joosten

scrupulously imitated the style of the divine instructions. Nevertheless, in the phraseology, he introduced a late element. Neither the pseudo-classical language of the Damascus Document nor the use of late expressions in Num 36:1 or Exod 36:10 proves that the idea of later scribes imitating CBH perfectly is impossible. Perhaps in other late passages the archaizing was more successful. But the available evidence does not confirm this possibility. There are late writings – certainly dating to the Persian period or beyond – that evince an effort to write Classical Hebrew. But to the best of my knowledge there are no late writings where this effort was consistently successful. The Stratified Nature of the Pentateuch Another frequent objection is that the linguistic approach is problematic because biblical texts are typically stratified and do not reflect one single period.45 Most knowledgeable scholars do indeed agree that the Pentateuch was not created in one go. Different parts and strata came into being over a rather long period of time, perhaps even over centuries. It is not clear, however, why the supplementary process of redaction should make the Pentateuch immune to linguistic analysis. If some of the strata were added in the postexilic period, one would expect those strata to reflect postclassical Hebrew.46 If no LBH elements can be located, this indicates that all the strata were composed during the period when CBH was in use. The monarchic era is a very long one. There is no problem of principle in postulating multiple rewritings before the cut-off point of 550 BCE or so. What about Textual Criticism? A more spurious type of objection is that the uncertainty of the text of the Hebrew Bible renders the linguistic approach ineffective.47 It is true that tex L evin , review of Wright (see n. 20), 1: “[T]he writings of the Old Testament, starting from their earliest compilation, developed unremittingly through internal biblical interpretation, and in this way they have become so complex from a literary point of view that it is hardly possible to isolate significant reference texts for a particular linguistic stage. The Pentateuch, for example, extends from the twelfth century, the possible date of the Song of Miriam (Exod 15:21), until the Hellenistic period.” 46  In Jeremiah, Masoretic pluses exhibit a later stage of Hebrew writing than the older parts of the book reflected in the Old Greek; see J. Joosten, “Lʼexcédent massorétique du livre de Jérémie et lʼhébreu post-classique,” in Conservatism and Innovation in the Hebrew Language of the Hellenistic Period: Proceedings of a Fourth International Symposium on the Hebrew of the Dead Sea Scrolls and Ben Sira (ed. J. Joosten and J.-S. Rey; STDJ 73; Leiden: Brill, 2008), 93–108. 47  See, e.g., Y oung et al., Linguistic Dating (see n. 16), 341–360; and in more extended form: I. Young and R. Rezetko, Historical Linguistics and Biblical Hebrew: Steps Toward 45

Diachronic Linguistics and the Date of the Pentateuch

343

tual problems interfere with the linguistic analysis of an ancient language like Hebrew. But they do not do so to the point of throwing the entire approach into doubt. In regard to the Pentateuch, two things must be said. Firstly, the MT of the Pentateuch is very conservative and well preserved. There are cases where a reading of the Septuagint, the Samaritan Pentateuch, or a Qumran fragment has to be preferred. But they are not very many. Secondly, where the Hebrew text is problematic, intelligent analysis very often confirms – rather than contradicts – the language-historical approach. A famous example is the ostensible presence of a Persian word, ‫ּדת‬,ָ “law,” in Deut 33:2 according to the qərê: “on his right hand was a fiery law for them.” Textual criticism will show that the qərê is a secondary development,48 even without taking account of the fact that the presence of a Persian word in this passage is implausible. Modernizations Finally, it has been alleged that the biblical text may have been “modernized” by later scribes and that the version accessible to us may not reflect the earliest stage of the writings. The answer to this objection is as before: perhaps modernization did happen, but if so, it must have happened over the period when CBH was in use. The MT of the Pentateuch shows very few signs of modernization during the postexilic period.49 In making this statement, I am disregarding matters of script, orthography, and pronunciation. What a modernized Pentateuch might look like can be seen in the Samaritan Pentateuch. Here we indeed have a large number of CBH forms that were changed so as to agree with norms of the Second Temple period.

Conclusions At the close of this article I will try to preempt two opposite reactions it may evoke. To some readers, the presentation of the language-historical approach may seem overly complex and unwieldy. What can linguistics contribute to the dating of biblical texts if the validity of every single feature involves so many different steps, each one of them submitting to various qualifications? Against this impression, I would like to underline that the diachronic approach to Biblical Hebrew as it has evolved over the last two hundred years has a certain an Integrated Approach (Atlanta: Society of Biblical Literature, 2014). 48  See, e.g., BHQ ad loc. 49  For some possible cases, see J. Joosten, “Textual Developments and Historical Linguistics,” in After Qumran: Old and Modern Editions of the Biblical Texts – The Historical Books (ed. H. Ausloos et al.; BETL 246; Leuven: Peeters, 2012), 21–31.

344

Jan Joosten

inbuilt robustness. The division into just two corpora – Genesis to 2 Kings as against Qoheleth, Esther, Ezra, Nehemiah, Chronicles, and Daniel – may be overly schematic, but it makes up for this by its sharpness. If any chapter of Ezra or Chronicles had been interpolated into the CBH corpus, linguists would be liable to catch it. To postulate that CBH and LBH are mere forms of style is fanciful. The two varieties reflect distinct – and as was argued above, not even contiguous – periods in the history of the language. There is no denying that the procedures of historical linguistics are complex. The complexities are well charted, however. When the linguistic method is used with circumspection it may lead to dependable results. To others, the linguistic approach may on the contrary appear to lack sophistication. A text written in LBH is postexilic and a text written in CBH is preexilic. Can it be so simple? The method seems to be out of step with recent research on the Hebrew Bible. I will plead guilty to feeling out of step with much current research. But I would take issue with the suggestion that the contribution of historical linguistics is unsophisticated. It is indeed coarse-grained, presenting the big picture, but this is a direct corollary of the robustness to which I already alluded. The attribution of CBH to the preexilic period is not precise at all. A text written in CBH may originate from the seventh, the eighth, or, for all we know, the ninth or tenth century BCE. Conversely, the date may be lowered to the early sixth century, or perhaps a little later still. However, to extend the period of creative writing in CBH to the fifth century or later means taking the approach beyond the breaking point. Dating a text like Gen 12:10–20 to the late Persian period can only be done if its language is left out of consideration. Leaving linguistic data out of consideration would be a mistake. At the same time, the linguistic approach can be, and must be, combined with other approaches. Textual and historical data may be expected to go hand in hand with linguistic clues – and they often do. Where they do not, the different criteria should be confronted with one another. Perhaps the linguist will have to bow his or her head and allow that some forms of CBH did continue to be practiced after 500 BCE. Or perhaps the historical critic could be brought to admit that some of her or his late dates originated as mere thought experiments and have little basis in fact. Barring truly astonishing epigraphic discoveries, the dating of biblical texts will remain difficult. But by bringing all available inputs to bear on the question, at least the extent of our ignorance will be more precisely defined.

Linguistic Dating, Writing Systems, and the Pentateuchal Sources William M. Schniedewind Languages change over time. The only question is how this universal linguistic fact may be applied to our study of the Hebrew Bible and, in the present case, the Pentateuch. To be sure, there have been challenges to linguistic dating,1 but these have not been convincing to most scholars who specialize in language.2 With regard to the Pentateuch, the linguistic dating of the texts has not advanced much beyond the studies of Hurvitz, Polzin, and Milgrom from decades ago.3 To be sure, diachronic studies on the words and syntax have certainly been important and needed, but they have not fundamentally changed the landscape. This is because it is the accumulation of linguistic features that is necessary for dating, as Hurvitz has emphasized in his methodology.4 Thus, a debate over a word or

 I. Y oung (ed.), Biblical Hebrew: Studies in Chronology and Typology (JSOTSup 369; London: T&T Clark, 2003); I. Y oung et al., Linguistic Dating of Biblical Texts: An Introduction to Approaches and Problems (2 vols.; London: Equinox, 2009); and R. Rezetko and I. Y oung , Historical Linguistics and Biblical Hebrew: Steps Towards an Integrated Approach (SBL Ancient Near East Monographs 9; Atlanta: Society of Biblical Literature, 2014). 2  For example, see C. Miller-Naudé and Z. Zevit (eds.), Diachrony in Biblical Hebrew (Linguistic Studies in Ancient West Semitic 8; Winona Lake, IN: Eisenbrauns, 2012); J. Joosten, review of Linguistic Dating of Biblical Texts, by I. Young et al., Bible und Babel 6 (2014), 535–542; N. Pat-El and A. Wilson-Wright, “Features of Archaic Biblical Hebrew and the Linguistic Dating Debate,” Hebrew Studies 54 (2013), 387–410. In the first book edited by Young, Biblical Hebrew (see n. 1), the diachronic study of Hebrew was posed as a scholarly dialogue. In the subsequent work of Young and R ezetko (see n. 1), they have become the advocates for a new approach to biblical Hebrew that dismisses the traditional diachronic understanding of Biblical Hebrew. 3  See the review of scholarship in A. Hornkohl, Ancient Hebrew Periodization and the Language of the Book of Jeremiah (Leiden: Brill, 2014), 1–50; D.-H. K im , Early Biblical Hebrew, Late Biblical Hebrew and Linguistic Variability: A Sociolinguistic Evaluation of the Linguistic Dating of Biblical Texts (Leiden: Brill, 2013), 11–44. 4  See most recently A. H urvitz , “Once Again: The Linguistic Profile of the Priestly Material in the Pentateuch and Its Historical Age: A Response to J. Blenkinsopp,” ZAW 112 (2000), 180–191. 1

346

William M. Schniedewind

syntagma does not substantively change the overall analysis.5 Likewise, there have been significant studies of selected chapters and sections of the Pentateuch, but they have not changed the overall understanding of the linguistic character of the Pentateuch and rely on the distinction between Standard Biblical Hebrew (SBH) and Late Biblical Hebrew (LBH).6 Perhaps it is for this reason that during the past several years the debate about linguistic dating has largely centered around methodology. The methodological and historical arguments do have something to add to the discussion of linguistic dating. In this paper, I review some of the methodological debates and make historical and methodological observations that should be germane to our dating of the pentateuchal sources, and in particular, the Priestly source. I highlight one methodological point – namely, the importance of the linguistics of writing systems – and one historical point – namely, the disruption in the Hebrew scribal infrastructure that helps separate Standard Biblical Hebrew (SBH) from Late Biblical Hebrew (LBH). Finally, three examples illustrate these observations. What are the causes of linguistic change? Most scholars recognize it begins with historical processes. So, for example, the use of the exile as a watershed for the dating of Biblical Hebrew has been standard.7 In addition, scholars have pointed out the phenomenon of “languages in contact,”8 namely, that during the exile and Persian period the Hebrew language came into contact with Aramaic.  For example, the debate over the words ʿedah and qahal as evidence of diachronic development or different idiolects or registers does not affect the overall picture. See, for example, E. Meyer, “Dating the Priestly Text in the Pre-Exilic Period: Some Remarks about Anachronistic Slips and Other Obstacles,” Verbum et Ecclesia 31/1 (2010) (doi: 10.4102/ ve.v31i1.423). Sperling has suggested that the term miknasayim really means “two-legged trousers,” which only were introduced in the Persian period, and that the term is a genuine anachronism in the P source; S. D. Sperling, “Pants, Persians and the Priestly Source,” in Ki Baruch Huʾ: Ancient Near Eastern, Biblical, and Judaic Studies in Honor of Baruch A. Levine (ed. R. Chazan et al.; Winona Lake, IN: Eisenbrauns, 2003), 373–385. I have reservations with even this example, as it seems to be reading late meanings back into the analysis of miknasayim, which itself is not a late term or loanword. 6  For example, A. Rofé, “An Enquiry into the Betrothal of Rebekah,” in Die hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag (ed. E. Blum et al.; Neukirchen-Vluyn: Neukirchener Verlag, 1990), 27–39; cf. G. A. Rendsburg, “Some False Leads in the Identification of Late Biblical Hebrew Texts: The Cases of Genesis 24 and 1 Samuel 2:27–36,” JBL 121 (2002), 23–35. 7  The history of scholarship is exhaustively reviewed by Young et al., Linguistic Dating of Biblical Texts (see n. 1), 1–71; an important contribution to the use of the exile as a watershed has been David Talshirʼs research, for example, D. Talshir , “The Habitat and History of Hebrew during the Second Temple Period,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 251–275. 8  Note, for example, A. Hurvitz, “Relevance of BH Linguistics for the Historical Study of Ancient Israel,” in Proceedings of the Twelfth World Congress of Jewish Studies: Division A; The Bible and Its World (Jerusalem, 1999), 28*. 5

Linguistic Dating, Writing Systems, and the Pentateuchal Sources

347

The use of these models, however, must be careful to distinguish between speech and writing systems.9 To be sure, the same historical forces are at work on speech and on writing; however, the way historical events shape vernacular language can be quite different from the way they shape writing systems. For example, vernacular Hebrew would have continued to be spoken in any Judean town or village that did not experience settlement dislocation.10 If there is continuity in settlement, then there is continuity in spoken language. Archaeologists estimate that there was as much as an 80 percent decline in settlements.11 Without going into the various regional variations, there was a great disjunction in the Hebrew speech communities following the Babylonian incursions in 597, 586, and 581 BCE. At the same time, there were a few villages that continued to be settled, and the general observation of speech communities is that they continue to speak their language until the community is dislocated. Thus, Judean villages not dislocated in the wake of the Babylonian campaigns would have continued to speak Hebrew. Written languages, in contrast, are dependent on social, political, and economic systems that transcend local settlements. Scribes are dependent on social and economic infrastructures for their continued training and replication. The scribal tradition did not immediately disappear as a result of the Babylonian invasions, but the infrastructure that supported it economically and politically no longer supported its continuation. In the present case, Aramaic immediately replaced Hebrew in the administration of the southern Levant by the Babylonians and later the Persians. For this reason, the fate of written and spoken Hebrew during the sixth through fourth centuries BCE was quite different. Another observation important to the methodological debate can be drawn from the burgeoning field of sociolinguistics.12 Language ideology, for example,  The distinction between writing and speech has long been recognized in the study of Hebrew. G. A. Rendsburg devoted a monograph to this phenomenon, Diglossia in Ancient Hebrew (AOS 72; New Haven: American Oriental Society, 1990). More recently, it is a central point in S. A. Kaufmanʼs review of Diachrony in Biblical Hebrew, ed. C. Miller-Naudé and Z. Zevit, in RBL [http://www.bookreviews.org] (2014). As Kaufman points out, not enough methodological attention is given to this distinction even though it is widely acknowledged. 10  See my discussion in W. M. S chniedewind , A Social History of Hebrew: Its Origins through the Rabbinic Period (New Haven, CT: Yale University Press, 2013), 128–138. 11  Estimates vary considerably, but the overall patterns are most critical. See recently A. Faust , Judah in the Neo-Babylonian Period: The Archaeology of Destruction (Atlanta: Society of Biblical Literature), 119–148; I. Finkelstein, “The Territorial Extent and Demography of Yehud/Judea in the Persian and Early Hellenistic Periods,” RB 117 (2010), 39–54. 12  Some important works include S. Romaine, Socio-Historical Linguistics: Its Status and Methodology (Cambridge: Cambridge University Press, 1982); P. Kroskrity (ed.), Regimes of Language: Ideologies, Polities, and Identities (School of American Research Advanced Seminar Series; Santa Fe: School of American Research, 2000); F. Coulmas, Writing Systems: An Introduction to Their Linguistic Analysis (Cambridge: Cambridge University Press, 2002); M. Sebba, Spelling and Society: The Culture and Politics of Orthography around the World (Cambridge: Cambridge University Press, 2007). 9

348

William M. Schniedewind

is often critical to the shaping of linguistic history. David Talshirʼs study “The Habitat and History of Hebrew during the Second Temple Period,” for example, identified the type of nationalistic ideology in personal names that may have encouraged the revitalization of a Hebrew scribal tradition after the exile,13 but the Hebrew epigraphic evidence for revitalization does not begin to emerge until the fourth century BCE. A recent book by Dong-Hyuk Kim summarizes the contributions of the sociolinguistic approach and emphasizes the concept of “linguistic variability” as crucial for our understanding of linguistic change.14 In this, Kim relies on the monumental work by the sociolinguist William Labov, Principles of Linguistic Change,15 which points out that linguistic change begins as synchronic linguistic variability. This is particularly noteworthy because scholars such as Ian Young and Robert Rezetko, as well as Joseph Blenkinsopp before them, have used linguistic variability to undermine diachronic analysis of Hebrew. In actual fact, linguistic variability is central to the process of linguistic change. Labov distinguished five stages of linguistic change: incipient, new and vigorous, mid-range, nearing completion, and completion.16 Linguistic variability may add complexity to our analysis of linguistic change, but it certainly does not undermine the fundamental division between SBH and LBH. The fundamental context of linguistic change is social and historical. Indeed, the discussion of the dating of the pentateuchal sources started with larger historical syntheses such as that of Wellhausen, and the preexilic dating of the Priestly source is associated with the grand synthesis of Yehezkiel Kaufmann. For example, Blenkinsopp begins with an extended critique of the historical arguments of Kaufmann and his followers.17 Blenkinsoppʼs article was followed by two responses, by Jacob Milgrom and Avi Hurvitz.18 Milgrom dealt mostly with Blenkinsoppʼs challenges to Kaufmannʼs historical and literary arguments, and Hurvitz responded to the issues of linguistic dating. Suffice it to say, this exchange did not seem to move the debate forward. Those arguing for a preexilic date for the Priestly source found Milgrom and Hurvitzʼs rebuttals more than adequate, whereas those holding to a late date for the Priestly source felt that Blenkinsopp had nicely summarized the case and critiqued the early dating.

 D. Talshir, “The Habitat and History of Hebrew during the Second Temple Period,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 253–254. 14  Kim, Early Biblical Hebrew, Late Biblical Hebrew and Linguistic Variability (see n. 3). 15  W. Labov, Principles of Linguistic Change (2 vols.; Oxford: Blackwell, 2001). 16  Labov, Principles of Linguistic Change, 2:166–171. 17  J. Blenkinsopp, “The Alleged Preexilic Date of the Priestly Material in the Pentateuch,” ZAW 108 (1996), 495–518. 18  J. M ilgrom , “The Antiquity of the Priestly Source: A Reply to J. Blenkinsopp,” ZAW 111 (1999), 10–22; Hurvitz , “Once Again: The Linguistic Profile of the Priestly Material” (see n. 4), 180–191. 13

Linguistic Dating, Writing Systems, and the Pentateuchal Sources

349

A central issue in the methodology of Robert Polzin as well as the critique of linguistic dating by Blenkinsopp and more recently Young, Rezetko, and others is the use of lexical items as opposed to syntax for measuring linguistic change.19 The debate is somewhat neutralized by the fact that both types of evidence yield essentially the same result. Still, Kimʼs review of methodology frames the debate in a helpful manner by dividing ongoing linguistic change into two types: “changes from below social awareness” and “changes from above social awareness.”20 These are categories borrowed from sociolinguistic approaches and particularly from the study of language ideology. “Changes from below” are changes that are unconscious and “socially unnoticed.” Changes in syntax invariably fall into this category, but most lexical changes also fall into this category. The change in the verbal system certainly falls into this category. In contrast, “changes from above” are socially constructed and may be motivated by a variety of linguistic ideologies. Examples of this type of change are especially evident in Qumran sectarian Hebrew.21 Changes of this type are often important markers for group identity and are ideologically shaped. Usually changes from above are generated by the dominant social classes – to which scribes would certainly belong. Though there has been a great deal of discussion about whether syntax or lexicon is preferable for studying linguistic change, this is not an either/or problem. To be sure, lexicon is more susceptible to linguistic change “from above social awareness,” but this does not make it any less an aspect of linguistic change.22 One useful example of linguistic change “from above social awareness” for pentateuchal studies is the term torah itself. The early meaning of the term was merely “teaching,” and later it came to also refer to a written text – first in some unspecified or debated sense and later for the Pentateuch. This was certainly a loaded term and is a semantic shift coming “from above social awareness.” For example, in Neh 8 the Torah is paraded as a text. This textualization of Torah is also evident in the book of Chronicles. In Chronicles, for instance, the term  See R. Polzin, Late Biblical Hebrew: Toward an Historical Typology of Biblical Hebrew Prose (Missoula, MT: Scholars Press, 1976); R. R ezetko , “Dating Biblical Hebrew,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 245–249; Joosten, “The Distinction between Classical and Late Biblical Hebrew as Reflected in Syntax,” Hebrew Studies 46 (2005), 327–339; Joosten, The Verbal System of Biblical Hebrew: A New Synthesis Elaborated on the Basis of Classical Prose (Jerusalem Biblical Studies 10; Jerusalem: Simor, 2012), 377–410. 20  Kim, Early Biblical Hebrew, Late Biblical Hebrew and Linguistic Variability (see n. 3), 88–94. 21  See W. M. Schniedewind, “Linguistic Ideology in Qumran Hebrew,” in Diggers at the Well: Proceedings of a Third International Symposium on the Hebrew of the Dead Sea Scrolls and Ben Sira (ed. T. Muraoka and J. F. Elwolde; Leiden: Brill, 2000), 245–255. 22  The main observation about changes “from above social awareness” is that they reflect the intentional use of language change for constructing group identities. 19

350

William M. Schniedewind

torah means a text without qualification. It is not necessary to qualify the term torah by saying “the scroll of the Torah.” This is in marked contrast with the DtrH, where torah must be intentionally marked as a text – that is, as “the scroll of the torah.” What about the Priestly literature in the Pentateuch? There, torah is still simply “teaching”; it is never marked as a text. In this sense, P stands in stark contrast to other Second Temple period priestly texts. Apparently, it was the Deuteronomic tradition that was trying to reshape the semantic meaning of torah. How do we explain this in P? Must P therefore be preexilic? Does P intentionally try to emphasize a nontextual torah (in contrast to texts like Neh 8)? This is certainly an observation that deserves more reflection in pentateuchal criticism.23 Sometimes the diachronic distinction between SBH and LBH is completely dismissed. For example, it has been suggested that all the books written in SBH, including the pentateuchal sources, may have come from the postexilic period.24 This is a rather radical suggestion that does not seem to have attracted many adherents. Philip Davies articulates a more typical approach: “Did Judaean scribes really forget, within two generations [from Haggai to Nehemiah], how to write CBH [that is, SBH]?”25 That is to say, SBH could have been written by learned scribes of the Persian period. Posing the issue as a simple question, he employs a wonderful rhetorical strategy. Yet, Daviesʼs innocent question sets up a straw man in several respects. First of all, it was not only two generations. It happened over the course of two centuries. Second, we should not accept the premise that Haggai and Zechariah have no traces of LBH. Hurvitz has already critiqued this commonly repeated notion that the postexilic prophetic literature is SBH, and Shalom Paul has pointed out LBH in Deutero-Isaiah.26 Several biblical books belong to the transitional phase of the Hebrew language, and transitional phases of language are necessarily complex. Nevertheless, Transitional Biblical  Note, for example, K. Finsterbusch, “Aufsummierte Tora: Zur Bedeutung von twrh als Bezeichnung für eine Gesetzessammlung im Pentateuch,” JAJ 2 (2011), 1–28. 24  E.g., I. Y oung et al., “Linguistic Dating,” in Linguistic Dating of Biblical Texts: An Introduction to Approaches and Problems (ed. I. Young et al.; 2 vols.; London: Equinox, 2009), 1:187. Some of this is explained by recourse to models for variability in vernacular languages; however, it should be noted again that scribal communities tend to standardize writing and do not have types of variability that are seen in dialect geography. To be sure, there is variability in writing standards as well, but it operates differently from vernacular dialects. 25  P. R. Davies, “Biblical Hebrew and the History of Ancient Judah: Typology, Chronology and Common Sense,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 150–163, here 155. 26  This is point also made specifically by A. Hurvitz, “The Recent Debate on Late Biblical Hebrew: Solid Data, Expertsʼ Opinions, and Inconclusive Arguments,” Hebrew Studies 47 (2006), 191–210, here 206–207; also see S. M. Paul, “Signs of Late Biblical Hebrew in Isaiah 40–66,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; LSAWS 8; Winona Lake, IN: Eisenbrauns, 2012), 293–300. 23

Linguistic Dating, Writing Systems, and the Pentateuchal Sources

351

Hebrew can be isolated with careful philological study.27 Finally, Davies seems to assume a robust and continuous culture of Judean scribes, which is something that cannot be assumed on the basis of the archaeological and historical evidence. The reasoning seems strikingly circular: namely, many SBH texts were written in the Persian period (assumed), therefore there must have be a robust Hebrew scribal culture at that time, therefore SBH must have been a feature of the robust Hebrew scribal culture of the Persian period. Yet, the historical background for a seminal linguistic change in Hebrew is quite compelling and located between the sixth and third centuries BCE. The disruption began with the Babylonian invasions and developed in the context of Babylonian and early Persian rule.28 The Hebrew scribal infrastructure was shattered and that must have resulted in the waning of the Hebrew scribal tradition. The separate histories of the scribal and speech communities especially come into play for understanding the distinction between SBH and LBH. Vernacular language changes gradually over time, whereas changes in writing systems tend to be punctuated by major transitions around historical and social events. As long as Hebrew scribal tradition remained unbroken, new generations of students learned the meaning of older texts even as vernacular Hebrew dialects underwent continual change. The disruption of the scribal tradition during the Babylonian period, however, would create a gap in the understanding of ancient Hebrew texts and a major transition in the written Hebrew language. On the one hand, the cessation of the Iron Age Hebrew scribal tradition began the end of SBH. On the other hand, LBH draws upon the continuous vernacular tradition of Hebrew from the Iron Age into the Persian period. But the history of Biblical Hebrew is ultimately a history of the writing system, and the changes in the writing system were conditioned by seminal historical events that left a linguistic trail on the pages of the Hebrew Bible. There are basic facts that must be acknowledged for the dating of the pentateuchal sources. First, there are no Persian loanwords in the Pentateuch (and more generally SBH); this is in marked contrast to LBH and Qumran Hebrew. This is widely acknowledged but generally dismissed. Yet, it should not be 27  See most recently, Hornkohl, Ancient Hebrew Periodization and the Language of the Book of Jeremiah (see n. 3), with an extensive review of the secondary literature as well as methodology. 28  Proposals for a continuing and robust Hebrew scribal tradition through the sixth and fifth centuries BCE at Bethel have also been critiqued and were based on flawed understanding of the archaeology of the region. See I. F inkelstein and L. S inger -Avitz , “Reevaluating Bethel,” ZDPV 125 (2009), 33–48, which gives a critique of the archaeological underpinnings of J. Blenkinsopp, “Bethel in the Neo-Babylonian Period,” in Judah and the Judeans in the Neo-Babylonian Periods (ed. O. Lipschits and J. Blenkinsopp; Winona Lake, IN: Eisenbrauns, 2003), 93–107, and E. A. K nauf , “Bethel: The Israelite Impact on Judean Language and Literature,” in Judah and the Judeans in the Persian Period (ed. O Lipschits and M. Oeming; Winona Lake, IN: Eisenbrauns, 2006), 275–294.

352

William M. Schniedewind

dismissed simply or quickly. Even when we have a strong Hebrew linguistic ideology among the sectarian Qumran documents, there are Persian loanwords.29 If the corpus is large enough, like the Pentateuch or P, then there should be some linguistic evidence. Second, there must be substantial Aramaic influence in any Hebrew composition from the Persian or Hellenistic period. This is, of course, the case with books like Ezra–Nehemiah, Chronicles, Esther, and Daniel, as well as the Dead Sea Scrolls and Hebrew epigraphic evidence from the Hellenistic and Roman periods. It is not pervasive in the Pentateuch or the Priestly source in particular. Third, there are striking similarities between pentateuchal literary genres (especially treaties and legal texts) and ancient Near Eastern literature of the preexilic period. In contrast, there are no convincing or extended comparisons in the Pentateuch with Persian or Hellenistic literary genres. Comparisons with Greek literature, such as suggested by John Van Seters,30 are best classified methodologically as comparative literature and are not evidence of direct literary influence. This was underscored, for example, in the well-received book by Karel van der Toorn, Scribal Culture and the Making of the Hebrew Bible, even though van der Toorn is apparently unaware of the tension it creates when he locates a formative period for the “making of the Hebrew Bible” in the Persian period.31 Fourth, there remains a troubling gap in external evidence for Hebrew writing between the early sixth and fourth centuries BCE. Related to this lacuna is the beginning of the use of the Aramaic script that was used for copying and later composing Hebrew texts – a telltale sign of a major change in the scribal culture. Along these lines, the phenomenon of difficult hapax legomena also illustrates the disruption of the Hebrew scribal infrastructure that separates SBH from LBH. It is not at all surprising that the Bible has hapax legomena, since they are characteristic of all literary works, but the question is how to explain the difficult words. For example, if this paper had some hapax legomena that could only be explained by recourse to Middle English or Old French, it would be quite strange and should generate some serious reflection about its authorship, and this is exactly the situation that we have in the Hebrew Bible. As illustrated in Chaim Cohenʼs research, the best comparative Semitic resources for under H urvitz responds (again) to the issue of Persian loanwords dismissed by Young and Rezetko in “The Recent Debate on Late Biblical Hebrew” (see n. 16), 203–204. 30  For example, J. Van Seters, The Life of Moses: The Yahwist as Historian in Exodus– Numbers (Louisville: Westminster John Knox, 1994), 34. To wit, David Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (New York: Oxford University Press, 2005), summarizes early Greek textuality and education for purely comparative purposes and does not suggest that there was direct “influence” as there was with Mesopotamian and Egyptian textuality and education (pp. 91–109). 31  K. van der Toorn, Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007); also see my review of van der Toorn in JHS 10/29 (2010), doi:10.5508/jhs.2010.v10.r29. 29

Linguistic Dating, Writing Systems, and the Pentateuchal Sources

353

standing the most difficult hapax legomena in the Hebrew Bible are Akkadian and Ugaritic – that is, linguistic resources dating from the thirteenth through the sixth centuries BCE.32 Cohen notes that there are about thirteen hundred hapax legomena in the Hebrew Bible, but only about four hundred are difficult to interpret. These difficult words derive primarily from biblical texts in SBH, including many words in the Pentateuch. It was the disjunction in the Hebrew scribal tradition in the Babylonian period that created a significant number of problematic words (as well as other linguistic features), because the meaning of many words was lost during the exilic and early postexilic periods. In contrast, the languages prevalent in the Near East during the Persian and Hellenistic periods – namely, Aramaic, Persian, and Greek – play little role in the philological problems related to the difficult hapax legomena. To illustrate, consider a few examples of hapax legomena in Esther, Daniel, Ezra, Nehemiah, and Chronicles – that is, books that have internal claims that date them to the later Persian or Hellenistic periods. In Esther, for example, there is the hapax word ʾōnēs, whose meaning, “to press, force, violate,” is well known from Rabbinic Hebrew. Turning to Ezra–Nehemiah, kənāt in Ezra 4:7 is an Aramaic loanword meaning “companion,” well known in contemporary texts like the Jewish Aramaic texts from Elephantine. The book of Chronicles yields similar examples; thus, in 2 Chr 36:16 the verb lʿb, “to deride, mock,” is well known from later Jewish texts. Examples such as these could be multiplied, but the observation is simply that the hapax legomena from LBH texts are almost invariably understood as loans from Persian or Aramaic or from their use in later Jewish texts. The hapax legomena from the main corpus of SBH literature require a much broader set of linguistic tools to decipher. Chaim Cohen summarized the evidence for twenty-eight difficult hapax legomena using Akkadian and Ugaritic. For example, in Gen 43:11 we have the word boṭnîm, which is a loanword from Akkadian buṭnātu, for “pistachios.” However, the LXX translators no longer understood the preexilic Akkadian loanword word and translated it as τερανδέμινθον, “terebinth tree.” It is worth observing that we do not find a cognate of the Persian word here, which is pista.33 The use of ancient Semitic languages to decipher the meaning of these rare words should be quite striking and telling. Akkadian was essentially unknown outside of scholastic circles by the fifth century BCE, and Ugaritic disappeared as a language around 1200 BCE. The hapax legomena were words whose meanings were presumably understood by ancient Hebrew scribes down through the sixth century, but these meanings were sometimes lost when the scribal tradition of Hebrew suffered a disjunction 32  C. Cohen, Biblical Hapax Legomena in the Light of Akkadian and Ugaritic (Missoula, MT: Scholars Press, 1978). 33  I acknowledge the assistance of my colleague Rahim Shayegan in helping me work through the various Persian cognates in this study. On pista, see J. C heung , Etymological Dictionary of the Iranian Verb (Leiden: Brill, 2007), 292–293.

354

William M. Schniedewind

by the end of the sixth century BCE.34 This is just one particularly illuminating example, but more generally the substantial number of hapax legomena in SBH texts elucidated only by Ugaritic and Akkadian further points to the periodization of Hebrew into SBH and LBH corpora. The absence of difficult hapax legomena in postexilic texts or LBH, on the other hand, indicates the continuity in Hebrew scribal tradition after the fourth century BCE. The philological problems created by linguistic change coupled with a gap in scribal tradition are not limited to hapax legomena. There were also older words and constructions that had gone out of use during the monarchic period but were preserved in the textual record. An interesting example of this is the asseverative lamed, which I wrote about in an article coauthored with Daniel Sivan many years ago.35 We pointed out that there were originally two separate terms in West Semitic: a negative particle /lā/ and an asseverative /lū/. These two particles can be illustrated through Ugaritic as well as the Amarna letters. The Canaanite shift caused the negative /lā/ to be pronounced as /lō/; as a result, this negative particle sounded similar to the positive asseverative particle /lū/. An ʾaleph was appended in Hebrew /lō/, and this created a graphemic distinction between the negative and the asseverative, but the asseverative nevertheless eventually disappeared in Hebrew and other Canaanite dialects. An excellent example of the problems that this disjunction created for later scribes, as well as the ways they attempted to resolve them, may be seen in Gen 23, the story of Abrahamʼs purchase of a burial plot for Sarah, which is usually identified with P. The repeated use of the asseverative lamed in this narrative, employed four times in vv. 5, 11, 13, and 14, has resulted in the variety of readings suggested by the qere-ketiv, the versions, and modern exegetes.36 For instance, vv. 5–6 in the Masoretic Text (MT) read: Genesis 23:5–6 And the Hittites answered Abraham saying to him (lʾmr lw), “Listen to us, my lord, you are a prince of God in our midst, bury your dead in our best burial plot.”

34  Although not a single one of these difficult hapax legomena is found in the corpus of LBH texts (for example, Esther, Daniel, Ezra, Nehemiah, and Chronicles), a few terms do come from the corpus of texts deriving from the Babylonian period, that is, Isa 40–66, Jer 26–52, and Ezekiel. These terms include štʿ, “to be afraid” (Isa 41:10); qbʿt, “cup” (Isa 51:22); zyz, “nipple” (Isa 66:11); ʾšyh, “tower” (Jer 50:15); lbh, “rage” (Ezek 16:30); swgr, “collar” (Ezek 19:9); and brmym, “two-colored fabric” (Ezek 27:24). 35  W. M. S chniedewind and D. Sivan , “Letting Your ‘Yes’ Be ‘No’ in Ancient Israel: A Study of the Asseverative lʾ and hlʾ in Hebrew,” JSS 38 (1993), 209–226. 36  For a comparison of the versions, see S. M. G rill , “Die alten Versionen und die Partikeln loʾ, lô, lû, lî,” BZ 1 (1957), 277–281, here 277–278. Note the emendations proposed by J. Skinner, A Critical and Exegetical Commentary on Genesis (ICC; New York: Scribnerʼs Sons, 1910), 336–338.

Linguistic Dating, Writing Systems, and the Pentateuchal Sources

355

The MT vocalized the asseverative as a prepositional phrase + suffix, lʾmr lw, “saying to him” (also in v. 14), even though this possibility is precluded by the fact that lʾmr, “saying,” invariably introduces a direct quote in Hebrew – it cannot grammatically be followed by lw, “to him.” The earlier meaning would be as follows: “And the Hittites answered Abraham saying, ‘Indeed, Listen to us, my lord [. . .]’ ” This solution is confirmed by comparing vv. 10–11, 13, and 14–15, where this construction is repeated with slight variations resulting from a misunderstanding of the asseverative lamed. When we recognize that this is merely the old asseverative lamed that was no longer understood by postexilic scribes, then the difficulties are resolved. More than that, it indicates a disruption in the scribal tradition that resulted in archaic Hebrew words and phrases that were no longer understood in the postexilic period. Another illustration of the gap in Hebrew scribal tradition is pseudoclassicisms. This phenomenon has been studied in particular by Jan Joosten, who has identified a number of examples that need not be repeated here. 37 Still, it is important to recognize the historical implications of pseudoclassicisms. They again demonstrate that later authors were studying the earlier texts but that the meanings of certain words and expressions from SBH were no longer understood or had evolved. Indeed, SBH was a dead language, even though a vernacular Hebrew continued to be spoken through the Babylonian period and into the Period period. This vernacular would inform the revival of the literary register during the postexilic period, but it did not completely fill the gaps in the scribal understanding of SBH. By way of conclusion, we should recall some of the common explanations for the linguistic character of the Pentateuch. Perhaps the oldest explanation for the SBH language in the pentateuchal sources, particularly the P source, is its supposed “semblance of antiquity.” This is supposedly the result of a learned, artificially archaizing style. This view was made standard by the classic work of Julius Wellhausen, who wrote that P “tries hard to imitate the costume of the Mosaic period and to disguise its own period.”38 Hurvitz, Milgrom, and others have countered that if P were postexilic then there should be some evidence of late language and anachronisms. These arguments, however, have not affected the consensus of scholars who are committed to seeing a significant postexilic component to the writing, redacting, and editing of the Priestly source. John Collins, for example, writes, “All of this shows that the language of P was not 37  J. J oosten , “The Evolution of Literary Hebrew in Biblical Times: The Evidence of Pseudo-Classicisms,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Linguistic Studies in Ancient West Semitic 8; Winona Lake, IN: Eisenbrauns, 2012), 281–292. 38  J. Wellhausen, Prolegomena to the History of Ancient Israel (trans. J. Sutherland Black and Allan Enzies; New York: Meridian Books, 1957), 9. Also see, for example, F. M. Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), 293–325.

356

William M. Schniedewind

invented in the exilic or postexilic period. But Priestly, liturgical language is often archaic, and terminology is often preserved in ritual contexts long after it has fallen out of use in popular speech.”39 Yet, the linguistic problem is not that the Priestly source or the Pentateuch has archaic elements but that it has very few late elements (e.g., Aramaisms, Persian loans, Mishnaisms). In contrast, the Hebrew version of the book of Jubilees and the Temple Scroll (and the Dead Sea Scrolls more generally) also have “archaic” elements, but they are mixed with all the signs of later Hebrew. Ironically, Collinsʼs argument underscores the fact that our overall assessment of the linguistic character of the Priestly source (as well as the Pentateuch as whole) still has not changed, but it has to be explained. There have been several explanations proffered; the main ones include the following: (1) the Pentateuch preserves SBH, (2) the Pentateuch is archaizing, or (3) SBH is just a dialect with no diachronic location. None of these explanations are particularly convincing given the overall historical context presented in this paper. Linguistic dating remains a crux for the pentateuchal sources. To be sure, this historical overview of the linguistic context of SBH and LBH just gives a general framework for the linguistic dating of the Pentateuch and its sources. Since linguistic dating requires an accumulation of features, it cannot be as easily applied to editorial work that shaped the final form of the biblical books. At the same time, claims of substantive SBH writing in the late Persian and Hellenistic periods are not supported by what we know from historical sociolinguistics or from the evidence of texts like Chronicles, Ezra–Nehemiah, Esther, Daniel, Jubilees, 1QIsaiah, or other Qumran texts. Indeed, there is a studied quality to LBH or Qumran Hebrew, but there is also always a diachronic dimension. After a disruption in the Hebrew scribal tradition, priestly scribes in the late Persian and Hellenistic periods revived the national language and literature using their Aramaic scribal training and an unbroken tradition of vernacular Hebrew.

39

 J. J. Collins, Introduction to the Hebrew Bible (Minneapolis: Fortress, 2004), 175.

How to Date Pentateuchal Texts Some Case Studies Thomas Römer What Facts Do We Have? If you asked the date of the Hebrew Bible in its present Masoretic form, the answer would be quite simple. The Hebrew Bible as we have it was edited in the Middle Ages – the St. Petersburg Codex in the eleventh century, the mutilated Aleppo Codex somewhat earlier, and the textus receptus some centuries later. This does not answer the question, of course, but shows us that we start with a very late form of the Biblical text when we work with the BHS or BHQ. Thanks to the discovery of the Qumran fragments, we have factual evidence for the existence of almost all biblical books in or around the second or first century BCE, although most of them are attested only in a very fragmentary form. The divergences among manuscripts of the same book – for instance, Jeremiah or Isaiah – also make it clear that these books could not have been written for the first time in this period and must be older. But how much older? One can try, for the Torah/Pentateuch, to go one step backward and refer to the Greek translation. According to the Letter of Aristeas, written about 150–100 BCE, the Greek translation of the Torah took place under Ptolemy II in Alexandria in the third century BCE. Although the story is legendary, it is quite plausible that there were Greek translations of the Pentateuch beginning in the third century. This is confirmed in the work of Demetrius the Chronographer (221–204), who comments on difficult texts of the Torah by apparently using a Greek translation. Aristobulos of Paneas (around 160 BCE) also attests the idea that the Pentateuch was translated in Greek under Ptolemy Philadelphus (fragment 3).1 In contrast, the fragments of Hecataeus of Abderaʼs Aegyptiaca (commonly dated around 320 BCE) show that the author knew traditions about Moses – for example, that he led the Hebrews out of Egypt and into the land, built Jerusalem, and issued laws – some of which are quite different from what can be found in the Pentateuch.2 1  For an English translation, see C. R. Holladay (ed.), Aristobulus, vol. 3 of Fragments from Hellenistic Jewish Authors (Atlanta: Scholars Press, 1995). 2  For more details, see C. Z amagni , “La tradition sur Moïse d’‘Hécatée dʼAbdère’ dʼaprès Diodore et Photius,” in Interprétations de Moïse: Égypte, Judée, Grèce et Rome

358

Thomas Römer

It appears reasonable, therefore, to date the beginning of the Greek translation of the Torah, which took place over several decades, around 270 BCE. 3 This presupposes an authoritative Hebrew Pentateuch at least at the end of the fourth century. After this date, we are on quite secure ground. Before this date, however, things become more complicated.

A Short Overview of the History of Research The traditional way to date the Torah was to start with the terminus a quo and, since the last event narrated is the death of Moses, the rabbis, who claimed that Moses was the author of the Torah, admitted that it was completed by Joshua, who added the last verses of the Pentateuch. Of course, they were not interested in a precise historical date of the Torah. Their aim was to affirm the Mosaic authority of the Law, but they also admitted some diachronic or logical considerations. One may also mention the “postmosaica” identified by Abraham Ibn Ezra (1089–1164), who pointed discretely to verses in the Pentateuch that presuppose a time much later than the time of Moses. In the seventeenth century, Spinoza also used the terminus a quo argument, although in a different way. He pointed out the existence of an Enneateuch by arguing that the books from Genesis to Kings “were written by a single historian, who wished to relate the antiquities of the Jews from their first beginning down to the first destruction of the city.”4 Thus, the Pentateuch cannot have been written before the destruction of Jerusalem and the Babylonian exile, and the best candidate for the redaction of these books Spinoza can think of is Ezra. Spinozaʼs emphasis on a single historian responsible for the Enneateuch was given up when stylistic and theological differences or contradictions in the literature came to the forefront. When de Wette wrote his Dissertatio Critica on the book of Deuteronomy in 1805, he was mainly interested in demonstrating how the book of Deuteronomy is different from the books of Genesis to Numbers.5 Since Lev 26 constitutes a clear conclusion to the Mosaic Law, and since Deut 28 (ed. P. Borgeaud et al.; Jerusalem Studies in Religion and Culture 10; Leiden: Brill, 2010), 133–169. 3  See E. Tov, Textual Criticism of the Hebrew Bible (3rd ed.; Minneapolis: Fortress, 2012), 131; M. Tilly, Einführung in die Septuaginta (Darmstadt: Wissenschaftliche Buchgesellschaft, 2005), 26–37; T. H. Lim, The Formation of the Jewish Canon (AYBRL; New Haven, CT: Yale University Press, 2013), 74–88. 4  B. de S pinoza , A Theologico-Political Treatise and a Political Treatise (trans. R. H. M. Elwes; New York: Dover Publications, 1951; repr., 2004), 128. 5  W. M. L. de Wette, Opuscula theologica (Berlin: G. Reimer, 1830), 149–168. The Latin original has been translated into German and is easily accessible in H.-P. Mathys, “Wilhelm Martin Leberecht de Wettes Dissertatio critico-exegetica von 1805,” in Biblische Theologie und historisches Denken, Wissenschaftsgeschichtliche Studien: Aus Anlass der 50. Wiederkehr

How to Date Pentateuchal Texts

359

repeats this chapter in a different style, de Wette draws the conclusion that Deuteronomy stems from a different author. The fact that Deuteronomy also contains parallels to the laws and narratives of the preceding books shows that its author was familiar with those traditions and that his work must therefore date from a later period. De Wette further observes that the main focus of the Deuteronomic law lies on the idea of a centralization of the cult, which, according to him, can refer only to the Jerusalem temple. He therefore suggests in a lengthy footnote that (the first edition of) Deuteronomy is the book of Josiahʼs reform. Though the identification of Deuteronomy with Josiahʼs law book was nothing new, the possibility that Deuteronomy could stem from the time of Josiah was, and it would allow a precise date for the book of Deuteronomy. With this claim, de Wette provided (as O. Eissfeldt put it) “Pentateuchal criticism with a ‘point of Archimedesʼ to which it could attach itself in order to deliver it from the bonds of church and synagogue tradition, and put in its place an alternative dating of the Pentateuch.”6 In his Beiträge zur Einleitung in das Alte Testament, de Wette demonstrates that the narrative of 2 Kgs 22 itself proves that the book was formerly unknown and probably hidden in the temple by the priest Hilkiah, 7 thus supporting the idea of a seventh-century BCE origin of the first edition of Deuteronomy. The seventh-century date of Deuteronomy then allowed dating the older sources in the books of Genesis to Numbers (J/E) between the tenth and eighth centuries. Since the 1830s, several scholars (Vatke, Popper, Reuss, et al.) observed that the Priestly law or parts of it was later than Deuteronomy and the older sources and unknown by the preexilic prophets.8 Graf built on these observations by claiming that the Priestly laws had been inserted into the Hexateuch only in the Persian period.9 De Wette thus provided the foundation for the late date of the P source on which the elaboration of the Documentary Hypothesis by Kuenen and Wellhau-

der Basler Promotion von Rudolf Smend (ed. M. Kessler and M. Walraff; Studien zur Geschichte der Wissenschaften in Basel, Neue Folge 5; Basel: Schwabe, 2008), 171–211. 6  O. Eissfeldt, The Old Testament: An Introduction (Oxford: Basil Blackwell, 1965), 171. 7  W. M. L. de Wette, Beiträge zur Einleitung in das Alte Testament (2 vols.; Halle: Schimmelpfennig und Compagnie, 1806), 1:170. But here he remains cautious: “Doch bin ich weit davon entfernt, diese Vermuthung zur Gewißheit erheben zu wollen. [. . .] Woher das Buch gekommen sey, darüber zieht die Geschichte ihren Vorhang und es würde vermessen seyn, ihn wegziehen zu wollen.” His doctoral dissertation, recently edited anew, puts it more assertively; see Mathys, “Wilhelm Martin Leberecht de Wette” (see n. 5). 8  For details, see T. R ömer , “ ‘Higher Criticism’: The Historical and Literary-Critical Approach – with Special Reference to the Pentateuch,” in Hebrew Bible/Old Testament, The History of Its Interpretation (ed. M. Sæbø; Göttingen: Vandenhoeck & Ruprecht, 2013), 3.1:393–423. 9  K. H. G raf , “Die sogenannte Grundschrift des Pentateuch,” Archiv für die wissen­ schaftliche Erforschung des Alten Testaments 1 (1869), 466–477.

360

Thomas Römer

sen was built.10 (Note, however, that the Kaufmann school did and still does not accept the late date of P.11) In the context of the Documentary Hypothesis, dating the Pentateuch thus became an easy – and circular – task. Once a passage was attributed to J, E, D, or P, its date was secured by the global theory underlying the sources of the Documentary Hypothesis. But the Wellhausenian framing of the Documentary Hypothesis came under increased criticism beginning in the 1970s, and it is no longer possible to take it for granted, at least in Europe.12 So the question of dating becomes crucial again. It is indeed fascinating to see how a passage like Gen 12:1–4a, used by G. von Rad and H. W. Wolff as the summary of the kerygma of a Yahwist dated to the reign of Solomon,13 can now be dated to the Persian period, as it is, for instance, in a fine article by J. L. Ska, who considers the passage to be postexilic, as it mixes Dtr and Priestly style and themes in order to integrate the Abraham traditions into the Pentateuch.14 The “democratization” of royal ideology and its application to Abraham (see the parallels in 2 Sam 7:9 and Ps 72:17), who is in several regards constructed as a predecessor not only of Moses but also of David, is possible only after the end of the Judean monarchy.15 I find myself sympathetic to this option, but I also recognize that our dating of pentateuchal texts is intimately linked to a global view about the formation of the Torah. So, before I address some possibilities for dating those texts, I would like to briefly address some ideological questions.  J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (1899) (Berlin: de Gruyter, 1963); A. Kuenen, A Historical-Critical Inquiry into the Origin and Composition of the Hexateuch (London: Macmillan, 1886). 11  Y. Kaufmann, The Religion of Israel: From Its Beginnings to the Babylonian Exile (Chicago: University of Chicago Press, 1960). Among his most famous disciples, see especially J. Milgrom, “The Antiquity of the Priestly Source: A Reply to Joseph Blenkinsopp,” ZAW 111 (1999), 10–22, and A. Hurvitz, “Once Again: The Linguistic Profile of the Priestly Material in the Pentateuch and Its Historical Age: A Response to J. Blenkinsopp,” ZAW 112 (2000), 180–191. For a critical reassessment, see S. Meyer, “Dating the Priestly Text in the Pre-Exilic Period: Some Remarks about Anachronistic Slips and Other Obstacles,” Verbum et Ecclesia 31 (2010), 6, http://www.ve.org.za/index.php/VE/article/view/423/475 (accessed 02/03/2015). 12  For an overview of developments in European pentateuchal research, see T. B. Dozeman and K. Schmid (eds.), A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (SBL Symposium Series 34; Atlanta: Society of Biblical Literature, 2006), and T. Römer, “Zwischen Urkunden, Fragmenten und Ergänzungen: Zum Stand der Pentateuchforschung,” ZAW 125 (2013), 2–24. 13  G. von R ad , The Problem of the Hexateuch and Other Essays (trans. E. A. Trueman; London: SCM, 1984; German orig., 1938), 1–78, esp. 71–73; H. W. Wolff, “The Kerygma of the Yahwist,” in The Vitality of Old Testament Tradition (ed. W. Brueggemann and H. W. Wolff; Atlanta: John Knox, 1975; German orig., 1964), 41–82, 56. 14  J. L. S ka , The Exegesis of the Pentateuch (FAT 66; Tübingen: Mohr Siebeck, 2009; French orig., 1997), 46–66. 15  M. Köckert, Vätergott und Väterverheißungen: Eine Auseinandersetzung mit Albrecht Alt und seinen Erben (FRLANT 142; Göttingen: Vandenhoeck & Ruprecht, 1988), 276–299. 10

How to Date Pentateuchal Texts

361

Why Do We Need to Date the Pentateuch? Contrary to the books of Kings, where many foreign rulers are named and historical details (coinciding with extrabiblical documents or not) are given, the Pentateuch seems not to be very keen on delivering such historiographical details. The pharaoh of the exodus, often claimed to be Ramses II, bears no name in the exodus story; the same holds true for the Egyptian king in the Abraham and Joseph stories. And if the foreign monarch bears a name like Abimelek, he cannot easily be identified with a historical Philistine ruler, although perhaps there is a link between Abimelek and Abdī-Milki, who is mentioned in an inscription of Esarhaddon from 673/672 as king of Ashdod. An identification of Abimelek with this king would give a terminus a quo in the seventh century for Gen 20 and 26.16 Most pentateuchal narratives seem to construct types or figures through which they want to describe YHWHʼs intervention in favor of his people. In Exod 1–15, Pharaoh represents Egypt, its gods, and its power, which will be defeated by the God of Israel.17 The interest in the historicity of the pentateuchal narratives and their date arose with the beginning of historical-critical scholarship of the Hebrew Bible, when historicity became a criterion for the truth of the Bible and when scholars believed that the older a text was, the more credible and valuable. Even today, many biblical scholars come from a religious background and have conscious or subconscious, hidden or open, theological agendas that may depend on certain views about the age of pentateuchal texts or traditions. We should be aware of these when we ask ourselves why it is so important to have “old” pentateuchal texts that come as close as possible to supposedly historical events. Of course, the question of the date of a text belongs to critical scholarship and should not be neglected because it helps to understand the formation of the Bible in its historical contexts. But we should not transform the question of dating into a question of faith.

Linguistic Evidence? Can the question of the date of pentateuchal or other biblical texts be resolved by the distinction between Classical Biblical Hebrew and Late Biblical Hebrew,  I. Finkelstein and T. Römer, “Comments on the Historical Background of the Abraham Narrative: Between ‘Realia’ and ‘Exegetica,’ ” HBAI 3, no. 1 (2014), 3–23, here 13. 17  There are, of course, some learned historical comments, as for instance in Num 13:22: “Hebron had been built seven years before Tanis in Egypt.” As B. A. Levine, Numbers 1–20 (AB 4; New York: Doubleday 1993), 354, points out, this information should not be taken in a historical sense: “This statement merely reflects the image of Tanis held in the first millennium B.C.E., when it was thought that this town in the northern delta had served as a capital city during the Ramesside period.” 16

362

Thomas Römer

which would give some allegedly objective criteria for the dating of biblical texts? It is not astonishing that this method has found a significant number of supporters, especially in North America and in Israel; I will not go into details here, as other papers will deal more extensively with this question.18 I would only like to raise some cautions about use of this tool. First of all, we must ask ourselves, with E. Ullendorf and E. A. Knauf, whether Biblical Hebrew was a spoken language.19 The evidence for so-called Classical Hebrew20 outside the Bible is limited to a few inscriptions and personal names that do not allow us to claim a unified Classical Hebrew in the monarchic period. We must admit dialectical differences in the spoken and written extrabiblical texts and, more important, differences between literary and spoken forms. Second, there is no doubt that some late books like Qoheleth differ from so-called Classical Biblical Hebrew, but texts that may well be as late as Qoheleth – like Zech 1–8 and the extracanonical Ps 151 – are written in perfect Classical Biblical Hebrew.21 Finally, it is also very difficult, if not impossible, to fix a clear line of demarcation between “Classical” and “Late” Biblical Hebrew. As C. Edenburg has recently observed, undisputed late (Persian period) texts share a preference for object suffixes directly affixed to verbs with Hebrew/Moabite inscriptions from the Iron Age.22 That means that we cannot claim a straightforward evolution.23 “Biblical Hebrew” is above all a literary language that outlived its spoken life (if it had any) and persisted in the scribal milieu. The distinction between Classical  See now on this topic also the very important comments of C. Edenburg, Dismembering the Whole: Composition and Purpose of Judges 19–21 (AIL: Atlanta: Society of Biblical Literature, 2016), 115–123. 19  E. U llendorf , Is Biblical Hebrew a Language? Studies in Semitic Languages and Civilizations (Wiesbaden: Harrassowitz, 1977), 3–17; E. A. Knauf, “War Biblisch-Hebräisch eine Sprache? Empirische Gesichtspunkte zur linguistischen Annäherung an die Sprache der althebräischen Literatur (1990),” in Data and Debates: Essays in the History and Culture of Israel and Its Neighbors in Antiquity (ed. K. Schmid et al.; AOAT 407; Münster: Ugarit-Verlag, 2013), 411–423, esp. 421: “Biblisch-Hebräisch war keine Sprache, die jemals von jemandem gesprochen worden wäre.” 20  Knauf, “War Biblisch-Hebräisch eine Sprache?” (see n. 19), even challenges the plausibility of such a concept. 21  I. Y oung , “What Is ‘Late Biblical Hebrew’?,” in A Palimpsest: Rhetoric, Ideology, Stylistics, and Language Relating to Persian Israel (ed. E. Ben Zvi et al.; PHSC 5; Piscataway: Gorgias Press, 2009), 253–268, here 258–259. 22  Edenburg, Dismembering (see n. 18), 120–121. 23  See I. Young, “What Do We Actually Know about Ancient Hebrew,” Australian Journal of Jewish Studies 27 (2013), 11–31, who challenges the so-called Three Stages Theory (Classical Hebrew, Late Biblical Hebrew, Mishnaic Hebrew). See also the different appreciation of the narrative parts of the book of Job by A. Hurvitz, “The Date of the Prose Tale of Job Linguistically Reconsidered,” HTR 67 (1974), 17–34, and I. Young, “Is the Prose Tale of Job in Late Biblical Hebrew?,” VT 59 (2009), 606–629. 18

How to Date Pentateuchal Texts

363

and Late Biblical Hebrew, especially when applied to a whole book, does not deal with the widely acknowledged fact that the text is the product of a long process of redaction and revision. Thus it appears that scribes can preserve or even partially invent a language that has not been spoken for many centuries. One should therefore be cautious about claiming that the whole Pentateuch was written before the exile because it is mostly written in Classical Biblical Hebrew.24 So, what other possibilities do we have for dating pentateuchal texts? I will start with what I consider to be weaker methods and continue with some more trustworthy ones.

Allegorical Dating This method is quite commonly used, even though it is never labeled this way. The promise to Abraham in Gen 12:1–4a, mentioned above, was dated by von Rad to the Solomonic period because, so went his argument, the promise of a great nation was fulfilled under his reign. But this argument works only with the assumption that J was written under Solomon. More recently, T. Veijola and K. Schmid have dated (probably rightly) the story of the Akedah in Gen 22 to the early Persian period, arguing that the divine order according to which Abraham should sacrifice his son should be read as reflecting the fear of the early Persian community in Judea about whether there would be any future (descendant) for Israel.25 Again this aspect is not obvious, albeit not excluded. Genesis 22 is first and foremost a divine testing of the patriarch, who passes the exam. The theological interpretation of this text should not, in my view, be used to date it. Another interesting case is the golden-calf episode in Exod 32 (// Deut 9), where allegorical dating is combined with intertextual arguments. Exodus 32:4 and 1 Kgs 12:28 are clearly related to one another; the question is how to use this parallel for dating the Exodus (or the Kings) account.26

 M. E hrensvärd , “Once Again: The Problem of Dating Biblical Hebrew,” SJOT 11 (1997), 29–40. 25  T. Veijola, “Das Opfer des Abraham – Paradigma des Glaubens aus dem nachexilischen Zeitalter,” ZTK 85 (1988), 129–164; K. Schmid , “Die Rückgabe der Verheißungsgabe: Der ‘heilsgeschichtliche’ Sinn von Gen 22 im Horizont innerbiblischer Exegese,” in Gott und Mensch im Dialog: Festschrift für Otto Kaiser zum 80. Geburtstag (ed. M. Witte; Berlin: de Gruyter, 2004), 271–300. 26  For a history of research until 2000, see K. S chmid , “Israel am Sinai: Etappen der Forschungsgeschichte zu Ex 32–34 in seinen Kontexten,” in Gottes Volk am Sinai: Untersuchungen zu Ex 32–34 und Dtn 9–10 (ed. M. Köckert and E. Blum; Veröffentlichungen der Wissenschaftlichen Gesellschaft für Theologie 18; Gütersloh: Kaiser, 2001), 9–39. 24

364 Exodus 32:4 1 Kings 12:28

Thomas Römer

‫ֹלהיָך יִ ְש ָר ֵאל ֲא ֶשר ֶה ֱעלּוָך ֵמ ֶא ֶרץ ִמ ְצ ָריִ ם‬ ֶ ‫ֵא ֶלה ֱא‬ ‫ֹלהיָך יִ ְש ָר ֵאל ֲא ֶשר ֶה ֱעלּוָך ֵמ ֶא ֶרץ ִמ ְצ ָריִ ם‬ ֶ ‫ִהנֵ ה ֱא‬

Was Exod 32 (in its original form) written during the existence of the Northern Kingdom in order to criticize the northern sanctuaries of Bethel and Dan?27 Does Exod 32 presuppose the fall of the Northern Kingdom and the reform of Josiah (see the parallels between Mosesʼs destruction of the calf in Exod 32 and Josiahʼs reform in 2 Kgs 23)?28 Or is Exod 32 an allegory for the fall of Judah?29 Taken alone, allegorical historical settings of pentateuchal narratives are too speculative to convince a majority of scholars. This approach can be used only as a supplement to other arguments.

Arguments from Silence This method, which is also used in other fields of classical studies, is based on the assumption that a tradition must be new if it is not mentioned in older texts. H. Vorländerʼs work about the formation of the “Yehowistic history” was based mainly on the argument that J/E needs to be dated to the exilic period because its traditions are not attested in the preexilic texts of the Former and Latter Prophets.30 This method is problematic not only because it is difficult to date many prophetic texts but also because of its reliance on quotation of or allusions to pentateuchal traditions outside the Torah. That said, an inquiry about the presence or absence of pentateuchal traditions outside the Pentateuch can sometimes be helpful, as is the case for references to Abraham compared to the much more frequent references to Jacob. The datable texts that mention Abraham outside the Torah probably belong to the Babylonian or early Persian period (Ezek 33:24).31 27  G. N. K noppers , “Aaronʼs Calf and Jeroboamʼs Calves,” in Fortunate the Eyes That See: Essays in Honor of David Noel Freedman in Celebration of His Seventieth Birthday (ed. A. B. Beck et al.; Grand Rapids, MI: Eerdmans, 1995), 92–104. 28  T. R ömer , “Le jugement de Dieu et la chute dʼIsraël selon Exode 32,” Foi et Vie 91 (Cahier biblique 31) (1992), 3–14. 29  T. B. D ozeman , Exodus (ECC; Grand Rapids, MI: Eerdmans, 2009), 700. See also T. B. Dozeman , “The Composition of Ex 32 within the Context of the Enneateuch,” in Auf dem Weg zur Endgestalt von Genesis bis II Regum: Festschrift für Hans-Christoph Schmitt zu seinem 65. Geburtstag (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 175–189. 30  H. Vorländer, Die Entstehungszeit des jehowistischen Geschichtswerkes (Europäische Hochschulschriften: Theologie 32/109; Bern: Lang, 1978). 31  For more details, see T. Römer, “Abraham Traditions in the Hebrew Bible outside the Book of Genesis,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; Leiden: Brill, 2012), 159–180.

How to Date Pentateuchal Texts

365

This could foster the theory, quite common now in German scholarship, that the literary formation of the Abraham tradition did not start much earlier than the sixth century BCE. A more intriguing case is the figure of Joseph. Even if the name Joseph occurs in several prophetic books as a reference to the Northern Kingdom, no clear allusions to the Joseph story (Gen 37–50) are to be found outside the Hexateuch, besides Ps 105, a text that is commonly considered to presuppose the whole Torah.32 Combined with observations made by the Egyptologist Donald Redford, who noted that allusions to Egyptian customs and names fit best in the Saite and later periods,33 this would support a late date for the Joseph story, which, in combination with an “allegorical” interpretation, could be understood as a diaspora novella written (in Egypt?) during the Persian period.34

Terminus a quo and terminus ad quem We have already mentioned that the terminus a quo criterion is not very useful for the dating of the Pentateuch. But the criterion must also be handled with care when applied to other texts. As is well-known, M. Noth used 2 Kgs 25:27–30 to date the so-called Deuteronomistic History around 560 BCE. Interestingly, Noth here arrived at this identification of a terminus a quo and terminus ad quem based on the idea that the Deuteronomist (Dtr) was an “honest broker” who transmitted all the information and sources that were available to him35 – in other words, Noth assumed that if Dtr had known of events from the Persian 32  See, recently, S. Ramond, Les leçons et les énigmes du passé: Une exégèse intra-biblique des psaumes historiques (BZAW 459; Berlin: de Gruyter, 2014), 154–159. 33  D. B. R edford , A Study of the Biblical Story of Joseph (Genesis 37–50) (VTSup 20; Leiden: Brill, 1970). 34  This date for the Joseph story is becoming increasingly common in European scholarship; see T. R ömer , “Joseph approché: Source du cycle, corpus, unité,” in Le livre de traverse: De lʼexégèse biblique à lʼanthropologie (ed. O. Abel and F. Smyth; Patrimoines; Paris: Cerf, 1992), 73–85; A. Catastini, Storia di Guiseppe (Genesi 37–50) (Venice: Marsilio, 1994); J.-M. Husser, “Lʼhistoire de Joseph,” in La Bible et sa culture: Ancien Testament (ed. M. Quesnel and P. Gruson; Paris: Desclée, 2000), 112–122; C. Uehlinger, “Fratrie, filiations et paternités dans lʼhistoire de Joseph (Genèse 37–50*),” in Jacob: Commentaire à plusieurs voix de Gen. 25–36: Mélanges offerts à Albert de Pury (ed. J.-D. Macchi and T. Römer; MdB 44; Genève: Labor et Fides, 2001), 303–328; M. F ieger and S. H odel -H oenes , Der Einzug in Ägypten: Ein Beitrag zur alttestamentlichen Josefsgeschichte (Das Alte Testament im Dialog 1; Bern: Lang, 2007), 375–376; J. Ebach, Genesis 37–50 (HThKAT; Freiburg im Breisgau: Herder, 2007), 693; D. Nocquet, “LʼÉgypte, une autre terre de salut? Une lecture de Gn 45,1–46,7,” ETR 84 (2009), 461–480; H. C. P. Kim, “Reading the Joseph Story (Genesis 37–50) as a Diaspora Narrative,” CBQ 75 (2013), 219–238. 35  M. N oth , Überlieferungsgeschichtliche Studien: Die sammelnden und bearbeitenden Geschichtswerke im Alten Testament (Darmstadt: Wissenschaftliche Buchgesellschaft, 1967;

366

Thomas Römer

period he would have included them. But, as Graeme Auld has half-mockingly stated: “The fact that Kings ends with the fate of Judahʼs last king tells us no more about the date of composition (generally believed to be exilic) than the fact that the Pentateuch ends with the death of Moses.” 36 Despite the exilic perspective of Deuteronomy–Kings, there is evidence that the Deuteronomistic History underwent one or several redactions in the Persian period with which I cannot deal here. There are, however, some cases in the Torah where a terminus a quo argument can be interesting. The expression ‫ אור כשדים‬is one such case. It occurs in the Hebrew Bible only in Gen 11:28, 31 (P); Gen 15:7; and Neh 9:7 (which apparently quotes Gen 15:7). The expression ‫ כשדים‬is widely attested, especially in the books of Kings and Jeremiah, where it refers in all cases to the Neo-Babylonians. The expression thus cannot have arisen before the end of the seventh century BCE, and a sixth century (or later) date of the Genesis texts is plausible. If Gen 11:28, 31 belongs to the same literary layer as other P texts, this observation has consequences for the terminus a quo of the Priestly writings of the Pentateuch. A similar approach can be made for the name ‫יון‬, which occurs in the Pentateuch in the Priestly table of nations in Gen 10:2, 4 and which may be related to the name Ionian. In the Hebrew Bible it is used in Isa 66:19; Ezek 27:13, 19; Zech 9:13; and Dan 10:20 and 11:2 to designate Greek populations either in Asia Minor or in Greece. All these texts are no earlier than the Persian period, which would then locate P in Gen 10 in the same context.37 The name Ionia (KUR. ia-man) does occur in an inscription of Esarhaddon (60, 10′; RINAP 4) and perhaps under Sennacherib (20, i′ 7, RINAP 3/1, although the name is restored here) and possibly in other texts from the Neo-Assyrian period, which suggests the possibility of an earlier terminus a quo.38 But in the context of the Hebrew Bible, a sixth-century BCE terminus a quo seems more plausible. 1st ed., 1943); English translation, idem, The Deuteronomistic History (trans. J. Doull et al.; 2nd ed.; JSOTSup 15; Sheffield: Sheffield Academic, 1991), 26, 128. 36  A. G. Auld, Samuel at the Threshold: Selected Works of Graeme Auld (Society of Old Testament Studies Monographs; Burlington: Ashgate, 2004), 45–61, esp. 61. 37  Such a date for Gen 10 is proposed by A. de Pury, “Sem, Cham et Japhet: De la fraternité à lʼesclavage,” in κορυφαίῳ ἀνδρί: Mélanges offerts à André Hurst (ed. A. Kolde et al.; Recherches et rencontres; Geneva: Droz, 2005), 495–508, who considers the non-P texts in this chapter to be post-Priestly. 38  C. T uplin , Achaemenid Studies (Historia Einzelschriften 99; Stuttgart: Steiner, 1996), 27. There may be other texts referring to the Ionians; see A. M. Bagg, Die Orts- und Gewässernamen der neuassyrischen Zeit, Teil 1: Die Levante (RGTC 7/1; Wiesbaden: L. Reichert, 2007), 123–124. For the historical context, see R. R ollinger , “The Ancient Greeks and the Impact of the Ancient Near East Textual Evidence and Historical Perspective (ca. 750–650 BC),” in Mythology and Mythologies: Methodological Approaches to Intercultural Influence (ed. R. M. Whiting; Melammu Symposia II; Helsinki: Neo-Assyrian Text Corpus Project, 2001), 233–264.

How to Date Pentateuchal Texts

367

The terminus ad quem for all major pentateuchal texts should be somewhere around 350–300, which does not mean that revisions were impossible from that time on, as shown by the variety of manuscripts in Qumran and the differences in the chronological framework (life spans, etc.) between the MT and the LXX.

Dating by External Comparisons Several texts of the Pentateuch relate to extrabiblical texts or names. The most obvious case is the Priestly and non-Priestly accounts of the flood, which have close parallels with the Mesopotamian flood narratives, the closest parallel being tablet 11 of the standard version of the Gilgamesh Epic from the library of Assurbanipal. This version is considered to be a copy of an older version. The question is when Judean scribes got to know this or a similar account. Were such accounts available when Israel and Judah were under Assyrian domination? Or would it be more plausible to think that Judean scribes conceived a Yahwistic version of the flood during the Babylonian exile?39 The original story of the birth of Ishmael could also be dated on the basis of external parallels if one is willing to follow Knauf, according to whom the name Ishmael reflects a federation of Arabic tribes attested during the seventh century in Assyrian documents.40 The original account of Gen 16 could therefore have arisen in the seventh century as an etiology of these tribes and an attempt to relate them to Abraham. The birth story of Moses in Exod 2:1–10 has a close parallel in the birth legend of Sargon,41 copies of which are attested from the eighth century.42 It is therefore plausible that the birth story of Moses was written in the seventh 39  See the discussion in J. C. G ertz , “Beobachtungen zum literarischen Charakter und zum geistesgeschichtlichen Ort der nichtpriesterlichen Sintfluterzählung,” in Auf dem Weg zur End­gestalt von Genesis bis II Regum: Festschrift für Hans-Christoph Schmitt zu seinem 65. Geburts­tag (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 41–57; T. R ömer , “The Creation of Humans and Their Multiplication: A Comparative Reading of Athra-Hasis, Gilgamesh XI and Genesis 1, 6–9,” Indian Theological Studies 50 (2013), 123–131. 40  E. A. Knauf, Ismael: Untersuchungen zur Geschichte Palästinas und Nordarabiens im 1. Jahrtausend v.Chr. (2nd ed.; ADPV; Wiesbaden: Harrassowitz, 1989), 1–16, 25–55; see also F inkelstein and R ömer , “Comments” (see n. 16), 13–14, and recently E. A. K nauf , “Ishmael, I: Hebrew Bible/Old Testament,” EBR 13 (2016), cols. 352–355, 353–354. 41  M. Gerhards, Die Aussetzungsgeschichte des Mose: Literar- und traditions­geschicht­ liche Untersuchungen zu einem Schlüsseltext des nichtpriesterlichen Tetrateuch (WMANT 109; Neukirchen-Vluyn: Neukirchener Verlag, 2006), 149–259. 42  B. Lewis, The Sargon Legend: A Study of the Akkadian Text of the Tale and the Tale of the Hero Who Was Exposed at Birth (ASOR Dissertation Series 4; Cambridge: American Schools of Oriental Research, 1980).

368

Thomas Römer

century (under Josiah?)43 in order to demonstrate that Moses was as important as the founder of the Assyrian dynasty. Another interesting case is the Priestly blessing of Num 6:22–26, which has an extrabiblical parallel in the Ketef Hinnom amulets. The amulets have been dated to the eighth or seventh century BCE,44 a date used by some scholars to claim a preexilic date for the Priestly Code, if Num 6 belongs to a “P-document.”45 This argument overlooks the possibility that the blessing was an originally independent poetic piece, which could well have been inserted into a more recent book of Numbers. In any case, consensus about the date of the amulets has dissolved.46 The first version of the book of Deuteronomy remains a centerpiece of efforts to date based on external comparisons. It has often been observed that the book bears the language and ideology of the Neo-Assyrian vassal treaties – or, better, loyalty oaths. Interestingly, the Neo-Assyrian documents use the Numeruswechsel, which occurs frequently in Deuteronomy. H.-U. Steymans has pushed these comparisons further and argued that the oldest edition of Deuteronomy had a specific Vorlage: the loyalty oath (adê) of Esarhaddon that was written in order to guarantee the succession of his son Assurbanipal. The parallels between the exhortations of loyalty and Deut 6:4–9* as well as Deut 13* and the curses of the adê and Deut 28 are too close to be coincidental, and the best solution may indeed be that the author of Deut 6; 12–18*; and 28* did use the Neo-Assyrian text, which can be dated quite precisely to 672.47 This would confirm de Wetteʼs impulse  E. Otto, Das Gesetz des Mose (Darmstadt: Wissenschaftliche Buchgesellschaft, 2007), 182–185. Sometimes scholars emphasize the parallels between Exod 2 and the birth legend of Cyrus, transmitted by Herodotus, e.g., H. Z lotnick-Sivan, “Moses the Persian? Exodus 2, the ‘Other’ and Biblical ‘Mnemohistory,’ ” ZAW 116 (2004), 189–205, but these parallels are less strong. 44  G. B arkay , “The Priestly Benediction on Silver Plaques from Ketef Hinnom in Jerusalem,” TA 19 (1992), 139–192, and G. Barkay et al., “The Amulets from Ketef Hinnom: A New Edition and Evaluation,” BASOR 334 (2004), 41–71. 45  E. Waaler , “A Revised Date for Pentateuchal Texts? The Evidence from Ketef Hinnom,” TB 53 (2002), 29–55; see also the presentation in J. D. Smoak, The Priestly Blessing in Inscription and Scripture: The Early History of Numbers 6:24–26 (New York: Oxford University Press, 2015), 61–88. 46  For a sixth- or fifth-century date, see, e.g., A. Berlejung, “Ein Programm fürs Leben: Theologisches Wort und anthropologischer Ort der Silberamulette von Ketef Hinnom,” ZAW 120 (2008), 204–230, and N. Na‍ʾaman, “A New Appraisal of the Silver Amulets from Ketef Hinnom,” IEJ 61 (2011), 184–195. 47  H. U. Steymans, Deuteronomium 28 und die adê zur Thronfolgeregelung Asarhaddons: Segen und Fluch im Alten Orient und in Israel (OBO 145; Göttingen: Vandenhoeck & Ruprecht, 1995); H. U. Steymans, “Die neuassyrische Vertragsrhetorik der ‘Vassal Treaties of Esarhaddon’ und das Deuteronomium,” in Das Deuteronomium (ed. G. Braulik; ÖBS 23; Frankfurt am Main: Lang, 2003), 89–152; E. Otto, Das Deuteronomium: Politische Theologie und Rechtsreform in Juda und Assyrien (BZAW 284; Berlin: de Gruyter, 1999). 43

How to Date Pentateuchal Texts

369

to situate the first edition of Deuteronomy in the time of King Josiah. Some of my German and Finnish colleagues may find this naïve and favor a date in the Babylonian or even Persian period,48 but I have never been able to understand why somebody at that particular time should construct the core of Deuteronomy according to Neo-Assyrian textual standards. This does not mean that all parts of Deuteronomy can be explained with Neo-Assyrian parallels, since Deuteronomy was newly edited and revised during the sixth and fifth centuries BCE, but the Neo-Assyrian parallels point to a seventh century date for the core of the book.49

Relative Dating by Internal Comparison Finally, the parallel traditions inside the Pentateuch allow a relative dating in many cases. The most obvious case, on which the majority of scholars agree, is the relationship between the Covenant Code in Exod 20–23* and the Deuteronomic Code. If the Deuteronomic Code was intended as a new edition of the Covenant Code, as argued, for instance, by B. M. Levinson,50 then it is clear that the Covenant Code must be at least forty or fifty years earlier than the first edition of the Deuteronomic Code. And if Lev 26 depends on Deut 28 and probably also on Priestly texts, it must be a later text,51 and the same, then, 48  R. G. Kratz, “Der literarische Ort des Deuteronomiums,” in Liebe und Gebot: Studien zum Deuteronomium; Festschrift zum 70. Geburtstag von Lothar Perlitt (ed. R. G. Kratz and H. Spieckermann; FRLANT 190; Göttingen: Vandenhoeck & Ruprecht, 2000), 101–120; J. Pakkala, “The Date of the Oldest Edition of Deuteronomy,” ZAW 121 (2009), 388–401. 49  This date can be supported by the recent discovery of a copy of Esarhaddonʼs loyalty oath in the temple of Tayinat. This discovery makes it very plausible that there was also a copy of Esarhaddonʼs vassal treaty (VTE) in Jerusalem. See H. U. Steymans, “Deuteronomy 28 and Tell Tayinat,” Verbum et Ecclesia 34 (2013), 13, http://www.ve.org.za/index.php/VE/article/ view/870 (accessed 02/16/2016). For the text, see J. Lauinger , “Esarhaddonʼs Succession Treaty at Tell Tayinat: Text and Commentary,” JCS 64 (2012), 87–123. Of course, one may also point out parallels between Deuteronomy and the Hittite vassal treaties, as has J. Berman, “Histories Twice Told: Deuteronomy 1–3 and the Hittite Treaty Prologue Tradition,” JBL 132 (2013), 229–250, but these parallels are not as close as those with VTE, and they show only that the Neo-Assyrian vassal rhetoric is probably influenced by Western forerunners, such as the Hittite treaties. 50  B. M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (New York: Oxford University Press, 1997). This idea is accepted by the majority of scholars; for different views, see J. Van Seters, A Law Book for the Diaspora: Revision in the Study of the Covenant Code (Oxford: Oxford University Press, 2003), and, with yet another perspective, W. Oswald, Israel am Gottesberg: Eine Untersuchung zur Literaturgeschichte der vorderen Sinaiperikope Ex 19–24 und deren historischem Hintergrund (OBO 159; Göttingen: Vandenhoeck & Ruprecht, 1998), and J. U. Ro, “The Portrayal of Judean Communities in Persian Era Palestine through the Lens of the Covenant Code,” Semitica 56 (2014), 249–289. 51  C. N ihan , “Heiligkeitsgesetz und Pentateuch: Traditions- und kompositionsge­ schichtliche Aspekte von Levitikus 26,” in Abschied von der Priesterschrift? Zum Stand der

370

Thomas Römer

holds true for the so-called Holiness Code (Lev 17–26), of which Lev 26 is the conclusion.52 This method can therefore provide important arguments for a relative chronology of pentateuchal texts, which can then tentatively fit into an absolute chronology. This dating of course remains hypothetical until we find documents amenable to radiocarbon dating.

Conclusion The most secure date for the existence of pentateuchal texts is the Persian period, because this setting can be deduced from hard evidence. The earlier one moves, the more complicated and hypothetical dates become. One should therefore start by considering whether a text fits the Persian period, as well as whether it is composite and what would allow for the identification of older layers. The first edition of Deuteronomy in the seventh century BCE has remained an apparently good point of comparison for the dating of older texts since the time of de Wette. As for P, I still think that a Persian period dating is the best option. But one should not dramatize the divergences. Even scholars who support a postexilic date acknowledge that the rituals and prescriptions in Lev 1–15 may well stem at least partially from the time of the First Temple.53 But it is methodologically more secure to consider the Priestly texts of the Torah first and foremost in the context of the Second Temple.

Pentateuchdebatte (ed. F. Hartenstein and K. Schmid; Veröffentlichungen der Wissenschaftlichen Gesellschaft für Theologie 40; Leipzig: Evangelische Verlagsanstalt, 2015), 186–218. 52  E. Otto, “Innerbiblische Exegese im Heiligkeitsgesetz: Levitikus 17–26,” in Levitikus als Buch (ed. H.-J. Fabry and H.-W. Jüngling; BBB 119; Berlin: Philo, 1999), 125–196; C. Nihan, “The Holiness Code between D and P: Some Comments on the Function and Significance of Leviticus 17–26 in the Composition of the Torah,” in Das Deuteronomium zwischen Pentateuch und Deuteronomistischem Geschichtswerk (ed. E. Otto and R. Achenbach; FRLANT 206; Göttingen: Vandenhoeck & Ruprecht, 2004), 81–122; J. Stackert, Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007). Although these authors have different ideas about the date and the role of the Holiness Code, they agree on the relative chronology. 53  C. Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 198–231.

The Numeral 11 and the Linguistic Dating of P Noam Mizrahi 1 Introduction The Hebrew Bible in general and the Pentateuch in particular are corpora of ancient literature that went through various stages of literary growth; they were canonized at some point of their history, and their texts were transmitted to us through the ages by the diligent work of countless scribes and tradents. Their language inevitably bears witness to the long and complicated process by which these texts were formed, reshaped, and transmitted. Like Homeric Greek or Avestan Iranian, the language of biblical texts can be submitted to historical-linguistic study, which reveals phases in the development of the language of the literary sources contained therein, thus affording assistance in chronologically situating texts that cannot be securely dated on other grounds. In the last generation, Avi Hurvitz, in particular, focused on developing a philologically sound method for distinguishing between early and late linguistic elements within Biblical Hebrew and hence for identifying the chronological setting of whole texts. He has paid particular attention to the methodological rigor of his procedure and integrated into it several controls that aim at filtering out all cases that might be late but cannot be proven to be so.1 Despite challenges of various sorts, most if not all historical linguists who specialize in Hebrew still consider Hurvitzʼs rigorous method one of the most compelling procedures available.2 1  See, e.g., A. Hurvitz, “Can Biblical Texts Be Dated Linguistically? Chronological Perspectives in the Historical Study of Biblical Hebrew,” in Congress Volume: Oslo, 1998 (ed. A. Lemaire and M. Sæbø; VTSup 80; Leiden: Brill, 2000), 143–160. For a full list of Hurvitzʼs publications, see A. H ornkohl , “Avi Hurvitz: List of Publications,” Studies in Language 11–12 (2009), *21–*27 (Hebrew). 2  For attempts to counter Hurvitzʼs method, see especially I. Y oung et al., Linguistic Dating of Biblical Texts (2 vols.; London: Equinox, 2008); see also R. Rezetko and I. Young, Historical Linguistics and Biblical Hebrew: Steps Toward an Integrated Approach (Atlanta: Society of Biblical Literature, 2014). Although it contains some points of valid criticism, however, their work was harshly criticized both for methodological fallacies and for betraying the lack of firsthand knowledge and in-depth understanding of the primary sources and their linguistic evidence; see, e.g., J. J oosten , review of Linguistic Dating of Biblical Texts, by I. Young et al., Babel und Bibel 6 (2012), 535–542, or the collection of studies edited by

372

Noam Mizrahi

Like any scientific method, this one too rests upon certain suppositions – which can be either accepted or rejected – and it has its own set of limitations. Both aspects are lucidly discussed in Hurvitzʼs publications. Each such element deserves detailed scrutiny, but the purpose of this essay is different. I would like to highlight a “blind spot” that is inherent in the method itself, a methodological problem that one is bound to tackle even if one accepts the methodʼs premises and acknowledges its limits, as I do. As we shall see, historical reconstruction of a certain chain of linguistic developments is no simple task, and its practical application in the form of linguistic dating of biblical texts is a complicated venture that requires one to carefully weigh conflicting data. The issue is best understood by way of illustration, and I have chosen to discuss a concrete example that has a direct bearing on the dating of P – a topic to which Hurvitz has dedicated a whole book3 and which has far-reaching implications for any attempt to historically appreciate the literary, social, cultural, and religious world of ancient Israel. To be sure, the example to be discussed has been briefly noted by Hurvitz himself,4 but without addressing all its complexities, which deserve sustained consideration. For the sake of the present discussion, I will work within the assumptions of Hurvitzʼs method, such as the treatment of P en bloc and the rough division of the linguistic history of Hebrew into an early, that is, preexilic stage (Standard or Classical Biblical Hebrew, abbreviated CBH), and a late, postexilic stage (Late Biblical Hebrew, or LBH), with the exilic period in the middle as a transitional stage. These and other assumptions are necessarily simplistic, and, in any case, they do not exactly reflect my own personal views of these matters.5 C. M iller -N audé and Z. Z evit, Diachrony in Biblical Hebrew (LSAWS 8; Winona Lake, IN: Eisenbrauns, 2012). An unnoticed yet important critique of Hurvitzʼs method, albeit from a literary-historical point of view, is that of Y. Hoffman, “The Lexicography of the P Document and the Problems Concerning its Dating,” in Studies in Judaica (ed. M. A. Friedman and M. Gil; Teʿuda 4; Tel Aviv: University Projects, 1986), 13–22 (Hebrew), reacting especially to A. H urvitz , “Studies in the Vocabulary of the Priestly Code: The Use of ‫ ְׁש ֵאר‬and ‫ֵׁש ָכר‬ in the Books of Leviticus and Numbers,” in Bible Studies: Y. M. Grintz in Memoriam (ed. B. Uffenheimer; Teʿuda 2; Tel Aviv: HaKibbutz HaMeuchad, 1982), 299–305 (Hebrew). 3  A. Hurvitz, A Linguistic Study of the Relationship between the Priestly Source and the Book of Ezekiel: A New Approach to an Old Problem (CahRB 20; Paris: Gabalda, 1982). For additional studies by Hurvitz on this topic, see also H ornkohl , “Avi Hurvitz: List of Publications” (see n. 1), items 8, 12, 17, 21, 25, 28, 29, 32, 37, 45. 4  Hurvitz, Linguistic Study (see n. 3) 145, with n. 6. 5  In my opinion, P is not a unified document but rather a literary complex, which contains both early and late strata. I hope to address this issue at greater length elsewhere; for the time being, see the debate between B. A. Levine, “Research in the Priestly Source: The Linguistic Factor,” ErIsr 16 (1982), 124–131 (Hebrew); idem , “Late Language in the Priestly Source: Some Literary and Historical Observations,” in Proceedings of the Eighth World Congress of Jewish Studies, Panel Sessions: Bible Studies and Hebrew Language (Jerusalem: World Union of Jewish Studies, 1983), 69–82, vs. the reaction of A. Hurvitz, “The Language of the

The Numeral 11 and the Linguistic Dating of P

373

Nevertheless, they will suffice for the present discussion, and, in any case, finer distinctions will not significantly alter the final conclusions.

2 Case Study 2.1 Evidence Hebrew, like other Semitic languages, denotes the numeral 11 with an additive composite expression, which encodes the notion of 1+10 but takes a syntactic form similar to that of a construct phrase.6 Biblical Hebrew presents two competing expressions for this morpho-lexical slot: A. The first makes use of the regular word for 1, namely, ‫( אחד עשר‬when counting masculine nouns) or ‫( אחת עשרה‬for feminine nouns).7 B. The alternative construction is similarly shaped, only that the digit is denoted by ‫עשתי‬, which is uninflected for gender and is restricted in use to this numeral alone: ‫ עשתי עשר‬for masculine nouns and ‫ עשתי עשרה‬for feminine ones.8 The full distribution of both compounds is presented in the following (tables A and B).9 Table A: ‫אחת עשרה‬/‫אחד עשר‬ Masculine 1. Genesis 32:23 ‫׃‬ ‫ת־א ַ ֥חד ָע ָ ׂ֖שר יְ ָל ָ ֑דיו ַו�ּֽיַ ֲע ֔בֹר ֵ ֖את ַ ֽמ ֲע ַ ֥בר יַ ּֽבֹק‬ ַ ‫ת־ׁש ֵ ּ֣תי ִׁש ְפח ָֹ֔תיו וְ ֶא‬ ְ ‫ת־ׁש ֵ ּ֤תי נָ ָׁש ֙יו וְ ֶא‬ ְ ‫וַ ָּי ָ�֣קם ׀ ַּב ַּל֣יְ ָלה ֗הּוא וַ ּיִ ַ ּ֞קח ֶא‬ The same night he got up and took his two wives, his two maids, and his eleven children, and crossed the ford of the Jabbok. 2. Genesis 37:9 ‫ּׁש ֶמׁש וְ ַהּיָ ֵ ֗ר ַח‬ ֣ ֶ ‫לֹום ֔עֹוד וְ ִה ֵּנ֧ה ַה‬ ֙ ‫אמר ִה ֵּ֨נה ָח ַל ְ֤מ ִ ּֽתי ֲח‬ ֶ ֹ ‫עֹוד ֲח ֣לֹום ַא ֵ֔חר וַ יְ ַס ֵ ּ֥פר א ֹ֖תֹו ְל ֶא ָ ֑חיו וַ ּ֗י‬ ֙ ‫וַ ַּי ֲֽח ֹ֥לם‬ ָ ‫וְ ַא ַ ֤חד ָע ָׂש ֙ר ּֽכ‬ ‫׃‬ ‫ֹוכ ִ֔בים ִ ֽמ ְׁש ַּת ֲחִ ֖וים ִ ֽלי‬ He had another dream, and told it to his brothers, saying, “Look, I have had another dream: the sun, the moon, and eleven stars were bowing down to me.”

Priestly Source and Its Historical Setting: The Case for an Early Date,” in Proceedings of the Eighth World Congress of Jewish Studies, Panel Sessions: Bible Studies and Hebrew Language (Jerusalem: World Union of Jewish Studies, 1983), 83–94. Not all of Levineʼs examples are equally persuasive, but neither are all of Hurvitzʼs objections equally valid. 6  For the linguistic typology of such numerical expressions, see P. W. van der H orst and G. M ussies , “Subtractive versus Additive Composite Numerals in Antiquity,” Illinois Classical Studies 13/1 (1988), 183–202. 7  Both expression are designated below collectively as “the ‫ אחד‬compound.” 8  Referred to hereafter as “the ‫ עשתי‬compound.” 9  The English translation of biblical passages follows the NRSV, unless noted otherwise.

374

Noam Mizrahi

3. Deuteronomy 1:2

‫׃‬ ‫ר־ׂש ִ ֑עיר ַ ֖עד ָק ֵ ֥דׁש ַּב ְר ֵנ ַֽע‬ ֵ ‫יֹום ֵ ֽמח ֵ ֹ֔רב ֶ ּ֖ד ֶרְך ַה‬ ֙ ‫ַא ַ֨חד ָע ָ ׂ֥שר‬ By the way of Mount Seir it takes eleven days to reach Kadesh-barnea from Horeb.

Feminine 4. Joshua 15:51 ‫׃) וְ ֹ֥ג ֶׁשן וְ ח ֹֹ֖לן וְ גִ ֹ֑לה‬ ‫מה וְ ָע ִ ֽנים‬ ֹ ֖ ‫׃ ַ ֽו ֲע ָנ֥ב וְ ֶא ְׁש ְּת‬ ‫ת־ס ָּנ֖ה ִ ֥היא ְד ִ ֽבר‬ ַ ַ‫׃ וְ ַד ָּנ֥ה וְ ִק ְרי‬ ‫(ּוב ָ ֑הר ָׁש ִ ֥מיר וְ יַ ִ ּ֖תיר וְ ׂשֹו ֽכֹה‬ ָ ‫׃‬ ‫יהן‬ ֽ ֶ ‫ת־ע ְׂש ֵ ֖רה וְ ַח ְצ ֵר‬ ֶ ‫ָע ִ ֥רים ַ ֽא ַח‬ (And in the hill country, Shamir, Jattir, Socoh, Dannah, Kiriath-sannah (that is, Debir), Anab, Eshtemoh, Anim,) Goshen, Holon, and Giloh: eleven towns with their villages. 5. 1 Kings 6:38 ‫ל־מ ְׁש ָּפ ָ ֑טו וַ ּיִ ְב ֵנ֖הּו‬ ִ ‫ּול ָכ‬ ְ ‫ל־ּד ָב ָ ֖ריו‬ ְ ‫ּוב ָּׁשנָ ֩ה ָ ֽה ַא ַ֨חת ֶע ְׂש ֵ ֜רה ְּב ֶי ַ�֣רח ּ֗בּול ֚הּוא ַה ֣חֹ ֶדׁש ַה ְּׁש ִמ ִ֔יני ָּכ ָל֣ה ַה ַּ֔ביִ ת ְל ָכ‬ ַ ‫׃‬ ‫ֶ ׁ֥ש ַבע ָׁש ִנֽים‬ In the eleventh year, in the month of Bul, which is the eighth month, the house was finished in all its parts, and according to all its specifications. He was seven years in building it. 6. 2 Kings 9:29

ֽ ָ ְ‫ן־א ְח ָ ֑אב ָמ ַ ֥לְך ֲא ַחזְ ָי֖ה ַעל־י‬ ַ ‫יֹורם ֶּב‬ ֖ ָ ‫ּוב ְׁשנַ ֙ת ַא ַ ֣חת ֶע ְׂש ֵ ֣רה ָׁש ָ֔נה ְל‬ ִ ‫׃‬ ‫הּודה‬ In the eleventh year of Joram son of Ahab, Ahaziah began to reign over Judah.

7. 2 Kings 23:36 ֣ ֵ ְ‫ּוׁש ָל֑םִ ו‬ ָ ‫ן־ע ְׂש ִ ֨רים וְ ָח ֵ ֤מׁש ָׁשנָ ֙ה יְ ֽהֹוי� ִ ָ֣קים ְּב ָמ ְל ֔כֹו וְ ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ָמ ַ ֖לְך ִּב ֽיר‬ ֶ ‫ֶּב‬ ]‫בּוּדה‬ ֥ ָ ְ‫ׁשם ִא ּ֔מֹו זבידה [ז‬ ‫׃‬ ‫ן־רּומה‬ ָֽ ‫ת־ּפ ָד ָי֖ה ִמ‬ ְ ‫ַב‬ Jehoiakim was twenty-five years old when he began to reign; he reigned eleven years in Jerusalem. His motherʼs name was Zebidah daughter of Pedaiah of Rumah.

= 2 Chronicles 36:5 ‫הו֥ה‬ ָ ְ‫ּוׁש ָל֑םִ וַ ַּי ַ֣עׂש ָה ַ ֔רע ְּב ֵע ֵינ֖י י‬ ָ ‫ן־ע ְׂש ִ ֨רים וְ ָח ֵ ֤מׁש ָׁשנָ ֙ה יְ ֽהֹוי� ִ ָ֣קים ְּב ָמ ְל ֔כֹו וְ ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ָמ ַ ֖לְך ִּב ֽיר‬ ֶ ‫ֶּב‬ ‫׃‬ ‫ֹלהיו‬ ֽ ָ ‫ֱא‬ Jehoiakim was twenty-five years old when he began to reign; he reigned eleven years in Jerusalem. He did what was evil in the sight of the LORD his God.

8. 2 Kings 24:18 ]‫מּוטל‬ ֥ ַ ‫[ח‬ ֲ ‫ׁשם ִא ּ֔מֹו חמיטל‬ ֣ ֵ ְ‫ּוׁש ָל֑םִ ו‬ ָ ‫ן־ע ְׂש ִ ֨רים וְ ַא ַ ֤חת ָׁשנָ ֙ה ִצ ְד ִק ָּי֣הּו ְב ָמ ְל ֔כֹו וְ ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ָמ ַ ֖לְך ִּב ֽיר‬ ֶ ‫ֶּב‬ ‫׃‬ ‫ַ ּֽבת־יִ ְר ְמָי֖הּו ִמ ִּל ְב ָנֽה‬ Zedekiah was twenty-one years old when he began to reign; he reigned eleven years in Jerusalem. His motherʼs name was Hamutal daughter of Jeremiah of Libnah.

= Jeremiah 52:1 ֲ ‫ׁשם ִא ּ֔מֹו חמיטל‬ ֣ ֵ ְ‫ּוׁש ָל֑םִ ו‬ ָ ‫ן־ע ְׂש ִ ֨רים וְ ַא ַ ֤חת ָׁשנָ ֙ה ִצ ְד ִק ָּי֣הּו ְב ָמ ְל ֔כֹו וְ ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ָמ ַ ֖לְך ִּב ֽיר‬ ֶ ‫ֶּב‬ ]‫מּוטל‬ ֥ ַ ‫[ח‬ ‫׃‬ ‫ַ ּֽבת־יִ ְר ְמָי֖הּו ִמ ִּל ְב ָנֽה‬ Zedekiah was twenty-one years old when he began to reign; he reigned eleven years in Jerusalem. His motherʼs name was Hamutal daughter of Jeremiah of Libnah.



= 2 Chronicles 36:11

‫׃‬ ִ‫ּוׁש ָ ֽלם‬ ָ ‫ן־ע ְׂש ִ ֧רים וְ ַא ַ ֛חת ָׁש ָנ֖ה ִצ ְד ִק ָּי֣הּו ְב ָמ ְל ֑כֹו וְ ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ָמ ַ ֖לְך ִּב ֽיר‬ ֶ ‫ֶּב‬ Zedekiah was twenty-one years old when he began to reign; he reigned eleven years in Jerusalem.

The Numeral 11 and the Linguistic Dating of P

375

9. Ezekiel 30:20

‫׃‬ ‫מר‬ ֹ ֽ ‫הו֖ה ֵא ַ ֥לי ֵלא‬ ָ ְ‫ח ֶדׁש ָה ָי֥ה ְד ַבר־י‬ ֹ ֑ ‫אׁשֹון ְּב ִׁש ְב ָ ֣עה ַל‬ ֖ ‫וַ יְ ִ֗הי ְּב ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ָ ּֽב ִר‬ In the eleventh year, in the first month, on the seventh day of the month, the word of the LORD came to me.

10. Ezekiel 31:1

‫׃‬ ‫מר‬ ֹ ֽ ‫הו֖ה ֵא ַ ֥לי ֵלא‬ ָ ְ‫ח ֶדׁש ָה ָי֥ה ְד ַבר־י‬ ֹ ֑ ‫יׁשי ְּב ֶא ָ ֣חד ַל‬ ֖ ִ ‫וַ יְ ִ֗הי ְּב ַא ַ ֤חת ֶע ְׂש ֵר ֙ה ָׁש ָ֔נה ַּב ְּׁש ִל‬ In the eleventh year, in the third month, on the first day of the month, the word of the LORD came to me.

Table B: ‫ה‬/‫עשתי עשר‬ Masculine 11. Numbers 7:72

‫׃‬ ‫ן־ע ְכ ָ ֽרן‬ ָ ‫יאל ֶּב‬ ֖ ֵ ‫יֹום ַע ְׁש ֵ ּ֣תי ָע ָ ׂ֣שר י֔ ֹום נָ ִ ׂ֖שיא ִל ְב ֵנ֣י ָא ֵ ׁ֑שר ַּפגְ ִע‬ ֙ ‫ְּב‬ On the eleventh day Pagiel son of Ochran, the leader of the Asherites [. . .].

12. Numbers 29:20 ֽ ִ ‫ֽי־ׁש ָנ֛ה ַא ְר ָּב ָ ֥עה ָע ָ ׂ֖שר ְּת ִמ‬ ָ ‫ילם ְׁש ָנ�֑יִ ם ְּכ ָב ִ ׂ֧שים ְּב ֵנ‬ ֣ ִ ‫י־ע ָ ׂ֖שר ֵא‬ ָ ‫יׁשי ָּפ ִ ֥רים ַע ְׁש ֵּת‬ ֛ ִ ‫ּובּי֧ ֹום ַה ְּׁש ִל‬ ַ ‫׃‬ ‫ימם‬ On the third day: eleven bulls, two rams, fourteen male lambs a year old without blemish [. . .]. 13. Deuteronomy 1:3 ‫הו֛ה‬ ָ ְ‫ל־ּב ֵנ֣י יִ ְׂש ָר ֵ֔אל ֠ ְּככֹל ֲא ֶׁ֨שר ִצָּו֧ה י‬ ְ ‫ח ֶדׁש ִּד ֶ ּ֤בר מ ֶֹׁש ֙ה ֶא‬ ֹ ֑ ‫ח ֶדׁש ְּב ֶא ָ ֣חד ַל‬ ֹ ֖ ‫י־ע ָ ׂ֥שר‬ ָ ‫הי ְּב ַא ְר ָּב ִ ֣עים ָׁש ָ֔נה ְּב ַע ְׁש ֵ ּֽת‬ ֙ ִ ְ‫ַו�ֽי‬ ‫׃‬ ‫א ֹ֖תֹו ֲא ֵל ֶ ֽהם‬ In the fortieth year, on the first day of the eleventh month, Moses spoke to the Israelites just as the LORD had commanded him to speak to them. 14. Zechariah 1:7 ָ ‫יֹום ֶע ְׂש ִ ֨רים וְ ַא ְר ָּב ָ֜עה ְל ַע ְׁש ֵ ּֽת‬ ֩ ‫ְּב‬ ‫הוה‬ ֗ ָ ְ‫י־ע ָ ׂ֥שר ֨חֹ ֶד ׁ֙ש הּוא־ ֣חֹ ֶדׁש ְׁש ָ֔בט ִּב ְׁש ַנ֥ת ְׁש ַ ּ֖תיִ ם ְל ָ ֽד ְר ָי�֑וֶ ׁש ָה ָי֣ה ְד ַבר־י‬ ‫׃‬ ‫מר‬ ֹ ֽ ‫ן־ע ּ֥דֹוא ַהּנָ ִ ֖ביא ֵלא‬ ִ ‫ן־ּב ֶ֣ר ְכ ָ֔יהּו ֶּב‬ ֶ ‫ֶאל־זְ ַכ ְריָ ֙ה ֶּב‬ On the twenty-fourth day of the eleventh month, the month of Shebat, in the second year of Darius, the word of the LORD came to the prophet Zechariah son of Berechiah son of Iddo; and Zechariah said [. . .]. 15. 1 Chronicles 12:14 ‫יאל‬ ֖ ֵ ‫ּתי ַה ִּׁש ִּׁ֔שי ֱא ִל‬ ֙ ַ ‫׃ ַע‬ ‫יעי יִ ְר ְמ ָי֖ה ַ ֽה ֲח ִמ ִ ֽׁשי‬ ִ֔ ‫׃ ִמ ְׁש ַמּנָ ֙ה ָ ֽה ְר ִב‬ ‫יאב ַה ְּׁש ִל ִ ֽׁשי‬ ֖ ָ ‫(עזֶ ר ָה ֑ר ֹאׁש ֽעֹ ַב ְדיָ ֙ה ַה ֵּׁש ִ֔ני ֱא ִל‬ ֵ֖ ‫׃‬ ‫ירי ַמ ְכ ַּב ַּנ֖י ַע ְׁש ֵ ּ֥תי ָע ָ ֽׂשר‬ ֔ ִ ‫הּו ָ ֽה ֲע ִׂש‬ ֙ ‫׃) יִ ְר ְמ ָ֨י‬ ‫יעי‬ ֽ ִ ‫ֹוחנָ ֙ן ַה ְּׁש ִמ ִ֔יני ֶא ְלזָ ָ ֖בד ַה ְּת ִׁש‬ ָ ֽ‫׃ י‬ ‫ַה ְּׁש ִב ִ ֽעי‬ (Ezer the chief, Obadiah second, Eliab third, Mishmannah fourth, Jeremiah fifth, Attai sixth, Eliel seventh, Johanan eighth, Elzabad ninth,) Jeremiah tenth, Machbannai eleventh. 16. 1 Chronicles 24:12 ֣ ‫אׁשֹון ִל‬ ֙ ‫ּגֹורל ָ ֽה ִר‬ ֤ ָ ‫(וַ ּיֵ ֵ֞צא ַה‬ ‫׃‬ ‫) ְל ֶא ְליָ ִׁש ֙יב ַע ְׁש ֵ ּ֣תי ָע ָׂ֔שר ְלי� ִ ָ֖קים ְׁש ֵנ֥ים ָע ָ ֽׂשר‬. . . ‫יהֹויָ ִ ֔ריב ִ ֽל ַיד ְע ָי֖ה ַה ֵּׁש ִנֽי‬ (The first lot fell to Jehoiarib, the second to Jedaiah, [. . .]) the eleventh to Eliashib, the twelfth to Jakim [. . .].

376

Noam Mizrahi

17. 1 Chronicles 25:18 ‫׃ ַה ְּׁש ִל ִ ׁ֣שי זַ ּ֔כּור ָּב ָנ֥יו וְ ֶא ָ ֖חיו‬ ‫ּוב ָנ֖יו ְׁש ֵנ֥ים ָע ָ ֽׂשר‬ ָ ‫הּו ַה ֵּׁש ִ֔ני ֽהּוא־וְ ֶא ָ ֥חיו‬ ֙ ‫יֹוסף ּגְ ַד ְל ָ֨י‬ ֑ ֵ ‫אׁשֹון ְל ָא ָ ֖סף ְל‬ ֛ ‫ּגֹורל ָ ֽה ִר‬ ֧ ָ ‫(וַ ּיֵ ֵ֞צא ַה‬ ‫׃‬ ‫י־ע ָ ׂ֣שר ֲעזַ ְר ֵ֔אל ָּב ָנ֥יו וְ ֶא ָ ֖חיו ְׁש ֵנ֥ים ָע ָ ֽׂשר‬ ָ ‫׃) ַע ְׁש ֵ ּֽת‬ ‫ירי ִׁש ְמ ִ֔עי ָּב ָנ֥יו וְ ֶא ָ ֖חיו ְׁש ֵנ֥ים ָע ָ ֽׂשר‬ ֣ ִ ‫ ָ ֽה ֲע ִׂש‬. . . ‫ְׁש ֵנ֥ים ָע ָ ֽׂשר‬ (The first lot fell for Asaph to Joseph; the second to Gedaliah, to him and his brothers and his sons, twelve; the third to Zaccur, his sons and his brothers, twelve; [. . .] the tenth to Shimei, his sons and his brothers, twelve;) the eleventh to Azarel, his sons and his brothers, twelve [. . .]. 18–19. 1 Chronicles 27:14 ‫׃‬ ‫יאל וְ ַעל֙ ַ ֽמ ֲח ֻל ְק ּ֔תֹו ֶע ְׂש ִ ֥רים וְ ַא ְר ָּב ָ ֖עה ָ ֽא ֶלף‬ ֑ ֵ ‫אׁשֹון ָי ָֽׁש ְב ָ ֖עם ֶּבן־זַ ְב ִּד‬ ֔ ‫(על ַה ַ ּֽמ ֲח ֹ֤ל ֶקת ָה ִ ֽראׁשֹונָ ֙ה ַל ֣חֹ ֶדׁש ָ ֽה ִר‬ ַ֞ ‫חֹוח֙י‬ ִ ‫ּדֹודי ָ ֽה ֲא‬ ֤ ַ ‫׃ וְ ַ֞על ַ ֽמ ֲח ֹ֣ל ֶקת ׀ ַה ֣חֹ ֶדׁש ַה ֵּׁש ִ֗ני‬ ‫אׁשֹון‬ ֽ ‫ל־ׂש ֵ ֥רי ַה ְּצ ָב ֖אֹות ַל ֥חֹ ֶדׁש ָ ֽה ִר‬ ָ ‫י־פ ֶרץ ָה ֛ר ֹאׁש ְל ָכ‬ ֶ ֗ ֵ‫ן־ּבנ‬ ְ ‫ִמ‬ ‫ירי ַמ ְה ַ ֥רי‬ ֔ ִ ‫ ָ ֽה ֲע ִׂש ִיר֙י ַל ֣חֹ ֶדׁש ָ ֽה ֲע ִׂש‬. . . ‫׃‬ ‫ּומ ְק ֖לֹות ַהּנָ ִג֑יד וְ ַעל֙ ַ ֽמ ֲח ֻל ְק ּ֔תֹו ֶע ְׂש ִ ֥רים וְ ַא ְר ָּב ָ ֖עה ָ ֽא ֶלף‬ ִ ‫ּומ ֲח ֻל ְק ּ֔תֹו‬ ַ֣ ‫י־ע ָׂש ֙ר ְל ַע ְׁש ֵ ּ֣תי ָע ָ ׂ֣שר ַה ֔חֹ ֶדׁש ְּבנָ ָי֥ה‬ ָ ‫׃) ַע ְׁש ֵ ּֽת‬ ‫ֹופ ִ ֖תי ַלּזַ ְר ִ ֑חי וְ ַעל֙ ַ ֽמ ֲח ֻל ְק ּ֔תֹו ֶע ְׂש ִ ֥רים וְ ַא ְר ָּב ָ ֖עה ָ ֽא ֶלף‬ ָ ‫ַהּנְ ֽט‬ ‫׃‬ ‫ן־ּב ֵנ֣י ֶא ְפ ָ ֑ריִ ם וְ ַעל֙ ַ ֽמ ֲח ֻל ְק ּ֔תֹו ֶע ְׂש ִ ֥רים וְ ַא ְר ָּב ָ ֖עה ָ ֽא ֶלף‬ ְ ‫תֹונ֖י ִמ‬ ִ ‫ַה ִּפ ְר ָ ֽע‬ (Jashobeam son of Zabdiel was in charge of the first division in the first month; in his division were twenty-four thousand. He was a descendant of Perez, and was chief of all the commanders of the army for the first month. Dodai the Ahohite was in charge of the division of the second month; Mikloth was the chief officer of his division. In his division were twenty-four thousand. [. . .] Tenth, for the tenth month, was Maharai of Netophah, of the Zerahites; in his division were twenty-four thousand.) Eleventh, for the eleventh month, was Benaiah of Pirathon, of the Ephraimites; in his division were twenty-four thousand.

Feminine 20. Exodus 26:7

‫׃‬ ‫י־ע ְׂש ֵ ֥רה יְ ִרי ֖עֹת ַ ּֽת ֲע ֶ ׂ֥שה א ָ ֹֽתם‬ ֶ ‫ל־ה ִּמ ְׁש ָ ּ֑כן ַע ְׁש ֵּת‬ ַ ‫א ֶהל ַע‬ ֹ ֖ ‫ית יְ ִרי ֣עֹת ִע ִּ֔זים ְל‬ ָ֙ ‫וְ ָע ִׂ֨ש‬ You shall also make curtains of goatsʼ hair for a tent over the tabernacle; you shall make eleven curtains.

21. Exodus 26:8 ‫יעה ָ ֽה ֶא ָ ֑חת ִמ ָ ּ֣דה ַא ַ֔חת ְל ַע ְׁש ֵ ּ֥תי ֶע ְׂש ֵ ֖רה‬ ֖ ָ ‫ים ָ ּֽב ַא ָּ֔מה וְ ֨ר ֹ ַח ֙ב ַא ְר ַּב֣ע ָ ּֽב ַא ָּ֔מה ַהיְ ִר‬ ֙ ‫ֹלׁש‬ ִ ‫יעה ָ ֽה ַא ַ֗חת ְׁש‬ ֣ ָ ‫א ֶרְך ׀ ַהיְ ִר‬ ֹ֣ ‫׃‬ ‫יְ ִרי ֽעֹת‬ The length of each curtain shall be thirty cubits, and the width of each curtain four cubits; the eleven curtains shall be of the same size. 22. Exodus 36:14

‫׃‬ ‫י־ע ְׂש ֵ ֥רה יְ ִרי ֖עֹת ָע ָ ׂ֥שה א ָ ֹֽתם‬ ֶ ‫ל־ה ִּמ ְׁש ָ ּ֑כן ַע ְׁש ֵ ּֽת‬ ַ ‫א ֶהל ַע‬ ֹ ֖ ‫וַ ַּ֨י ַע ׂ֙ש יְ ִרי ֣עֹת ִע ִּ֔זים ְל‬ He also made curtains of goatsʼ hair for a tent over the tabernacle; he made eleven curtains.

23. Exodus 36:15 ‫׃‬ ‫יעה ָ ֽה ֶא ָ ֑חת ִמ ָ ּ֣דה ַא ַ֔חת ְל ַע ְׁש ֵ ּ֥תי ֶע ְׂש ֵ ֖רה יְ ִרי ֽעֹת‬ ֣ ָ ‫ֹלׁש ֙ים ָ ּֽב ַא ָּ֔מה וְ ַא ְר ַּב֣ע ַא ּ֔מֹות ֖ר ֹ ַחב ַהיְ ִר‬ ִ ‫יעה ָ ֽה ַא ַ֗חת ְׁש‬ ֣ ָ ‫א ֶרְך ַהיְ ִר‬ ֹ֜ The length of each curtain was thirty cubits, and the width of each curtain four cubits; the eleven curtains were of the same size. 24. 2 Kings 25:2

‫׃‬ ‫וַ ָּת ֥ב ֹא ָה ִ ֖עיר ַּב ָּמ ֑צֹור ַ ֚עד ַע ְׁש ֵ ּ֣תי ֶע ְׂש ֵ ֣רה ָׁש ָ֔נה ַל ֶ ּ֖מ ֶלְך ִצ ְד ִק ָּיֽהּו‬ So the city was besieged until the eleventh year of King Zedekiah.

The Numeral 11 and the Linguistic Dating of P



377

= Jeremiah 52:5

‫׃‬ ‫וַ ָּת ֥ב ֹא ָה ִ ֖עיר ַּב ָּמ ֑צֹור ַ ֚עד ַע ְׁש ֵ ּ֣תי ֶע ְׂש ֵ ֣רה ָׁש ָ֔נה ַל ֶ ּ֖מ ֶלְך ִצ ְד ִק ָּיֽהּו‬ So the city was besieged until the eleventh year of King Zedekiah.

25. Jeremiah 1:3 ‫הּודה‬ ֑ ָ ְ‫אׁש ָּי֖הּו ֶ ֣מ ֶלְך י‬ ִ ֹ ‫ן־י‬ ֽ ‫הּודה ַעד־ּת ֹ֙ם ַע ְׁש ֵ ּ֣תי ֶע ְׂש ֵ ֣רה ָׁש ָ֔נה ְל ִצ ְד ִק ָּי֥הּו ֶב‬ ֔ ָ ְ‫הּו ֶ ֣מ ֶלְך י‬ ֙ ‫אׁש ָּ֨י‬ ִ ֹ ‫ן־י‬ ֽ ‫ימי יְ ֽהֹוי� ִ ָ֤קים ֶּב‬ ֵ֨ ‫וַ יְ ִ֗הי ִּב‬ ‫׃‬ ‫יׁשי‬ ֽ ִ ‫ח ֶדׁש ַ ֽה ֲח ִמ‬ ֹ ֥ ‫ּוׁש ַל֖םִ ַּב‬ ָ ‫ַעד־ּגְ ֥לֹות יְ ֽר‬ It came also in the days of King Jehoiakim son of Josiah of Judah, and until the end of the eleventh year of King Zedekiah son of Josiah of Judah, until the captivity of Jerusalem in the fifth month. 26. Jeremiah 39:2

‫׃‬ ‫ח ֶדׁש ָה ְב ְק ָ ֖עה ָה ִ ֽעיר‬ ֹ ֑ ‫יעי ְּב ִת ְׁש ָ ֣עה ַל‬ ֖ ִ ‫ח ֶדׁש ָ ֽה ְר ִב‬ ֹ ֥ ‫י־ע ְׂש ֵ ֤רה ָׁשנָ ֙ה ְל ִצ ְד ִק ָּ֔יהּו ַּב‬ ֶ ‫ְּב ַע ְׁש ֵ ּֽת‬ In the eleventh year of Zedekiah, in the fourth month, on the ninth day of the month, a breach was made in the city.

27. Ezekiel 26:1

‫׃‬ ‫מר‬ ֹ ֽ ‫הו֖ה ֵא ַ ֥לי ֵלא‬ ָ ְ‫ח ֶדׁש ָה ָי֥ה ְד ַבר־י‬ ֹ ֑ ‫י־ע ְׂש ֵ ֥רה ָׁש ָנ֖ה ְּב ֶא ָ ֣חד ַל‬ ֶ ‫וַ יְ ִ ֛הי ְּב ַע ְׁש ֵ ּֽת‬ In the eleventh year, on the first day of the month, the word of the LORD came to me.

28. Ezekiel 40:49 ‫ל־ה ֵא ֔ ִילים‬ ֣ ָ ‫ים ֶא‬ ֙ ‫ּוב ַּמ ֲע ֔לֹות ֲא ֶ ׁ֥שר ַי ֲֽע ֖לּו ֵא ָ ֑ליו וְ ַע ֻּמ ִד‬ ֽ ַ ‫א ֶרְך ָ ֽה ֻא ֞ ָלם ֶע ְׂש ִ ֣רים ַא ָּ֗מה וְ ֨ר ֹ ַח ֙ב ַע ְׁש ֵ ּ֣תי ֶע ְׂש ֵ ֣רה ַא ָּ֔מה‬ ֹ֣ ‫׃‬ ‫ֶא ָ ֥חד ִמ ֖ ֹּפה וְ ֶא ָ ֥חד ִמ ֽ ֹּפה‬ The length of the porch was twenty cubits, and the breadth eleven cubits; and it was by steps that it was ascended; and there were pillars by the posts, one on this side, and another on that side.10

Critical scholars long ago noticed that the distribution of the two compounds within the Tetrateuch is not incidental but rather aligns with literary, theological, and other stylistic features that distinguish between the Priestly (P) and nonPriestly (non-P) sources. The ‫ עשתי‬compound is attested in passages universally attributed to P,11 whereas tokens of the ‫ אחד‬compound, few as they are, are found only in non-P passages.12 The ‫ עשתי‬compound is found only in passages whose Priestly nature can be demonstrated on independent grounds. Furthermore, nowhere in P do we find the ‫ אחד‬compound; the ‫ עשתי‬compound is the exclusive means of expressing either the cardinal eleven or the ordinal eleventh. Hence, within the Tetrateuch, the ‫ עשתי‬construction is both typical of P and unique to it. Deuteronomy 1 presents equivocal testimony, since it includes the two compounds in consecutive verses.13 But an independent piece of evidence, namely, the peculiar way of fixing the date by reckoning the month and the day – a  Following the JPS (1917 edition). The NRSV for this passages renders not the MT but the LXX: “The depth of the vestibule was twenty cubits, and the width twelve cubits; ten steps led up to it; and there were pillars beside the pilasters on either side.” 11  Table B, no. 11–13, 20–23. See, e.g., J. E. Carpenter and G. Harford-Battersby, The Composition of the Hexateuch (London: Longmans & Green, 1902), 413, no. 57. 12  Table A, no. 1–3. The classical, so-called New Documentary Hypothesis assigns these passages to E. 13  Table A, no. 3; table B, no. 13. 10

378

Noam Mizrahi

formula that is unparalleled in D but is very common in P – suggests that v. 3 has its origin in P.14 By contrast, v. 2 has no characteristic of P in it, and there is no reason to deny it to D.15 If so, the ‫ עשתי‬compound is a distinctive linguistic feature of P, and one wonders if there is any conditioning factor that dictates the preference for the employment of one compound over the other. More specifically, can chronological development be discerned in the available sources, leading diachronically from one option to the other? 2.2 Analysis A ready answer for this query is offered by Hurvitzʼs method, which consists of three procedures: examination of distribution, identifying pertinent linguistic contrasts, and adducing external corroboration. Examination of distribution means that one should demonstrate that an element suspected of being late is otherwise attested, either exclusively or at least overwhelmingly, in biblical works that are indisputably late. Contrast with an equivalent element in CBH is required in order to show that speakers in former periods were indeed able to say the same thing but did so by other means, thus ensuring that the distribution is not a random product of chance but rather reflects actual replacement of an old element by a younger alternative. External corroboration controls for the limited scope of the biblical corpus by demonstrating that the suspicious element is similarly attested in late extrabiblical corpora. Particular importance is attached to the detection of parallels in Aramaic sources, since such evidence suggests the potential source of many items of LBH without denying that other, internally motivated changes could have been operative as well. It takes little to no effort to see that outside the Pentateuch, the ‫ עשתי‬compound becomes frequent in passages that relate the fall of Jerusalem, which  See, e.g., S. R. Driver, A Critical and Exegetical Commentary on Deuteronomy (3rd ed.; ICC; Edinburgh: T&T Clark, 1902), 7. This conclusion is not dependent on any specific model of the nature of P in general and of the precise construal of Deut 1:3 in particular, e.g., whether it had originally formed part of a continuous Priestly narrative or should rather be taken as a solitary element or redactional retouching. 15  Or from its putative non-P source(s), which some again identify with E. The idea that D relies on E has been advocated especially by M. Haran, “The Exodus Routes in the Pentateuchal Sources,” Tarbiz 40/2 (1971), 113–143 (Hebrew); idem , The Biblical Collection (4 vols.; Jerusalem: Bialik Institute, 1996–2014), 2:200–206 (Hebrew). Be that as it may, the matter discussed here is not affected by the precise identification of the literary source(s) from which these passages stem. The only distinction that is relevant for our concern – which ideally commands broad consensus among critical scholars – is that between P and non-P, regardless of the question whether such texts are to be considered continuous documents, disconnected fragments, or accumulated supplementations. Put differently, the present analysis should be evaluated independently of any specific theory concerning the compositional history of the Pentateuch. 14

The Numeral 11 and the Linguistic Dating of P

379

were obviously written only after this event had taken place, namely, in the exilic period at the earliest.16 It also prevails in the nonsynoptic passages of Chronicles,17 whereas the ‫ אחד‬compound is found only in synoptic passages, where they are copied verbatim from their source text in Kings. 18 The ‫עשתי‬ compound is also found in a superscription to a prophecy of First Zechariah, that is, not earlier than the Persian period.19 In other words, indisputably postexilic sources exhibit a distinct preference for the ‫ עשתי‬compound, whereas presumably preexilic sources prefer the ‫ אחד‬compound. The book of Ezekiel, which presents itself as the words of an exilic prophet, takes an intermediate position by evincing two tokens of either construction.20 This even distribution can be easily interpreted as reflecting a transitional stage from an older time, during which the ‫ אחד‬compound was the common way of expressing 11, to an era in which the ‫ עשתי‬compound took over. No wonder, then, that of the numerous arguments for a late date adduced by Friedrich Giesebrecht in his detailed monograph on the language of P,21 the ‫ עשתי‬compound is one of the very few examples that were left uncontested in S. R. Driverʼs devastating rejoinder to Giesebrechtʼs work22 and one of the only two examples that Driver himself cited as testifying to the late date of P.23 The evidence satisfies Hurvitzʼs first and second criteria, that is, distribution and linguistic contrast. The third criterion of external corroboration can easily be met with the help of the Dead Sea Scrolls, which attest the ‫ עשתי‬compound in a wide range of texts and contexts. It is found, for instance, in the War Scrollʼs description of the military banners: 1QM IV, 15–1724 ]‫ אות כול העדה אורך ארבע עשרה אמה אות ֗של[ושת השבטים אורך שלוש‬:‫[אורך הא]ו֯ ֗תות‬ ‫עש[רה אמה אות האלף עשר‬ ֗ ‫הר ֗בוא עשתי‬ ֗ ‫֯עשרה אמה [אות השבט ]שתים עשרה אמה ֗אות‬ ]. . . ‫שב[ע אמות‬ ֗ ‫[המ]אה ֗תשע אמות [אות החמשים שמ]ו֯ נ֯ ֗ה אמות אות העשרה‬ ֗ ]‫אמות אות‬  Table B, no. 24–26.  Table B, no. 15–19. 18  Table A, no. 7–8. 19  Table B, no. 14. 20  Table A, no. 9–10; table B, no. 27–28. 21  F. G iesebrecht , “Zur Hexateuchkritik: Der Sprachgebrauch des hexateuchischen Elohisten,” ZAW 1 (1881), 177–276, here 226. Giesebrechtʼs work was a reaction against C. V. Ryssel, De Elohistae Pentateuchici sermone: Commentatio historico-critica (Leipzig: Fernau, 1878), who opted for an earlier date for P on linguistic grounds. However, Ryssel did not mention the ‫ עשתי‬compound, and Giesebrecht appears to be the first to have brought it into the discussion of the linguistic dating of P. 22  S. R. D river , “On Some Alleged Linguistic Affinities of the Elohist,” Journal of Philology 11 (1882): 201–236. 23  S. R. Driver, An Introduction to the Literature of the Old Testament (9th ed.; Edinburgh: T&T Clark, 1913; repr., New York: Meridian, 1957), 156. 24  Y. Yadin, The Scroll of the War of the Sons of Light against the Sons of Darkness, trans. B. Rabin and C. Rabin (Oxford: Oxford University Press, 1962), 276–279. 16 17

380

Noam Mizrahi

[Length of the Bann]ers: The banner of the whole congregation: length fourteen cubits. The banner of th[ree tribes: length thir]teen cubits. [The banner of a tribe:] twelve cubits. The banner of a myriad: eleven [cubits. The banner of a thousand: ten cubits. The banner of a hundre]d: nine cubits. [The banner of a fifty: eigh]t cubits. The banner of a ten: sev[en cubits].

In addition, the ‫ עשתי‬compound is common in calendrical texts from Qumran.25 Last but not least, there is the telling treatment of the flood story in the Commentary on Genesis A (4Q252). The text of 4Q252 closely follows that of Genesis, but it inserts additional calendrical information in vv. 6 and 11, dating the narrated events to the “eleventh month” (4Q252, col. I, 11–18; table C).26 Table C: MT vs. 4Q252 MT Genesis 8:5–11 ‫וְ ַה ַּ֗מיִ ם ָה ֙יּו ָה ֣לֹוְך וְ ָח ֔סֹור‬ ‫ירי‬ ֑ ִ ‫ַ ֖עד ַה ֣חֹ ֶדׁש ָ ֽה ֲע ִׂש‬

Commentary on Genesis A ‫]הלוך וחסור‬ ֯ ‫והמים הי[ו‬11 And the waters continued to decrease ‫ עד החודש [הע]שירי‬until the [te]nth month (Gen 8:5a), ‫ באחד בו‬its first day,

‫ָ ּֽב ֲע ִׂש ִיר֙י ְּב ֶא ָ ֣חד ַל ֔חֹ ֶדׁש‬

‫ יום רביעי לשבת‬the fourth day of the week 12

‫׃‬ ‫אׁשי ֶ ֽה ָה ִ ֽרים‬ ֥ ֵ ‫נִ ְר ֖אּו ָר‬ ‫וַ יְ ִ֕הי ִמ ֵ ּ֖קץ ַא ְר ָּב ִ ֣עים י֑ ֹום‬

‫נר ֯או ראשי ההרים‬ ֗ the tops of the mountains appeared (Gen 8:5b). ‫ ויהי ֗מ ֯קץ ארבעים יום‬And it was at the end of forty days (Gen 8:6a) ‫ההר[ים‬ ֯ 13 ‫להראות ראשי‬ ֗ when the tops of the mountain[s] were visible

‫וַ ּיִ ְפ ַ ּ֣תח ֔נ ֹ ַח‬ ‫ת־ח ּ֥לֹון ַה ֵּת ָ ֖בה‬ ַ ‫ֶא‬

‫ויפ]תח נוח‬ ֗ [that] Noah [op]ened ‫ את חלון התבה‬the window of the ark (Gen 8:6b),

‫׃‬ ‫ֲא ֶ ׁ֥שר ָע ָ ֽׂשה‬ ‫ יום אחד בשבת‬on the first day of the week, ‫ הוא יום עשרה‬that is, the tenth day ‫בעש[תי עשר] החודש‬ ֯ 14 in the elev[enth] month. ‫ת־הע ֵ ֹ֑רב‬ ֽ ָ ‫וַ יְ ַׁש ַּל֖ח ֶא‬ ֙ ָ‫וַ ּיֵ ֵצ֤א י‬ ‫צֹוא וָ ׁ֔שֹוב‬  See specially 4Q317 9, 6 (M. A begg , “4Q317,” in The Dead Sea Scrolls Reader, Vol. 4: Calendrical and Sapiential Texts [ed. E. Tov and D. W. Parry; Leiden: Brill, 2004], 58–72, here 65); 4Q320 1 ii 2; 2, 13; 4Q321 I, 4; II, 7; III, 6; V, 3; VII, 5 (S. Talmon et al., Qumran Cave 4, XVI: Calendrical Texts [DJD 21; Oxford: Clarendon, 2001], 48, 49, 69, 71, 72, 75, 78, respectively). 26  The transcription and English translation follow G. J. B rooke , “252: 4QCommentary on Genesis A,” in G. J. Brooke et al., Qumran Cave 4, XVII: Parabiblical Texts, Part 3 (DJD XXII; Oxford, 1996), 185–207, here 194, 196. 25

The Numeral 11 and the Linguistic Dating of P

381

‫ַעד־יְ ֥בֹ ֶׁשת ַה ַ ּ֖מיִ ם‬ ‫׃‬ ‫ֵמ ַ ֥על ָה ָ ֽא ֶרץ‬ ‫ּיֹונ֖ה ֵ ֽמ ִא ּ֑תֹו‬ ָ ‫ת־ה‬ ַ ‫וַ יְ ַׁש ַ ּ֥לח ֶא‬ ֣ ַ ‫אֹות ה‬ ֙ ‫ִל ְר‬ ‫�ֲקּלּו ַה ַּ֔מיִ ם‬

‫ וישלח ֯את היונה‬And he sent the dove (Gen 8:8a) ‫ לראות הקלו המים‬to see whether the waters had abated (Gen 8:8b)

‫׃‬ ‫ֵמ ַ ֖על ְּפ ֵנ֥י ָ ֽה ֲא ָד ָ ֽמה‬ ‫ֹוח‬ ַ ֜‫ּיֹונה ָמנ‬ ָ֨ ‫א־מ ְצ ָא ֩ה ַה‬ ָ ֹ ‫וְ ֽל‬

‫מצאה מנוח‬15 ‫ ולוא‬but it did not find a resting place

‫ף־רגְ ֗ ָלּה‬ ַ ‫ְל ַכ‬ ‫ל־ה ֵּת ָ֔בה‬ ַ ‫וַ ָ ּ֤ת ָׁשב ֵא ָל ֙יו ֶא‬

‫]התבה‬ ֯ ‫ ותבוא אליו [אל‬and came back to him [to] the ark (Gen 8:9).

‫ל־ה ָ ֑א ֶרץ‬ ָ ‫ל־ּפ ֵנ֣י ָכ‬ ְ ‫י־מיִ ם ַע‬ ֖ ַ ‫ִּכ‬ ‫דֹו וַ ּיִ ָּק ֶ֔ח ָה‬ ֙ ָ‫וַ ּיִ ְׁש ַל֤ח י‬ ‫׃‬ ‫ל־ה ֵּת ָ ֽבה‬ ַ ‫ּיָבא א ָ ֹ֛תּה ֵא ָ ֖ליו ֶא‬ ֥ ֵ ַ‫ו‬ ‫וַ ָּי ֶ֣חל ֔עֹוד‬ ‫ִׁש ְב ַ ֥עת יָ ִ ֖מים ֲא ֵח ִ ֑רים‬ ‫ּיֹונ֖ה‬ ָ ‫ת־ה‬ ַ ‫וַ ֹּ֛י ֶסף ַׁש ַ ּ֥לח ֶא‬

‫ ויחל עוד‬And he again waited ]‫ שבעת ימים ֗א[חרים‬a[nother] seven days, ‫ויוסף לשלחה‬16 and he sent it out again (Gen 8:10)

‫׃‬ ‫ן־ה ֵּת ָ ֽבה‬ ַ ‫ִמ‬ ‫וַ ָּת ֨ב ֹא ֵא ָל֤יו ַהּיֹונָ ֙ה‬

‫ ותבוא אליו‬and it came back to him

‫ְל ֵע֣ת ֶ֔ע ֶרב‬ ‫ה־ז�֖יִ ת‬ ַ ‫וְ ִה ֵּנ֥ה ֲע ֵל‬ ‫יה‬ ָ ‫ָט ָ ֣רף ְּב ִ ֑פ‬

‫ ועלי זית‬and leaves of an olive tree, ‫ טרף בפיה‬newly plucked, were in its beak (Gen 8:11a). ‫וארבעה‬17 ]‫[ [הוא יום עשרים‬This was the twenty-]fourth day ‫ לעשתי עשר החודש‬of the eleventh month, ‫בשב[ת‬ ֗ ‫ באחד‬on the first day of the wee[k.

‫וַ ֵּי ַ�֣דע ֔נ ֹ ַח‬ ֥ ַ ‫ִּכ‬ ‫י־קּלּו ַה ַ ּ֖מיִ ם‬ ‫׃‬ ‫ֵמ ַ ֥על ָה ָ ֽא ֶרץ‬

‫ וידע נוח‬And Noah knew ]‫ כי קלו המים‬that the waters had abated] ‫מעל הארץ‬18 from upon the earth (Gen 8:11b).

The scriptural text underlying 4Q252 generally aligns with the LXX,27 but the Vorlage of this part of the commentary does not appear to diverge significantly from the MT or ancient versions. Indeed, virtually none of its distinctive pluses or minuses – in this particular passage – are supported by the other textual witnesses of Genesis.28 There is no reason, therefore, to assume that the calendrical  G. J. Brooke, “Some Remarks on 4Q252 and the Text of Genesis,” Textus 19 (1998), 1–25.  See also Brooke, “252: 4QCommentary on Genesis A” (see n. 26), 197–198, and corresponding notes on the biblical text. 27

28

382

Noam Mizrahi

insertions of 4Q252 are derived from a deviating scriptural Vorlage; they more probably stem from the hand of a Second Temple interpreter. The picture that emerges from all Hebrew sources, as analyzed using Hurvitzʼs method, is that a linguistic change took place with respect to the marking of the digit in the compound numeral for 11: CBH had used ‫ אחד‬or ‫ אחת‬for this purpose, but LBH replaced it with ‫עשתי‬. P, so it seems, aligns in this case with LBH – contrary to the thesis advanced by Hurvitz himself regarding the early linguistic date of P. 2.3 Problems The presence of any linguistic element in any given chronolect can be accounted for by one of three main options: inheritance, borrowing, or innovation. Inheritance means that the use of this element simply continues a historically preceding phase. Borrowing means that the element in question did not exist in the immediate ancestor of the variety in which it is attested but was rather copied by the speakers from a contemporary variety or language with which they were in contact. If the element under scrutiny was neither inherited from previous generations nor borrowed from a contemporaneous language or a different dialect, it most probably evolved independently within the speech community, and it should be regarded as an internal innovation. Which one of these options is relevant for the case of the ‫ עשתי‬compound? As mentioned above, many peculiar features of LBH are due to lexical borrowing from – or grammatical interference of – Aramaic, with which speakers of Hebrew came into intense contact from the exilic period onwards. This usual suspect, however, has a compelling alibi in this particular case: despite the misleading information supplied by some dictionaries, Aramaic has no cognate of ‫ עשתי‬functioning as a numeral.29 Nevertheless, most scholars who have dealt with the issue favored the idea of borrowing, identifying the source language with Akkadian, in which we find the word ištēn, which denotes both the cardinal one and the ordinal first, also marking the digit in the compound numeral 11. According to this hypothesis, which has its roots in a comment

29  See, e.g., HALOT, 898, s.v. ‫עׁשתי‬. The alleged cognate from Egyptian Aramaic that is mentioned there (‫ )עשתא‬is a substantive, not a numeral, and it is very doubtful that the two words are etymologically related. See, e.g., A. E. Cowley, Aramaic Papyri of the Fifth Century B.C. (Oxford: Clarendon, 1923), 24. See also J. Hoftijzer and K. Jongeling, Dictionary of the North-West Semitic Inscriptions (2 vols.; Leiden: Brill, 1995), 2:895–896, s.v. ˁštˀ. The Common Aramaic numeral for 1 is ‫חד‬, which is also the universal means for marking the digit in 11 in the various Aramaic dialects. A. Bendavid, Biblical Hebrew and Mishnaic Hebrew (2 vols.; Tel Aviv: Dvir, 1967–1971), 2:470 (Hebrew), even suggested that the total absence of an Aramaic cognate of Hebrew ‫ עשתי‬might account for the disappearance of the latter from Mishnaic Hebrew.

The Numeral 11 and the Linguistic Dating of P

383

made by Wellhausen,30 speakers of Hebrew came into contact with Akkadian in the Babylonian exile, and it is this situation that explains the appearance of ‫ עשתי‬in LBH.31 This proposal, however, is untenable. Even if one accepts the implicit assumption that Akkadian was still a widely spoken vernacular in Babylonia during the middle of the sixth century BCE (which is debatable), the evidence demonstrates that Akkadian could not have been the source of the Hebrew word for phonological reasons, since Akkadian lost all word-initial gutturals no later than the early second millennium BCE.32 Hence, had speakers of Hebrew adopted this word from Akkadian in the middle of the first millennium, they would have had no way of knowing that it should begin with the pharyngeal fricative /ʕ/ (represented by the letter ‫)ע‬. Yet this initial consonant is surely original, as demonstrated by the grammatical form of the cognate words in Ugaritic and some dialects of Old South Arabian (table D). Table D: Comparative Evidence Akkadian33 1 11

Masculine ištēn (cardinal and ordinal)34 ištēššeret

Feminine išteat, ištēt ištēššer

30  Giesebrecht, “Zur Hexateuchkritik” (see n. 21), 226, n. 1: “Beachtenswerth erscheint bei den schwierigen Fragen, welche dieses Wort zu beantworten giebt, eine mir von Prof. Wellhausen mitgetheilte Vermuthung, daß dasselbe von dem assyrischen Zahlwort für eins ‘iśtin’ [. . .] abhänge. Hätten die Israeliten dieses Wort von den Euphratsemiten angenommen, dann wäre das späte Auftreten desselben in der hebräischen Literatur mit einem Schlage erklärt.” 31  H. Zimmern, Akkadische Fremdwörter als Beweis für Babylonischen Kultureinfluß (2nd ed.; Leipzig: Hinrichs, 1917), 65, despite the important reservation mentioned in his earlier work, idem, Vergleichende Grammatik der Semitischen Sprachen (Berlin: Reuther & Reichard, 1898), 181 (see also J. Oppert, Éléments de la Grammaire Assyrienne [2nd ed.; Paris: Franck, 1868], 38, § 93). For the view that Hebrew ‫ עשתי‬is an Akkadian loan, see also GKC, 290, § 97e, n. 1; M. Ellenbogen, Foreign Words in the Old Testament: Their Origin and Etymology (London: Luzac, 1962), 129. Contrast the view of P. V. Mankowski, Akkadian Loanwords in Biblical Hebrew (HSS 47; Winona Lake, IN: Eisenbrauns, 2000), 126, who rightly concludes from the comparative evidence that the Hebrew word must go back independently to Proto-Semitic. 32  For the historical phonology of the gutturals in Akkadian, see especially N. J. C. Kouwenberg, “The Reflexes of the Proto-Semitic Gutturals in Old Assyrian,” in The Akkadian Language in Its Semitic Context: Studies in the Akkadian of the Third and Second Millennium BC (ed. G. Deutscher and N. J. C. Kouwenberg; Leiden: Nederlands Instituut voor het Nabije Oosten, 2006), 150–176. 33  See, e.g., W. von S oden , Grundriss der akkadischen Grammatik (3rd ed.; AnOr 33; Rome: Pontifical Biblical Institute, 1995), 113–114. 34  Alongside maḫrû(m) and pānû(m). Note also (w)ēdu(m), “single, sole, alone.” See also J. Huehnergard, A Grammar of Akkadian (2nd ed.; HSS 45; Winona Lake, IN: Eisenbrauns,

384

Noam Mizrahi

Ugaritic35 1 11

Masculine ʾaḥd (cardinal and ordinal) ʿšty (cardinal and ordinal?) ʿšt ʿšrh (ʿšrt)

Feminine ʾaḥt ʿšt ʿšr

Old South Arabian36 1 11

Sabaic Minaic Qatabanic Sabaic Qatabanic

Masculine ʾḥd (cardinal), qdm (ordinal) ʿs1t ʿs1tn ʾḥd ʿs2r ṭd ʿs2r

Feminine Sabaic ʾḥt

Similarly, neither Ugaritic nor Old South Arabian can be the source language for such a loan. Ugaritic had apparently died out before Hebrew split from Canaanite to become an independent language, and we know of no substantial contact between Hebrew and South Arabian that could have resulted in borrowing a grammatically entrenched word such as a numeral, in contrast to isolated lexical items such as culture words and names of exotic products. According to the rules of the comparative method in historical linguistics,37 the existence of independent cognates of the word ‫ עשתי‬in languages that represent three distinct branches of the Semitic phylum must be interpreted as testifying to a shared inheritance that goes back to Proto-Semitic.38 Alternatively, 2005), 239, § 23.2c. 35  See, e.g., J. T ropper , Ugaritische Grammatik (AOAT 273; Münster: Ugarit-Verlag, 2000), 344–345, § 62.11; 368, § 63.211 (note also 370, § 63.31); see also P. B ordreuil and D. Pardee, A Manual of Ugaritic (Winona Lake, IN: Eisenbrauns, 2009), 35–36; G. del Olmo Lete and J. Sanmartín, A Dictionary of the Ugaritic Language in the Alphabetic Tradition (trans. W. G. E. Watson; 2 vols.; Leiden: Brill, 2003), 1:190, s.v. ʕšt(y). The implications of the Ugaritic form for the history of its Hebrew cognate were first noted by S. E. Loewenstamm, “The Numerals in Ugaritic,” in Proceedings of the International Conference of Semitic Studies (Jerusalem: The Israel Academy of Sciences and the Humanities, 1965), 172–179, here 177. 36  See, e.g., A. F. L. Beeston, A Descriptive Grammar of Epigraphic South Arabian (London: Luzac, 1962), 40, § 35:1–2; 41, § 35:12; P. Stein, “Ancient South Arabian,” in The Semitic Languages: An International Handbook (ed. S. Weninger et al.; HSK 36; Berlin: de Gruyter Mouton, 2011), 1042–1072, here 1052–1054. 37  For its outline, see, e.g., R. L. Rankin, “The Comparative Method,” in The Handbook of Historical Linguistics (ed. B. D. Joseph and R. D. Janda; Oxford: Blackwell, 2003), 183–212. For contemporary reflections on its principles, see, e.g., M. Durie and M. Ross (eds.), The Comparative Method Reviewed: Regularity and Irregularity in Language Change (New York: Oxford University Press, 1996). 38  A. W ilson -W right , “The Word for ‘One’ in Proto-Semitic,” JSS 59/1 (2014), 1–13. Corroboration for this hypothesis is furnished by possible Afroasiatic cognates, whose existence suggests that the word in question belongs to the vocabulary that Proto-Semitic

The Numeral 11 and the Linguistic Dating of P

385

it was proposed that ‫ עשתי‬might be a very early loan from Akkadian into Central Semitic.39 But even if so, it had been borrowed at least a millennium before Hebrew came into existence. In any case, its presence in Ugaritic confirms that ‫ עשתי‬was present in Northwest Semitic, a branch to which Hebrew also belongs. In other words, ‫ עשתי‬must have been part and parcel of the vocabulary of Hebrew since its earliest phase, and it must have been an integral part of preexilic Hebrew as well – contrary to our previous conclusion. If so, the ‫ עשתי‬compound cannot be used as an argument in favor of dating P to exilic or postexilic times; the form in question was evidently available for speakers of preexilic Hebrew as well.40 Although this conclusion has its own intrinsic value, its importance in the context of the present discussion lies in the fact that it exposes a grave methodological problem: conflicting conclusions were reached on the basis of two different sets of linguistic information and analytical procedures. Philological analysis of the Hebrew sources suggests that ‫ עשתי‬is peculiar to LBH, whereas the comparative Semitic evidence requires the opposite conclusion, namely, that the word was present in preexilic Hebrew as well. Evidently, this case managed to pass all the controls of Hurvitzʼs method. The theoretical background of this failure lies in the fact that although Hurvitzʼs methodology makes use of the external corroboration criterion, it is basically an application of internal reconstruction (IR), focusing especially on lexical replacements. Internal reconstruction is an established method of historical linguistics, but its reliability is sometimes limited compared to comparative reconstruction (CR) because within the scope of any given language, processes of linguistic change may irrecoverably eliminate traces of older forms and intermediate stages, whereas cognate languages – whose matrices of changes are never exactly the same with respect to all related dialects – may preserve the missing evidence.41 inherited from its own ancestor, Proto-Afroasiatic; see V. B lažek , “Etymologizing the Semitic Cardinal Numerals of the First Decad,” in New Data and New Methods in Afroasiatic Linguistics: Robert Hetzron in Memoriam (ed. A. Zaborski; Wiesbaden: Harrassowitz, 2001), 13–37, here 15–16, § 1.2. There is no basis for the conjecture that Akkadian ištēn is borrowed from another, non-Semitic language such as Hurrian, as suggested by F. A. Dombrowski and B. W. W. D ombrowski , “Numerals and Numeral Systems in the Hamito-Semitic and Other Language Groups,” in Semitic Studies in Honor of Wolf Leslau (ed. A. S. Kaye; 2 vols.; Wies­ baden: Harrassowitz, 1991), 1:340–381, here 351–358. Equally baseless is their attempt to argue that Akkadian retained an initial pharyngeal in this particular word until later than is usually assumed (347–349). 39  R. Hetzron, “Innovations in the Semitic Numeral System,” JSS 22/2 (1977), 167–201, here 178–179. 40  See also C. Brockelmann, Grundriss der vergleichenden Grammatik der semitischen Sprachen (2 vols.; Berlin: Reuther & Reichard / New York: Lemcke & Buechner, 1908–1913), 1:490. 41  See, e.g., H. H. H ock , Principles of Linguistic Change (2nd ed.; Berlin: Mouton de

386

Noam Mizrahi

2.4 Potential Solutions Several factors are commonly regarded as having the potential to interrupt the expected linear diachronic continuity between successive chronolects. (a) One such factor is dialectal variation, since distinct dialects of a language may differ from one another in their preservation or loss of inherited features. Such an explanation has indeed been evoked for our case by Stanley Gevirtz.42 According to this hypothesis, CBH represents a dialect of Hebrew in which ‫עשתי‬ had been rendered obsolete, but it was presumably preserved in another, contemporaneous dialect of Hebrew – more specifically of a northern provenance, according to Gevirtz – and resurfaced for one reason or another in LBH, which was no longer tied to the dialect underlying CBH. I doubt, however, that this explanation can be applied to our case. There is no evidence linking this feature with northern Israel, as none of the passages in which the ‫ עשתי‬compound occurs deal with northern places, characters, or events. It is no coincidence that this particular feature is missing altogether even from Gary Rendsburgʼs maximalist list of lexical and grammatical features of what he terms Israelian Hebrew.43 Nor is it easily assumed that Pʼs language should systematically reflect northern features while passages traditionally ascribed to E – held by early critics to stem from northern Israel – exhibit the equivalent yet presumably southern form. Such a scenario stretches the limits of reasonability. (b) Another possible solution for the contradictory deployment of diachronic markers is to apply to the notion of archaizing, assuming that late (exilic or postexilic) authors made a conscious effort to stylize their language following classical models, in much the same way that Latin was used in medieval Europe while the actual romance vernaculars of the time were already far removed from the classical language.44 Gruyter, 1991), 532–555, esp. 549–550; L. Campbell, Historical Linguistics: An Introduction (3rd ed.; Cambridge, MA: MIT Press, 2013), 198–220, esp. 211–212; D. R inge , “Internal Reconstruction,” in The Handbook of Historical Linguistics (ed. B. D. Joseph and R. D. Janda; Oxford: Blackwell, 2003), 244–261. 42  S. Gevirtz, “Of Syntax and Style in the ‘Late Biblical Hebrew’–‘Old Cannanite’ Connection,” JANES 18 (1986), 25–29, here 26, n. 7. 43  G. A. R endsburg , “A Comprehensive Guide to Israelian Hebrew: Grammar and Lexicon,” Orient 38 (2003), 5–35. Note that Brockelmann, Grundriss (see n. 40), 1:490, also speaks about “eine alte Dialektform,” but without committing to any geographical identification. 44  As a matter of fact, scholars often invoke this analogy without being fully aware that no serious Latinist could mistake the language of any medieval Latin text; late authors furnish the historical linguist with a plethora of linguistic indications for their time, and the varieties of Latin in the Middle Ages were not at all identical to Classical Latin; see, e.g., J. Clackson and G. Horrocks, The Blackwell History of the Latin Language (Oxford: Blackwell, 2007), 265–304.

The Numeral 11 and the Linguistic Dating of P

387

Such a solution, however, is again not easily applicable to our case. Though archaizing can indeed blur the linguistic profile of our records, this mechanism too depends on the availability of the archaic forms to the users, often on the basis of learned study of ancient texts that were cherished as an authoritative writ. But if we insist on an exilic or postexilic date for P, we are left with virtually no (potentially) preexilic biblical texts that utilize the ‫ עשתי‬compound. The question then arises: What is the old model on which late authors were relying when archaizing their language? Of course, one can argue that these authors were emulating the language of other old texts that are now lost to us. However, such an assumption only further complicates the picture, since it requires us to assume that late authors were influenced – for the matter concerned here – only by such lost sources. But even if this highly unlikely scenario is assumed, we are still forced to admit that preexilic Hebrew had the form ‫ עשתי‬included in its lexicon, as otherwise it could not have been taken up by later archaizers. If archaizing was at all operative in employing the ‫ עשתי‬compound, it is actually much more reasonable to assume that P is a preexilic source, because P is the only source utilizing ‫ עשתי‬whose date is ambiguous. This is the only explanation that would supply us with the missing link, that is, an old model that later authors (such as the Chronicler) were able to follow. (c) If both the ‫ אחד‬and the ‫ עשתי‬compounds were available for speakers of preexilic Hebrew, the choice between them was apparently conditioned by a nonchronological factor. All languages exhibit synchronic variation, which is motivated by numerous factors of all sorts, including sociolinguistic variables. For instance, the notion of register or style (in the sociolinguistic sense of the term) typically establishes a range of linguistic usages between the opposing poles of formal and informal language, sometimes selecting conspicuously archaic forms for the former and innovated, colloquial forms for the latter.45 It is noteworthy, in this regard, that several occurrences of the ‫ עשתי‬compound are embedded in a chronological formula that relates to a single date: the eleventh year of either Jehoiachin or Zedekiah,46 who were both crowned in the same year (2 Kgs 24:8–17). Significantly, this is also the year of the dramatic fall of Jerusalem. It is conceivable that this date was linguistically marked as a formal

45  This hypothesis is not to be confused with the notion of style as invoked by Y oung et al., Linguistic Dating of Biblical Texts (see n. 2), esp. 2:72–105, who use it as a cover term for LBH as a whole without supplying any explanation for why particular choices were made in given contexts. For a concise survey of various sociolinguistic approaches to style, see, e.g., C. F ought , “Style and Style Shifting,” in Encyclopedia of Language and Linguistics (ed. K. Brown; 2nd ed.; Boston: Elsevier, 2006), 210–212, http://www.sciencedirect.com/ science/article/pii/B0080448542015236 (accessed 03/17/16). For a broader perspective, see, e.g., N. C oupland , “The Sociolinguistics of Style,” in The Cambridge Handbook of Sociolinguistics (ed. R. Mesthrie; Cambridge: Cambridge University Press, 2011), 138–156. 46  See table B, no. 24–27.

388

Noam Mizrahi

calendrical designation due to its historical importance in the collective memory of (former) Judeans. Contemporary analogy can be adduced from Modern Hebrew. Speakers of Israeli Hebrew make use of two parallel systems of calendrical designation: A. The so-called Hebrew calendar: an inherited system that makes use of the Babylonian months, whose days are denoted by Hebrew letters (e.g., ‫ט״ז‬ ‫באדר א׳ תשע״ו‬,47 “I Adar 15, 5776”) B. The so-called civil calendar: the international, Gregorian calendar, whose days are denoted by numerals (e.g., 2016 ‫ בפברואר‬25,48 “February 25, 2016”) The alternate use of these systems is conditioned by a set of sociolinguistic variables, especially the expression of religious vs. secular identity. But it also entails register, since secular speakers, who normally prefer the civil calendar for daily and mundane purposes, resort to the Hebrew calendar for formal occasions and usage. Most conspicuously, the civil calendar is usually used when narrating history or alluding to historical events, but a few dates of particular national importance are denoted only by the Hebrew calendar, such as Independence Day (‫ה׳ באייר תש״ח‬, rather than the equivalent Gregorian date of May 14, 1948), or by a mixture of elements drawn from the Hebrew and civil calendars (‫כ״ט בנובמבר‬ 1947, the day on which the UN approved the partition plan for Palestine). Such dates underwent a process of idiomatization in Modern Hebrew, and they are treated as bound expressions or even phraseological units. The biblical designation of ‫עשתי עשרה שנה‬, “the eleventh year,” might function in a way comparable to the modern ‫כ״ט בנובמבר‬, in the sense that it is a recurrent date formula that linguistically marks the counted element – the year in the case of the former, or the day in the case of the latter – by a conspicuously archaic numeral, thus encoding the ideological importance of these dates in the historical discourse and cultural memory of biblical authors and their ancient audience.49 Admittedly, this solution is purely conjectural. The nature of the factor(s) conditioning the distribution of the ‫ אחד‬and ‫ עשתי‬compounds remains obscure, as the limited linguistic data at our disposal does not permit full elucidation of the issue, and any hypothesis in this regard is bound to remain speculative. One can only hope that epigraphic evidence will eventually come to light and help  Pronounced as [teːt ˈzajin bəʔaˈdaːr ˈʔaːlɛf taːv ˌʃiːn ʕajin ˈva:v].  Pronounced as [ʕɛsˌriːm vəχamiˈʃaː bəˌfɛbruːˈʔaːr] (or: [ʕɛsˈriːm vəχamiˈʃiː ləˌfɛbruː­ ˈʔaːr]) [ʔalˌpajim vəˌʃeːʃ ʕɛsˈreː]. 49  The use of different dating formulae and calendrical systems is also attested in other periods in the history of Hebrew and Aramaic, albeit in a variety of constellations and realizations. See also, e.g., H. Misgav, “Dating Formulae and Their Use in Jewish Documents from the Second Temple, Mishnaic, and Talmudic Periods,” in Measuring and Weighing in Ancient Times (Haifa: Reuben and Edith Hecht Museum, 2001), exhibition catalog, 41*–45*. 47

48

The Numeral 11 and the Linguistic Dating of P

389

us discern more clearly some of the variables that were operative in governing the distribution of the two competing numerical expressions in ancient Hebrew.

3 Conclusion The ‫ עשתי‬compound appears, upon first and even second glance, to be one of the few cases that present a compelling case for the linguistic lateness of P. Closer scrutiny, however, reveals that it is, at best, a chronologically ambiguous marker, which cannot bear the burden of demonstrating the affiliation of the language of P with LBH.50 On the contrary, typological and historical considerations point to the conclusion that the ‫ עשתי‬compound goes back to Proto-Hebrew and cannot be considered a newcomer in the exilic or postexilic age. At the same time, although this numeral is most probably an inherited item of preexilic Hebrew, it cannot be taken as a proof for dating P to the monarchic period, since its use continued well into the Second Temple period. Given the present state of our knowledge, it is difficult to identify the factors that conditioned the distribution and use of the ‫ אחד‬and ‫ עשתי‬compounds in preexilic Hebrew. This issue, however, is immaterial for our methodological concern; the crucial fact for the present discussion is that this case is tellingly overlooked by our main philological tool, and a totally mistaken conclusion would have been reached unless the comparative evidence had been independently scrutinized. If one wishes to trace the history of the ‫ עשתי‬compound – or any other linguistic element suspected of being a marker of a late date – despite these difficulties, then one cannot be satisfied with using only one analytical procedure. As powerful and persuasive as Hurvitzʼs method may be, it must be supplemented with other, well-tested procedures, such as the comparative method, that help us determine the historical development that linguistic elements underwent within broader contexts, such as the language family. Hurvitzʼs philological method focuses on uncovering the diachronic relation between competing elements within the internal history of Hebrew, but one should also carefully examine their status as elements of Hebrew in the first place. One should ask what the potential source of each such element is, that is, whether it has been inherited, borrowed, or innovated. Without attempting to trace as accurately as possible its path of development, one runs the risk of falsely identifying items as being late. For those of us who are committed to the critical, historically oriented approach to Biblical Hebrew, there is always room for rechecking and improving our methods in order to test and, with luck, validate our conclusions.  Compare, in this particular regard, the case of the priestly epithet ‫הכהן הגדול‬, “the high priest,” analyzed in N. M izrahi , “The History and Linguistic Background of Two Hebrew Titles for the High Priest,” JBL 130/4 (2011), 687–705. 50

Thereʼs No Master Key! The Literary Character of the Priestly Stratum and the Formation of the Pentateuch Jakob Wöhrle Introduction The existence of a Priestly stratum of texts in the Pentateuch is one of the oldest and most important insights into the formation of the Hebrew Bible. From the beginning of the critical investigation of the Pentateuch to the most recent times, the distinction between at least two literary strands, Priestly and non-Priestly, is one of the cornerstones of pentateuchal research. However, there is an ongoing debate about the literary character of the Priestly stratum of the Pentateuch.1 At this point there is no consensus as to whether the Priestly passages have to be seen as parts of a formerly independent source that was later compiled with the non-Priestly strata or whether the Priestly passages are a redactional layer that was from the outset written to appear in the context of an otherwise non-Priestly version of the Pentateuch. In the last 150 years of research, since the early challenge to the Documentary Hypothesis by Karl Heinrich Graf,2 the arguments for both assumptions have remained nearly the same. The proponents of the Documentary Hypothesis especially point to the many doublets between the Priestly and the non-Priestly stratum, such as those in the flood story or in the story of the crossing of the sea.

1  See for the following, for example, the overviews in D. M. Carr, Reading the Fractures of Genesis: Historical and Literary Approaches (Louisville: Westminster John Knox, 1996), 43–47; J. C. Gertz, “Tora und Vordere Propheten,” in Grundinformation Altes Testament: Eine Einführung in Literatur, Religion und Geschichte des Alten Testaments (ed. J. C. Gertz; 4th ed.; UTB 2745; Göttingen: Vandenhoeck & Ruprecht, 2010), 236–239; J. Wöhrle, Fremdlinge im eigenen Land: Zur Entstehung und Intention der priesterlichen Passagen der Vätergeschichte (FRLANT 246; Göttingen: Vandenhoeck & Ruprecht, 2012), 12–15; E. Zenger, Einleitung in das Alte Testament (ed. C. Frevel; 8th ed.; Stuttgart: Kohlhammer, 2012), 194–196. 2  K. H. G raf , “Die s. g. Grundschrift des Pentateuchs,” Archiv für wissenschaftliche Erforschung des Alten Testaments 1 (1867–1869), 466–477.

392

Jakob Wöhrle

Additionally, they refer to the fact that the Priestly passages – at least in large parts – can be read as a discrete and continuous entity.3 The proponents of the redaction model, in contrast, mainly point to the narrative gaps within the Priestly stratum. Especially within the ancestral narratives, the Priestly passages – compared to the non-Priestly stratum – lack several important traditions, and often they do not really connect with each other.4 Reevaluating the debate on the literary character of the Priestly stratum, it is noteworthy that the proponents of the two models develop their arguments on the basis of rather global observations from different areas of the Pentateuch. The doublets, which are of importance for the source model, are mainly found within the primeval history and the exodus account. The narrative gaps, to which the proponents of the redaction model refer, are mainly found in the ancestral narratives. Thus, it seems appropriate to examine Priestly passages from the different parts of the Pentateuch separately. The following considerations will therefore at first treat exemplary passages from the ancestral narratives and then from the primeval history and the exodus account.5 3  W. Eichrodt, Die Priesterschrift in der Genesis (Halle: Waisenhaus, 1915); N. Lohfink, “Die Priesterschrift und die Geschichte,” in Congress Volume: Göttingen 1977 (VTSup 29; Leiden: Brill, 1978), 189–225, here 199–200, n. 31; K. Koch, “P – Kein Redaktor!,” VT 37 (1987), 446–467; J. A. Emerton, “The Priestly Writer in Genesis,” JTS 39 (1988), 381–400, here 392–398; Carr, Fractures (see n. 1), 114–120; idem, The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011), 292–297; W. H. C. Propp, “The Priestly Source Recovered Intact?,” VT 46 (1996), 458–478; R. G. Kratz, Die Komposition der erzählenden Bücher des Alten Testaments: Grundwissen der Bibelkritik (Göttingen: Vandenhoeck & Ruprecht, 2000), 247; T. Römer, “Hauptprobleme der gegenwärtigen Pentateuchforschung,” ThZ 60 (2004), 289–307, here 291–292; Gertz, “Tora” (see n. 1), 234–236; J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (New Haven, CT: Yale University Press, 2012), 177–188; Zenger, Einleitung, 194–196. 4  B. D. E erdmans , Alttestamentliche Studien, Vol. 1: Die Komposition der Genesis (Gießen: Töpelmann, 1908); M. Löhr, Der Priesterkodex in der Genesis (BZAW 38; Gießen: Töpelmann, 1924); P. V olz , “P ist kein Erzähler,” in Der Elohist als Erzähler: Ein Irrweg der Pentateuchkritik; An der Genesis erläutert (ed. P. Volz and W. Rudolph; BZAW 63; Gießen: Töpelmann, 1933), 135–142; F. M. Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), 301–321; J. Van Seters, Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975), 279–287; R. R endtorff , Das überlieferungsgeschichtliche Problem des Pentateuch (BZAW 147; Berlin: de Gruyter, 1976), 112–146; E. Blum, Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 420–458; R. A lbertz , Religionsgeschichte Israels in alttestamentlicher Zeit, Vol. 2: Vom Exil bis zu den Makkabäern (GAT 8/2; Göttingen: Vandenhoeck & Ruprecht, 1997), 495–535; B. Ziemer, Abram – Abraham: Kompositionsgeschichtliche Untersuchungen zu Genesis 14, 15 und 17 (BZAW 350; Berlin: de Gruyter, 2005), 283–290, 372–374; W. O swald , Staatstheorie im Alten Israel: Der politische Diskurs im Pentateuch und in den Geschichtsbüchern des Alten Testaments (Stuttgart: Kohlhammer, 2009), 185–203. 5  For the following, see also the comprehensive analyses in Wöhrle, Fremdlinge (see n. 1).

Thereʼs No Master Key!

393

1 The Priestly Passages of the Ancestral Narratives Turning to the Priestly passages in the ancestral narratives, three short passages shall be treated first: Gen 11:31–32; Gen 12:4b, 5; and Gen 13:6. According to the classical form of the Documentary Hypothesis, these three passages within the independent Priestly source once followed directly upon each other and constituted the introduction to the Priestly version of the ancestral narratives.6 The first of these texts reads as follows: Genesis 11:31–32 Terah took his son Abram and his grandson Lot, the son of Haran, and his daughter-inlaw Sarai, his son Abramʼs wife, and they went out with them from Ur Chasdim to go to the land of Canaan; and they came to Haran and settled there. And the days of Terah were 250 years; and Terah died in Haran.

This Priestly text describes the departure of Terah and his family from Ur Chasdim in Gen 11:31. Remarkably, this verse mentions the land of Canaan as the destination of their journey. However, Terah and his family do not take the normal route from Ur to Canaan via Mari and Aleppo.7 Rather, they make a detour via Haran in northern Mesopotamia.8 There in Haran they make a stopover, and they do not leave this place until the death of Terah, mentioned in 11:32. However, the ancestorsʼ detour and their stopover in Haran are not the only noteworthy details in this passage. It is also noteworthy that this Priestly verse uses the geographic term Haran at all. Outside of Gen 11:31–32 and the closely connected Priestly verses 12:4b, 5, the Priestly passages do not use the place name Haran for the eastern territories of the ancestorsʼ relatives. Instead, 6  T. N öldeke , Untersuchungen zur Kritik des Alten Testaments (Kiel: Schwers, 1869), 17–18; M. N oth , Überlieferungsgeschichte des Pentateuch (Stuttgart: Kohlhammer, 1948), 17; C arr , Fractures (see n. 1), 105–106; P ropp , “Priestly Source” (see n. 3), 462; K ratz , Komposition (see n. 3), 240; A. de P ury , “PG as the Absolute Beginning,” in Les dernières rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (ed. T. Römer and K. Schmid; BETL 203; Leuven: Leuven University Press, 2007), 99–128, here 109; Baden, Composition (see n. 3), 179. 7  See N. M. S arna , Genesis (JPS Torah Commentary; Philadelphia: Jewish Publication Society, 1989), 88; P. B erlyn , “The Journey of Terah: To Ur-Kasdim or Urkesh,” JBQ 33 (2005), 73–80, here 78. 8  That Abrahamʼs journey from Ur to Canaan via Haran is a detour has already been recognized by G. von R ad , Das erste Buch Mose: Genesis (10th ed.; ATD 2–4; Göttingen: Vandenhoeck & Ruprecht, 1976), 120; C. W estermann , Genesis 12–36 (BKAT 1/2; Neukirchen-Vluyn: Neukirchener Verlag, 1981), 156; S arna , Genesis (see n. 7), 88, or J. M. Miller, “Ur of the Chaldeans and Interdisciplinary Research,” in Relating to the Text: Interdisciplinary and Form-Critical Insights on the Bible (ed. T. J. Sandoval and C. Mandolfo; JSOTSup 384; London: T&T Clark, 2003), 307–320, here 313. However, these authors do not draw any conclusions from this insight.

394

Jakob Wöhrle

they use the specific Priestly term Paddan Aram (‫)פדן ארם‬.9 The term Haran is documented within the subsequent text of the ancestral narratives only in the non-Priestly verses Gen 27:43; 28:10; 29:4. The previously mentioned peculiarities of Gen 11:31–32 – the detour and the stopover in Haran as well as the fact that these verses use the geographic term Haran – can hardly be explained within an independent Priestly source. In such an independent Priestly source, the intention of the ancestorsʼ detour via Haran would remain unclear, since in the subsequent Priestly text Haran is not mentioned anymore. However, the peculiarities of Gen 11:31–32 can easily be explained if the Priestly passages are a redactional layer written for an otherwise non-Priestly version of the ancestral narratives. Then Gen 11:31–32, with its detour and stopover in Haran, offers a narrative preparation for the subsequent non-Priestly parts of the ancestral narratives, according to which the ancestorsʼ relatives live right there, in Haran.10 The next passage commonly attributed to the Priestly stratum is Gen 12:4b, 5: Genesis 12:4b–5 And Abram (‫ )ואברם‬was seventy-five years old when he departed (‫ )בצאתו‬from Haran. Abram took his wife Sarai, his brotherʼs son Lot, and all their possesions that they had gathered and the people whom they had acquired in Haran and they departed (‫ )ויצאו‬to go to the land of Canaan; and they came (‫ )ויבאו‬to the land of Canaan.

According to the traditional source model, Gen 12:4b, 5 once followed directly upon 11:32.11 Thus, after Gen 11:32 mentioned the death of Terah, Gen 12:4b, 5 would have presented Abrahamʼs departure from Haran and his arrival in Canaan. This assumption is, however, countered by the dating formula documented in Gen 12:4b. Previous research has not adequately taken into account that within the Priestly passages two different types of dating formulas can be distinguished:12

9  The term ‫ פדן ארם‬is documented in Gen 25:20; 28:2, 5, 6, 7; 31:18; 33:18; 35:9, 26; 46:15 (and see Gen 48:7, which has ‫)פדן‬, all Priestly texts. 10  The background of the Priestly description can be seen in the fact that the Priestly authors relocated Abrahamʼs place of origin to “Ur of the Chaldeans.” Since the subsequent non-Priestly text of this account, which was taken up and reworked by the Priestly authors, located the ancestorsʼ eastern relatives in Haran, the Priestly authors were forced to show how Abraham and his relatives on their way to Canaan came to Haran, which only Abraham and his family left, while his relatives stayed there; see Wöhrle, Fremdlinge (see n. 1), 28–30, 176–177. 11  See above, n. 6. 12  Thus already R endtorff , Problem (see n. 4), 131–136; P. W eimar , “Struktur und Komposition der priesterschriftlichen Geschichtsdarstellung,” BN 23 (1984), 81–134, here 106–107.

Thereʼs No Master Key!

Type 1

Type 2

Form

narr. past of ‫ היה‬+ PN + age + narr. past

‫ ו‬+ PN + age + ‫ ב‬+ infinitive

Texts

Gen 11:32; 17:1; 26:34

Gen 12:4b; 16:16; 17:24–25; 21:5; 25:26b; 41:46

395

Priestly dating formula type 1 (with minor variations) begins with the narrative past-tense of the verb ‫היה‬, followed by the name of the ancestor and his age, after which a second narrative past-tense verbal clause describes the event, which happened at this time. Type 2, by contrast, begins with the conjunction ‫ ו‬and the name of the ancestor. It then gives his age, and through a prepositional construction with ‫ ב‬plus infinitive it finally mentions the event, which happened at this time. Remarkably, these two types of dating formulas fulfill different functions. Type 1 introduces a new event that has not been mentioned before. This is true of the death of Terah in Gen 11:32, the theophany in 17:1, and the marriages of Esau in 26:34. Type 2, however, is invariably used to date an event that has already been mentioned. For example, the dating formula in Gen 16:16 dates the birth of Ishmael, mentioned before in 16:15; Gen 17:24–25 dates the circumcision of Abraham and Ishmael, mentioned in 17:23; and Gen 21:5 dates the birth of Isaac, mentioned in 21:3. The distinction between the two different types of dating formulas within the Priestly passages is of importance for understanding Gen 12:4b, since this verse contains a dating formula of type 2. That means, however, that Gen 12:4b cannot be read as introducing the departure from Haran as a new fact that has not been mentioned before. The dating formula in Gen 12:4b rather presupposes that this event has already been introduced and is now subsequently connected with a certain date.13 Thus, the Priestly verse Gen 12:4b cannot be understood, as it often has been, as a direct continuation of Gen 11:32, which mentions the death of Terah. Genesis 12:4b rather depends on the first, non-Priestly, part of this verse, Gen 12:4a, since this part of the verse presents Abrahamʼs departure, which is subsequently connected with a particular date in the Priestly notice Gen 12:4b. Thus, like Gen 11:31–32, the Priestly passage Gen 12:4b, 5 presupposes its non-Priestly context, which is again best explained by the assumption that this Priestly passage was originally written for exactly this non-Priestly context.14  Weimar, “Struktur” (see n. 12), 107, realized that Priestly dating-formula type 2 is used in order to date a previously mentioned event. But Weimar considers Gen 12:4b an exception to this rule and supposes that the dating formula in this verse – part of an independent Priestly source – introduced a new event. This assumption, however, is evidently caused by an unquestioned adoption of the Documentary Hypothesis. 14  Propp, “Priestly Source” (see n. 3), 462, with n. 22, and H. Seebass, Vätergeschichte 13

396

Jakob Wöhrle

The next Priestly verse is Gen 13:6: Genesis 13:6 The land could not support them (‫ )אתם‬that they might dwell together, for their possessions (‫ )רכושם‬were great so that they could not (‫ )יכלו‬dwell together.

The proponents of the Documentary Hypothesis see Gen 13:6 as the direct continuation of Gen 12:4b, 5.15 This assumption, however, is also fraught with problems. It is remarkable that Gen 13:6 does not mention a subject for the plural verband-suffix forms in this verse. This verse does not specify whom the land could not support, whose possessions were too great, and who consequently could not live together. The last verse ascribed to the Priestly stratum, Gen 12:5, however, says that Abraham took Sarai, Lot, and additional persons and that they departed and came to Canaan. The plural forms at the end of this verse thus relate to Abraham, Sarah, Lot, and the other persons. Strictly speaking, after Gen 12:5, the plural forms of 13:6 should also relate to this group of persons, that is, to Abraham, Sarah, Lot, and the other persons. Yet this can hardly be the case. Based upon the subsequent text of Gen 13 – the Priestly as well as the non-Priestly portions – the issue described in Gen 13:6 surely relates to Abraham and Lot. Genesis 13:6 thus cannot be the original continuation of the Priestly verse 12:5. This verse rather presupposes its nonPriestly context, especially the preceding verse 13:5, which mentions only Abraham and Lot. Thus, a close examination of the first three texts of the Priestly stratum of the ancestral narratives shows that these texts can hardly be understood as parts of an originally independent Priestly source. They are rather redactional elements written for the context of an otherwise non-Priestly version of the ancestral narratives. One further example from the Priestly passages in the ancestral narratives shall be treated: the beginning of the Jacob story in Gen 25:19–26. In Gen 25:20, a verse commonly attributed to the Priestly stratum, one finds the following note:

II (23,1–36,43) (Neukirchen-Vluyn: Neukirchener Verlag, 1999), 17, also concede that Gen 12:4b presupposes its preceding context. However, this does not lead them to further conclusions regarding the literary character of the Priestly work. R. Smend, Die Erzählung des Hexateuch: Auf ihre Quellen untersucht (Berlin: Reimer, 1912), 12, and more recently C. Levin, Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993), 140, who also recognize that Gen 12:4b does not really connect with 11:32, suppose that Gen 12:4b is a late, post-Priestly addition to the text. But due to the fact that the dating formula in this verse is identical to that of other Priestly texts, this seems to be a rather unlikely solution, which is only based upon the assumption that the Priestly passages in the ancestral narratives are part of an originally independent source. 15  See above, n. 6.

Thereʼs No Master Key!

397

Genesis 25:20 Isaac was forty years old when he took Rebekah as wife, the daughter of Bethuel, the Aramean of Paddan-Aram, the sister of Laban the Aramean.

Subsequent to this note about the marriage of Isaac and Rebekah, the nonPriestly verses Gen 25:21–26a mention Rebekahʼs barrenness, the overcoming of this state, and the birth of Jacob and Esau.16 The next text commonly ascribed to the Priestly stratum is Gen 25:26b: Genesis 25:26b And Isaac (‫ )ויצחק‬was sixty years old when they were born (‫)בלדת אתם‬.

Genesis 25:26 dates the birth of Isaacʼs children, Jacob and Esau. Remarkably, this verse is formulated with a nominal clause followed by ‫ ב‬plus infinitive. It thus conforms to type 2 of the above-mentioned types of dating formulas. As has been shown, this type is used to date an event that has already been narrated. Thus, Gen 25:26b cannot be the direct continuation of the last verse ascribed to the Priestly stratum, Gen 25:20, which mentions the marriage of Isaac and Rebekah. Genesis 25:26b rather presupposes that in the preceding context the birth of Jacob and Esau has been narrated, which can also be shown by the mere fact that Gen 25:26b does not mention the newborns by name but only with the plural suffix of ‫אתם‬.17 Not without reason, proponents of the Documentary Hypothesis also concede that Gen 25:26b does not connect with 25:20 but rather presupposes a preceding birth notice.18 However, they presume that the Priestly source originally had such a birth notice between 25:20 and 25:26b and that this notice was dropped in the course of the compilation of the Priestly and non-Priestly sources, since the birth of a person can only be mentioned once. Such an omission of a Priestly birth notice does not, however, sufficiently explain the peculiarities of this passage. The concrete statements of age in 25:20  Regarding this classical view on the formation of Gen 25:19–26, according to which 25:19–20, 26b are attributed to the Priestly stratum and 25:21–26a to the non-Priestly stratum, see, for example, N öldeke , Untersuchungen (see n. 6), 26; H. G unkel , Genesis (3rd ed.; HKAT 1/1; Göttingen: Vandenhoeck & Ruprecht, 1910), 385; Noth, Überlieferungsgeschichte (see n. 6), 17; von Rad, Erste Buch Mose (see n. 8), 212; Westermann, Genesis 12–36 (see n. 8), 502; Blum, Komposition (see n. 4), 79; Levin, Jahwist (see n. 14), 200; Carr, Fractures (see n. 1), 108; Seebass, Vätergeschichte II (see n. 14), 271; Baden, Composition (see n. 3), 171. 17  Thus already W estermann , Genesis 12–36 (see n. 8), 502; J. L. S ka , “Genèse 25,19–34 – Ouverture du cycle de Jacob,” in Jacob: Commentaire à plusieurs voix de Gen 25–36; Mélanges offerts à Albert de Pury (ed. J.-D. Macchi and T. Römer; MdB 44; Genève: Labor et Fides, 2001), 11–21, here 14, n. 13. 18  N öldeke , Untersuchungen (see n. 6), 26; G unkel , Genesis (see n. 16), 385; N oth , Überlieferungsgeschichte (see n. 6), 13; von R ad , Erste Buch Mose (see n. 8), 212; Westermann, Genesis 12–36 (see n. 8), 502; Levin, Jahwist (see n. 14), 200; Carr, Fractures (see n. 1), 98; Baden, Composition (see n. 3), 183. 16

398

Jakob Wöhrle

and 25:26b remain remarkable. According to 25:20, Isaac is forty years old when he marries Rebekah. But according to 25:26b, at the time of Jacobʼs and Esauʼs birth he is already sixty years old. This span of twenty years between marriage and birth is best explained by the assumption that the Priestly passages also presuppose that Isaac and Rebekah initially could not have children.19 Thus, more than just a short birth notice must have stood between the Priestly verses 25:20 and 25:26b. There must also have been a reference to the initial childlessness of Isaac and Rebekah and to the overcoming of this situation. That means, however, that within an independent Priestly source, the Priestly version of Jacob and Esauʼs birth would have had precisely the same content as the non-Priestly version in Gen 25:21–26a, which mentions Rebekahʼs barrenness, her pregnancy, and finally the birth of the twins. Then, however, the question arises as to why, following the compilation of the Priestly and non-Priestly sources, between the two Priestly dating formulas of 25:20 and 25:26b nothing of the Priestly version but solely the non-Priestly version has been preserved, even though the Priestly version had more or less the same narrative substance. It is more probable, therefore, that the Priestly verses 25:20 and 25:26b do not stem from an independent Priestly source but were originally written for their non-Priestly context. Comparable observations can be made all over the ancestral narratives. The assumption that the Priestly passages trace back to an originally independent source is problematic nearly everywhere.20 Thus, within the ancestral narratives, the assumption of an originally independent Priestly source is not supported. The Priestly passages do not really connect with each other. They rather presuppose their narrower and broader non-Priestly context. They are thus to be taken as a redactional layer written for the context of an otherwise non-Priestly version of the ancestral narratives.

2 The Priestly Passages of the Primeval History and the Exodus Account Coming from the ancestral narratives, one may find it surprising that the nature of the Priestly stratum in the primeval history and the exodus account is quite different.21 Within the primeval history and the exodus account, two nearly complete narrative strands, a Priestly strand and a non-Priestly strand, can be distinguished. 19  Thus already von Rad, Erste Buch Mose (see n. 8), 212; Seebass, Vätergeschichte II (see n. 14), 269. 20  See Wöhrle, Fremdlinge (see n. 1), 25–146. 21  For the following, see also Wöhrle, Fremdlinge (see n. 1), 150–153.

Thereʼs No Master Key!

399

Within the primeval history it is possible to reconstruct two parallel flood stories (Gen 6–9), a completely preserved Priestly version and a nearly completely preserved non-Priestly version.22 The same applies in the book of Exodus for the stories of the plagues (Exod 7–12) and the crossing of the sea (Exod 14).23 Remarkably, in these parts of the Pentateuch even the smallest narrative details are preserved twice. For example, the flood story in Gen 6–9 gives two nearly identical commands to enter the ark: Genesis 6:18b (P) You shall go into the ark, you, your sons, your wife, and your sonsʼ wives with you. Genesis 7:1 (non-P) Go into the ark, you and all your household.

To give just one more example, within the story about the crossing of the sea in Exod 14, the notice that the waters of the sea came back appears twice: Exodus 14:27 (P) [. . .] The sea returned to its bed when the morning appeared [. . .] Exodus 14:28 (non-P) The waters of the sea returned [. . .]

Such doublets of the smallest details strongly speak against the assumption that the Priestly passages can be understood as a redactional layer. There is no plausible explanation for why a Priestly redactor should have added to the flood story a second request to enter the ark or why he should have added to the story about the crossing of the sea a second notice about the returning waters.24 22  See, for example, G unkel , Genesis (see n. 16), 137–140; C arr , Fractures (see n. 1), 76–77; M. W itte , Die biblische Urgeschichte: Redaktions- und theologiegeschichtliche Beobachtungen zu Genesis 1,1–11,26 (BZAW 265; Berlin: de Gruyter, 1998), 325–331; J. C. Gertz, “Beobachtungen zum literarischen Charakter und zum geistesgeschichtlichen Ort der nichtpriesterschriftlichen Sintfluterzählung,” in Auf dem Weg zur Endgestalt von Genesis bis II Regum; Festschrift Hans-Christoph Schmitt zum 65. Geburtstag am 11.11.2006 (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 41–57; idem, “Source Criticism in the Primeval History of Genesis: An Outdated Paradigm for the Study of the Pentateuch?,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr, 2011), 169–180; idem, “The Formation of the Primeval History,” in The Book of Genesis: Composition, Reception, and Interpretation (VTSup 152; Leiden: Brill, 2012), 107–135. 23  See L. S chmidt , Studien zur Priesterschrift (BZAW 214; Berlin: de Gruyter, 1993), 1–34; F. Kohata, Jahwist und Priesterschrift in Exodus 3–14 (BZAW 166; Berlin: de Gruyter, 1986), 93–126; W. H. Schmidt, Exodus 7–24 (BKAT 2/2; Neukirchen-Vluyn: Neukirchener Verlag, 1995ff.), 335–468; J. C. G ertz , Tradition und Redaktion in der Exoduserzählung: Untersuchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000), 74–232. 24  This observation in itself strongly speaks against those approaches that see the Priestly passages in the primeval history or in the exodus account as a redactional layer written for an

400

Jakob Wöhrle

In fact, it can be shown in detail that the Priestly passages in the primeval history and the Priestly passages in the exodus account were not written for their non-Priestly context. This is evident, for example, from the Priestly verses Gen 7:11 and 7:13, which now surround the non-Priestly verse 7:12: Genesis 7:11 (P) In the six hundredth year of Noahʼs life, in the second month, on the seventeenth day (‫ )בשבעה עשר יום‬of the month, on that day (‫)ביום הזה‬, all the fountains of the great deep were burst through and the windows of heaven opened. Genesis 7:12 (non-P) And the rain was on the earth forty days (‫ )ארבעים יום‬and forty nights. Genesis 7:13 (P) On the very same day (‫ )בעצם היום הזה‬Noah, his sons, Shem, Ham, and Japheth, and Noahʼs wife and the three wives of his sons came into the ark.

The Priestly verse Gen 7:11 connects the beginning of the flood with a certain day of Noahʼs life. The Priestly verse Gen 7:13, with its formula ‫בעצם היום הזה‬, “on the very same day,” refers back to this and states that Noah and his family now, on this day, went into the ark. The non-Priestly verse Gen 7:12, mentioning a time span of forty days of rain from the beginning of the flood to the flood itself, then disrupts the solid connection between the Priestly verses 7:11 and 7:13.25 Thus, it seems that the Priestly verses 7:11 and 7:13 originally connected with each other and were secondarily broken up by the non-Priestly verse 7:12. Within the book of Exodus, the beginning of the story about Israelʼs departure from Egypt is also noteworthy. The non-Priestly verse Exod 12:37 gives the following notice:

otherwise non-Priestly version of these parts of the Pentateuch; see, for the primeval history, for example, V olz , “P ist kein Erzähler” (see n. 4), 140–142; C ross , Canaanite Myth (see n. 4), 301–307; R. Rendtorff, “Lʼhistoire biblique des origines (Gen 1–11) dans le contexte de la rédaction ‘sacerdotale’ du Pentateuque,” in Le Pentateuque en question: Les origines et la composition des cinq premiers livres de la Bible à la lumière des recherches récentes (ed. A. de Pury; MdB 19; Geneva: Labor et Fides, 1989), 83–94; E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 278–285; J. Van Seters, Prologue to History: The Yahwist as Historian in Genesis (Zürich: Theologischer Verlag, 1992), 160–165, and for the exodus account C ross , Canaanite Myth (see n. 4), 307–321; B lum , Studien zur Komposition des Pentateuch, 232–262; R. A lbertz , “Der Beginn der vorpriesterlichen Exoduskomposition (KEX): Eine Kompositions- und Redaktionsgeschichte von Ex 1–5,” TZ 67 (2011), 223–262; idem, Exodus, Vol. 1: Exodus 1–18 (ZBK 2/1; Zürich: Theologischer Verlag, 2012), 21–22; C. B erner , Die Exoduserzählung: Das literarische Werden einer Ursprungslegende Israels (FAT 73; Tübingen: Mohr, 2010); H. Utzschneider and W. Oswald, Exodus 1–15 (Internationaler exegetischer Kommentar zum Alten Testament; Stuttgart: Kohlhammer, 2013), 50–52. 25  See Carr, Fractures (see n. 1), 56: “To find the day to which 7:13 is referring, one must jump in the present form of the text from 7:13 back to the exact dating in 7:11.”

Thereʼs No Master Key!

401

Exodus 12:37 (non-P) The Israelites set out from Rameses for Succoth, about six hundred thousand men on foot, besides children.

Within the subsequent text, the Priestly verses Exod 12:40–41 state: Exodus 12:40–41 (P) The time that the Israelites spent in Egypt was four hundred and thirty years. And it came to pass (‫ )ויהי‬at the end of the four hundred and thirty years, on that very same day, it came to pass (‫ )ויהי‬that all the armies of YHWH went out from the land of Egypt.

Remarkably, Exod 12:41 dates the Exodus with a dating formula that is close to dating-formula type 1, mentioned above. It begins with the narrative past-tense form of ‫היה‬. Due to the context, instead of the name of a person and his age, a reference to the time in Egypt follows. Finally, with a second narrative pasttense verb, the event to be dated is mentioned. Interestingly enough, as shown above, dating-formula type 1 is consistently used in order to introduce a new, previously unmentioned event or fact. Exodus 12:41 thus introduces the departure from Egypt as such a new, previously unmentioned event. That means, however, that the non-Priestly departure notice in 12:37 is not presupposed in this verse. All in all, in the primeval history and the exodus account, the Priestly passages build complete or nearly complete narratives. Here, one finds doublets between the Priestly and the non-Priestly stratum of even smallest narrative details. And here, it can be shown that the Priestly passages do not presuppose their non-Priestly context. Thus, here, the Priestly passages indeed seem to be parts of an originally independent source.

3 Conclusions Regarding the Priestly passages of the Pentateuch, the findings in the ancestral narratives on the one hand and in the primeval history and the exodus account on the other are quite different. In the ancestral narratives it is not possible to reconstruct a coherent Priestly narrative. The Priestly texts mostly do not connect to each other. And they obviously presuppose the narrower or broader non-Priestly context. In the primeval history and in the exodus account, in contrast, it is possible to reconstruct a nearly complete Priestly strand and a nearly complete non-Priestly strand. Here, one finds doublets of the smallest narrative details. And it can be shown that the Priestly passages do not presuppose their non-Priestly context. These different findings lead to the conclusion that the juxtaposition of Priestly and non-Priestly texts in the ancestral narratives on the one hand and in the primeval history and the exodus account on the other hand cannot be explained by one and the same model. It is simply impossible that all parts of the

402

Jakob Wöhrle

Pentateuch are to be explained as a compilation of a Priestly source and a nonPriestly source. This assumption would mean that one and the same compiler, who in the primeval history and the exodus account preserved even the smallest details from both sources, in the ancestral narratives – for example, in the birth story of Jacob and Esau – was willing to make comprehensive omissions. Additionally, the fact that the Priestly passages of the ancestral narratives obviously presuppose their narrower and broader non-Priestly context strongly speaks against this assumption. It is, however, also not possible that the Priestly passages are to be explained as a redactional layer in all parts of the Pentateuch. The two nearly complete and independent narrative strands in the primeval history and the exodus story as well as the fact that the Priestly passages in these parts of the Pentateuch do not presuppose their non-Priestly context speak against this view. Thus, the juxtaposition of the Priestly and non-Priestly texts has to be explained by a differentiated model. In the primeval history and in the exodus account it has to be explained as a compilation of two previously independent sources, whereas in the ancestral narratives the Priestly passages have to be explained as a redactional layer written for an otherwise non-Priestly version of the ancestral narratives. For the formation of the Pentateuch, this leads to the conclusion that the Priestly authors formulated their own independent primeval history, to which they added the ancestral narratives, which they composed by reworking older, non-Priestly traditions. Finally, probably from Exod 6 on,26 they formulated their own independent exodus story. Later, the originally independent non-Priestly primeval history and the originally independent non-Priestly exodus story were integrated into this work. These considerations align with the assumptions of Konrad Schmid, Jan Gertz, and Erhard Blum, according to whom the primeval history, the ancestral narratives, and the exodus account were originally independent traditions and according to whom these traditions were for the first time combined in the Priestly work.27 They differ, however, from previous research insofar as different models are used to explain the juxtaposition of Priestly and non-Priestly traditions in the different parts of the Pentateuch.28  See Wöhrle, Fremdlinge (see n. 1), 153–158.  K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 358–359; Gertz, Tradition (see n. 23), 380–388; E. Blum, “Die literarische Verbindung von Erzvätern und Exodus: Ein Gespräch mit neueren Endredaktionshypothesen,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 119–156. 28  See, however, Gertz, Tradition (see n. 23), 391, who, but rather cautiously, considers the possibility that the Priestly passages of the Pentateuch could be understood as a redactional layer in the ancestral narratives but as a source in the primeval history and the exodus account. 26

27

Thereʼs No Master Key!

403

Thus, there is no master key for explaining the formation of the Pentateuch. Further research should therefore refrain from global models explaining all parts of the Pentateuch with the same approach and rather look for differentiated models that take into account the different forms of the different parts of the Pentateuch.

Oral Platform and Language Usage in the Abraham Narrative Frank H. Polak This study argues that the tales of the patriarchs preserve residues of an underlying oral-epic substratum, an overarching platform that is situated in the tradition stream of the Northwest Semitic epic and narrative poetry and in which the narrative in its present, written form is anchored. The orality of this substratum is indicated, in my view, by a number of stylistic features that are close to the characteristics of spoken language and by a series of formulaic phrases. By contrast, linguistic characteristics of the Deuteronomic/Priestly/postexilic strata are far less frequent. By epic I do not mean a Nationalepos from a so-called heroic age but only an extended narrative in poetic form, concerning noble figures such as great heroes or esteemed ancestors, worthy of the “fame of women and men.” This delineation is intended to be purely formal, without any essentialist intentions or historic pretensions. The term overarching platform indicates a narrative complex that is known to and used by different narrators, “singers of tales,” in different variations but all around the same central theme. As I stated in a discussion of the Abraham-Jacob narrative: “What is traditional, and ultimately based on oral narrative/poetry, is the underlying unity of the overarching narrative, whereas the problems of repetition and contradiction originate with the activity of various different narrators, whether oral or writing, within this tradition.”1 I will focus on the Abraham narrative,2 in view of the features matched by Ugaritic epic poetry, as representative of the Northwest Semitic epic tradition, and because of its coherent plot structure as analyzed by Heinrich Ewald. The thematic relationship between the Ugaritic epic and the Abraham narrative (1.1 and 1.2) will be sharpened by the syntactic-stylistic profile of the Abraham tales (1.3) and significant residues of oral-formulaic phraseology that 1  See F. P olak , “Oral Substratum, Stylistic-Syntactic Profile and Thematic Flow in the Abraham-Jacob Narrative,” in Contextualizing Israelʼs Sacred Writings: Ancient Literacy, Orality, and Literary Production (ed. B. Schmidt; Atlanta: Society of Biblical Literature, 2015), 217–238, here 217. 2  When viewing the various tales in Gen 11:27–25:10 in a piecemeal perspective, I will speak of tales or stories, whereas I will reserve the term narrative for the Abraham story as an epic platform.

406

Frank H. Polak

stand out in some of the tales (2.1 and 2.2). An important part of this discussion will concern the methodological problems surrounding the role of linguistic characterization for literary stratification and periodization (2.3–4 and 3–4). In the end I will point to some phenomena that may still betray the narratorʼs voice and gesturing (5) and thus give the sōpēr – both scribe and author – a place within the oral arena. It is to be granted that it is impossible to reconstruct the oral traditions in the background of biblical texts as they were committed to writing and transmitted in written form.3 It is, however, my intention to show that the oral background has affected the written form of the Abraham narrative to a very large extent.

1 From Ewald to Ilimilku: Plot Structure and Encompassing Epic 1.1 Plot Structure: Ancient and Modern Here I invoke two names. I invoke the name of Ewald in honor of his insight that patriarchal narrative embodies a literary plan, a plot structure in my terms. Thus he characterizes the tale of Abraham: Zuerst also ist Sarah unfruchtbar und das Versprechen ist umsonst gegeben. Aber je höher hier der Erzähler durch Schürzung des verworrenen Knoten alles spannt, und je absichtlicher er alles aufsucht, die Unmöglichkeit dieses Versprechens zu zeigen, und je weiter er die Erfüllung selbst hinaus schiebt, desto überraschender ist endlich das erfüllte Versprechen. Abrahams Treue soll durch den Verzug und das zwar oft versprochene aber noch immer ferne Ziel geprüft werden (15,6. ‫וְ ֶה ֱא ִמן ַבּיהוָ ה וַ יַּ ְח ְשׁ ֶב ָה לֹּו‬ ‫)צ ָד ָקה‬, ְ und je höher das alles durch das feierliche Hinhalten und Zögern gestellt wird, desto kräftiger ist der Erfolg.4

In modern biblical criticism, the view proposed in the youthful work of this towering Hebraist and orientalist must raise the question whether one has the right to postulate an encompassing plot structure of this type. Ever since the work of Gunkel, precedence has been given to the small unit.5 In this vista an  Recently, see A. S chellenberg , “A ‘Lying Pen of the Scribes’ (Jer 8:8)? Orality and Writing in the Prophetic Books,” in The Interface of Orality and Writing: Speaking, Seeing, Writing in the Shaping of New Genres (ed. A. Weissenrieder and R. B. Coote; Tübingen: Mohr Siebeck, 2010), 285–309, here 286. 4  G. H. A. Ewald, Die Komposition der Genesis kritisch untersucht (Braunschweig: Ludwig Lucius, 1823), 272. In modern biblical research, see in particular L. R. H elyer , “The Separation of Abram and Lot: Its Significance in the Patriarchal Narratives,” JSOT 26 (1983), 77–88, here 80–81. For the comments of Naomi Steinberg, David Clines, Joel Baden, and David Carr, see n. 18, below. 5  See J. Van S eters , Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975), 158–164, 310–311, and D. M. C arr , Reading the Fractures of Genesis: 3

Oral Platform and Language Usage in the Abraham Narrative

407

encompassing plot structure necessarily is attributed to the redaction strata,6 or to a late narrator, as proposed by John Van Seters and, in a sense, Ronald Whybray.7 David Carr recognizes the role of the overarching narrative but places it in the framework of a later stage in the redaction-transmission process (Proto-Genesis, between 722 and 560 BCE).8 At this juncture I want to invoke the name of Ilimilku, who is mentioned as the copyist/author of the epics of Kirta, ʾAqhat and Baʿal.9 Ilimilkuʼs work exemplifies the form of a large-scale epic text in prosody that resembles the diction of Hebrew biblical poetry and language that is cognate to Biblical Hebrew. These epics prove that in the context of the Levant an extended narrative is not necessarily the product of a late redaction.10 So much has been argued by Frank

Historical and Literary Approaches (Louisville, KY: Westminster John Knox, 1996), both of whom follow H. Gunkel, Genesis (HKAT 1/1; 3rd ed.; Göttingen: Vandenhoeck & Ruprecht, 1910 [1st ed., 1901]). 6  See in particular J. L. Ska, The Exegesis of the Pentateuch: Exegetical Studies and Basic Questions (Tübingen: Mohr Siebeck, 2009), 27–29. Ska highlights the redactional chronology, covering a hundred years from Abrahamʼs migration until his death, but does not find “a single chain of events.” He regards the posterity theme as one of two “narrative programmes,” whereas Ewald presents it as the central issue par excellence. On the stylistic characterization of the introductory sections, 11:27–12:5, see below, pp. 413–415. 7  R. N. Whybray, The Making of the Pentateuch: A Methodological Study (JSOTSup 53; Sheffield: Sheffield Academic Press, 1987), 232–242. 8  C arr , Fractures (see n. 5), 201–203, 226–229; idem , An Introduction to the Old Testament: Sacred Texts and Imperial Contexts of the Hebrew Bible (Oxford: Blackwell, 2010), 37–39. An aperçu of recent scholarship is offered by T. Römer, “Zwischen Urkunden, Fragmenten und Ergänzungen: Zum Stand der Pentateuchforschung,” ZAW 125 (2013), 2–24, here 9–18; idem , “Recherches actuelles sur le cycle dʼAbraham,” in Studies in the Book of Genesis: Literature, Redaction and History (ed. A. Wénin; BETL 155; Leuven: Leuven University Press, 2001), 179–211. 9  D. Pardee , The Ugaritic Texts and the Origins of West Semitic Literary Composition (Schweich Lectures 2007; Oxford: Oxford University Press, 2012), 42–49; A. C urtis , “Ilimilku of Ugarit: Copyist or Creator?,” in Writing the Bible (ed. P. R. Davies and T. Römer; London: Acumen, 2013), 10–24; M. J. Korpel, “Exegesis in the Work of Ilimilku of Ugarit,” in Intertextuality in Ugarit and Israel (ed. J. C. de Moor; OTS 40; Leiden: Brill, 1998), 86–111. 10  For epigraphic literary texts from the southern Levant, specifically from Kuntillet ʿAjrud and Tell Deir ʿAlla, see S. Aḥituv, E. Eshel, and Z. Meshel, “The Inscriptions,” in Z. Meshel, Kuntillet ʿAjrud (Ḥorvat Teman): An Iron Age II Religious Site on the Judah-Sinai Border (Jerusalem: Israel Exploration Society, 2012), 73–142, here 108–113; P. K. McCarter, “Kuntillet ʿAjrud: Plaster Inscription (2.47D),” in The Context of Scripture, Volume II: Monumental Inscriptions from the Biblical World (ed. W. W. Hallo and K. Lawson Younger; Leiden: Brill, 2003), 173; J. A. H ackett , The Balaam texts from Deir ʿAlla (Chico, CA: Scholars Press, 1984); B. A. Levine, “The Deir ʿAlla Plaster Inscriptions (2.27),” in The Context of Scripture, Volume II: Monumental Inscriptions from the Biblical World (ed. W. W. Hallo and K. Lawson Younger; Leiden: Brill, 2003), 140–145.

408

Frank H. Polak

M. Cross, Ronald Hendel, and Simon Parker,11 but their protestations have not been heeded in most recent scholarship. However, the similarities between patriarchal narrative and the texts found in Ugarit d close to one of the the central themes o not suffer explaining away. Significantly, two Ugaritic epics, the tales of Kirta and Danʾilu-ʾAqhat, center on the theme of the sonless/childless king and his prayer (Danʾilu) or lament (Kirta), a theme that is close to one of the central themes of the narrative of Abraham and Sarah, who remained childless (Gen 11:30).12 A further common feature is the divine blessing and promise of posterity. Two details merit particular attention. First, as noted by Claus Westermann, the dream (‫מ ֲחזֶ ה‬, ַ Gen 15:1)13 in which Abram complains of the lack of a son (vv. 2–3) is matched by the dream in which Kirta gives vent to his despair to Ilu, his divine father. Second, the theme of the tale of Mamre, the divine visit to the human couple and the supper prepared for the guest, is found in the Ugaritic epic of ʾAqhat: the divine craftsman, Koṯar wa-ḫasis, visits Danʾilu and his wife in order to present the newborn ʾAqhat with a bow and arrows. This scene, as in the tale of Mamre, highlights the role of the woman in the preparation of the meal and is likewise connected to the birth theme. If Sarah and Abraham are rewarded with the promise of a son, then in the ʾAqhat tale the son receives his bow. This parallel has often been noted14 but is to be 11  F. M. Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), viii–ix; idem, “The Epic Traditions of Ancient Israel: Epic Narrative and the Reconstruction of Early Israelite Institutions,” in The Poet and the Historian: Essays in Literary and Historical Biblical Criticism (ed. R. E. Friedman; Chico, CA: Scholars Press, 1983), 13–39, here 14–16; idem , From Epic to Canon: History and Literature in Ancient Israel (Baltimore: Johns Hopkins University Press, 1998), 22–52; followed by S. B. Parker, The Pre-Biblical Narrative Tradition: Essays on the Ugaritic Poems Keret and Aqhat (Atlanta: Scholars Press, 1988); R. H endel , The Epic of the Patriarch: The Jacob Cycle and the Narrative Traditions of Canaan and Israel (Harvard Semitic Monographs 42; Atlanta: Scholars Press, 1987), 26–32, 39–59. 12  K. Koch, “Die Sohnesverheißung an den Ugaritischen Daniel,” ZA 58 (1967), 211–221, here 221; C. Westermann, The Promises to the Fathers (trans. D. E. Green; Philadelphia: Fortress, 1980), 132–134, 167–180 (with detailed comparison of the narrative patterns); Hendel, Epic of the Patriarch (see n. 11), 39–59. The thematic details discussed by Westermann are entirely disregarded by Van Seters, who considers childbirth tales ubiquitous: J. Van Seters, Prologue to History: The Yahwist as Historian in Genesis (Louisville, KY: Westminster John Knox, 1992), 247, 271–272, n. 20. 13  M. Lichtenstein, “Dream Theophany and the E Document,” JANES 1 (1969), 45–54. 14  See particularly Y. Avishur , Studies in Biblical Narrative: Style, Structure and the Ancient Near Eastern Literary Background (Tel Aviv-Jaffa: Archeological Center Publication, 1999), 57–74. In view of the Orion myth presented in Ovidʼs Fasti (V 495–544), T. Römer places this tale in the Persian era, which witnesses the meeting of Greek and Levantine culture. See T. Römer, “Quand les dieux rendent visite aux hommes (Gen 18–19),” in Dans le laboratoire de lʼhistorien des religions: Mélanges offerts à Philippe Borgeaud (ed. F. Prescendi and Y. Volokhine; Geneva: Labor et Fides, 2011), 615–626, here 617–620, 626. However, the ambiguity surrounding divine/angelic participation in a meal prepared and presented by

Oral Platform and Language Usage in the Abraham Narrative

409

mentioned again because of the many epic formulae encountered in the tale of Mamre.15 1.2 The Abraham Narrative: The Growth of a Symbolic Household In the epic vista, the plot of the Abraham narrative, episodic though it be, reveals a basic unity. All episodes are subordinate to the contrast between the promise to Abraham and the fundamental problem indicated by Ewald: since Sarah was childless, the divine promise to Abraham seems pointless. Since the promise to Abraham is often presented as a late theological motif,16 some comments are called for. First, in view of the development of the narrative, the divine promise of “a great ethnos” (12:2) remains problematic, since the motif of childlessness lies at the basis of a number of narratives. This motif, made explicit by Abrahamʼs complaint (15:2–3), is the motivation for the role of Hagar (16:1–2), is stated again in the tale of Mamre (18:11), and is implied by the reactions of Abraham and Sarah to the promise of a son (17:17; 18:12). The divine promise, then, augments the tension created by the note on Sarai (11:30) and should be viewed as classical case of literary prolepsis,17 foreshadowing a future perspective but not so transparent that one may know what is going to happen. Only after the narration of the event itself can the reader attain full comprehension of the meaning of the divine announcement. Thus the opening of the Abraham narrative should not be viewed primarily as a theological statement. It is the indication of a perspective and at the same time a way to create an insuperable tension: when is the promise to come true?18

humans hardly fits this period (contrast Judg 6:20–21; 13:16–20). The Greek and Levantine cultures meet in Anatolia, Crete, and Cyprus in place, in time by the late Bronze Age, and in writing by the Greek borrowing of the Phoenician alphabet. 15  See Polak, “Abraham-Jacob Narrative” (see n. 1), 229, 233–234. 16  J. L. S ka , Exegesis (see n. 6), 47–66. The problems concerning the command/nonexecution pattern, raised by Ska, were already solved by W. B aumgartner , “Ein Kapitel vom hebräischen Erzählungsstil,” in Eucharisterion: Studien zur Religion und Literatur des Alten und Neuen Testaments Hermann Gunkel zum 60. Geburtstage (ed. H. Schmidt; 2 vols.; FRLANT 36; Göttingen: Vandenhoeck & Ruprecht, 1923), 1:145–157, here 146–148. When a request, proposal, or command is not followed by the mention of its execution, the execution is implied, as in, e.g., Gen 12:13. 17  The role of the prolepsis as roter Faden is discussed by E. L ämmert , Bauformen des Erzählens (3rd ed.; Stuttgart: Metzler, 1968), 139–143, 146–47; see also S. B ar -E frat , Narrative Art in the Bible (Sheffield: Almond Press, 1989), 179–180. 18  D. J. A. C lines , The Theme of the Pentateuch (JSOTSup 10; Sheffield: Sheffield Academic Press, 1978), 45; Helyer , “Separation” (see n. 4); W estermann , Promises (see n. 12), 132–134; N. Steinberg, Kinship and Marriage in Genesis: A Household Economics Perspective (Minneapolis, MN: Fortress, 1993), 36, 45–48; Carr, Fractures (see n. 5), 184– 185; J. S. Baden, The Promise to the Patriarchs (Oxford: Oxford University Press, 2013), 14.

410

Frank H. Polak

This is the problem of opening, closure, and main stages of the Abraham narrative. As the narrator notes in a powerful bicolon:19 “Now Sarai was barren; she had no child” (11:30).20 The closure of the Abraham narrative signals the fundamental change in the fate of the patriarchal family: at this juncture their son, Isaac, lets his new bride, Rebecca, enter into his maternal tent (24:67). These developments introduce two of the central cornerstones of the Abraham narrative: wife and son (or in a wider vista, posterity). Two additional cornerstones are introduced by the episode of the divine promise and the trek: a land that Abraham is to move into (12:1) and the property he is taking with him (12:5).21 These cornerstones define the house as a constitutive element of the symbolic universe of this tale and thus as a symbolic household.22 In biblical scholarship, these issues are subsumed under the themes of the promise of land and posterity. However, in patriarchal narrative these matters are fully integrated into and dependent on the development of the tale and should not be taken as separate entities. The divine promise near the terebinth of Moreh and the ensuing construction of the altar are dependent on and subservient to Abrahamʼs arrival (12:7).23 The promise of land and posterity following Lotʼs departure (13:14–17) is related to the isolation caused by the break up. The double promise in Gen 15 is the response to Abrahamʼs questions and doubts concerning the divine assurances (15:3–5, 8). These doubts, which, as already noted, are matched by Kirtaʼs complaints in the Ugaritic epic, are an expression of the patriarchʼs abandonment and as such are to be viewed as an essential part of a large-scale narrative.24 The formulation of the double promise in the Lot  On this noteʼs linguistic features and parallelism, see below, p. 415.  For the attribution of 11:30 to the Priestly genealogical framework, see Ska, Exegesis (see n. 6), 48–49; J. W öhrle , Fremdlinge im eigenen Land: Zur Entstehung und Intention der priesterlichen Passagen der Vätergeschichte (FRLANT 246; Göttingen: Vandenhoeck & Ruprecht, 2012), 25–38; against this view, see J. A. E merton , “The Source Analysis of Genesis xi 27–32,” VT 42 (1992), 37–46, here 44–45. The stylistic shaping and narrative patterns of 11:29–30, discussed below, p. 415, do not have a match in this framework. 21  D. Steinmetz, From Father to Son: Kinship, Conflict and Continuity in Genesis (Louisville, KY: Westminster John Knox, 1991), 5–7, 10–11, 15–30; R. Polzin, “ ‘The Ancestress of Israel in Danger’ in Danger,” Semeia 3 (1975), 81–98, here 83–85. 22  For this vista, see Steinberg, Kinship and Marriage (see n. 18). 23  A similar view is urged by B aden , Promise (see n. 18), 57–58. Since Shechem is the “first-visited” place in the Abraham narrative, one has to assume that this narrative was primarily contemplated and composed in the north (see also 12:8), although its present form centers on Judean Hebron and Beer-sheba. 24  Against R. G. K ratz , Die Komposition der erzählenden Bücher des Alten Testaments (Göttingen: Vandenhoeck & Ruprecht, 2000), 263–264. The example from the Kirta narrative shows conclusively that the raising and answering of doubts are constitutive elements of narrative and should not be discarded in the way proposed by Ska. See S ka , Exegesis (see n. 6), 67–81, here 71–72; M. Köckert, “Gen 15: vom ‘Urgestein’ der Väterüberlieferung zum ‘theologischen Programmtext’ der späten Perserzeit,” ZAW 125 (2013), 25–48, here 28–30. 19

20

Oral Platform and Language Usage in the Abraham Narrative

411

narrative (13:14–17) may have served to answer these doubts in advance, on a redactional level. The narrative in all its stages is related to gain and loss in terms of these cornerstones, in accordance with David Clinesʼs insight that in the Pentateuch the promise is never realized in full and is still awaited with Mosesʼs death.25 Gain made in one element leads to a loss in another corner, and vice versa. Thus, when Abraham arrives in Canaan, a famine compels him to move to Egypt (loss of land), where his wife is taken to Pharaohʼs court (endangering his wife), but he receives compensation in terms of property (12:16). With his return to Canaan as a wealthy man, the household seems complete, but now his nephew, who symbolizes posterity, abandons him. The final promise of a son at the terebinths of Mamre (18:9–15; contrasted by the destruction of Sodom and Gomorrah and Lotʼs flight) is offset by the famine that sends Abrahamʼs household to Gerar (ch. 20), but his losses are counterbalanced by the gain in property in the form of the compensation Sarah receives. This narrative forms a parallel to the tale of the stay in Egypt (12:10–20) and thus demonstrates the absorption of doublets and narrative expansions within the overall structure. By the same dynamics, the achievement signaled by the birth of Isaac is offset by the expulsion of Hagar and her son, and an even greater threat in the form of the divine command to sacrifice Isaac as an “offering-up” (22:3), which would put an end to all hope.26 The perpetual balancing of gain and loss in the symbolic household, until the final success when the son marries a woman of valor who belongs to Abrahamʼs family in Upper Mesopotamia, suggests an overarching framework rather than the mere accumulation of independent tales or their creative redaction. This is the structure of the basic epic platform that was common to many storytellers, who recounted the narrative in different ways. In my view, then, this platform is anchored in the oral performance by the singer of tales, or at least the artist of oral literature.27 Important residues of orality are revealed by a number of syntactic and stylistic features that dominate most episodes of the Abraham tale. The ceremonies described in 15:9–10, 17 contain a series of elements that are foreign to pentateuchal ritual; the male calf of Jer 34:18 is quite unlike the three-year-old heifer, goat, and ram mentioned in Gen 15:9. 25  Clines, Theme (see n. 18), 45–47, 106–112, 117. Clinesʼs thesis has been taken up again by Baden, Promise (see n. 18), 17–22. 26  See S teinmetz , Kinship (see n. 21), 82–85; C lines , Theme (see n. 18), 45; E. B lum , “Abraham,” in Religion Past and Present (ed. H. D. Betz et al.; 14 vols.; Leiden: Brill, 2007–2013), 1:11–13, here 12a. 27  The artistry of oral narrative is demonstrated by a number of important studies, e.g., L. D égh , Folktales & Society: Story-Telling in a Hungarian Peasant Community (Bloomington: Indiana University Press, 1989), 165–285; R. F innegan , Oral Literature in Africa (Oxford: Clarendon Press, 1970), 2–22, 82–108, 243–269; G. Henssen , Überlieferung und Persönlichkeit: Die Erzählungen und Lieder des Egbert Gerrits (Münster: Aschendorffsche Verlagsbuchhandlung, 1951), 9–42; I. Okpewho, The Epic in Africa: Toward a Poetics of the

412

Frank H. Polak

1.3 Narrative Style, Scribal Desk, and Spoken Discourse: A Typology The Abraham narrative contains episodes in two different styles that can be diagnosed by means of syntactic analysis.28 One group of tales is characterized by long noun groups, a relatively large number of subordinate clauses, and, in general, elaborate sentence structure that establishes not only subject and object but also, for instance, beneficiary (or addressee) and/or coordinates of place or time, for example: Genesis 23:9b

‫ת־ק ֶבר‬ ָ ַ‫תֹוכ ֶכם ַל ֲא ֻחזּ‬ ְ ‫ְבּ ֶכ ֶסף ָמ ֵלא יִ ְתּנֶ נָּ ה ִלי ְבּ‬ For the full silver worth let him give it to me in your midst for a burial holding.

This elaborate clause includes an indirect object, indicating the price (‫ְבּ ֶכ ֶסף‬ ‫)מ ֵלא‬, ָ and an indication of the beneficiary (‫)לי‬, ִ of the place (‫תֹוכ ֶכם‬ ְ ‫)ּב‬, ְ and of ַ We note two noun groups: ‫ּב ֶכ ֶסף ָמ ֵלא‬, ְ and ‫ת־ק ֶבר‬ ָ ַ‫ל ֲא ֻחּז‬.ַ An the aim (‫ת־ק ֶבר‬ ָ ַ‫)ל ֲא ֻחּז‬. example for a subordinate clause is as follows: Genesis 25:10

‫אַב ָר ָהם וְ ָשׂ ָרה ִא ְשׁתֹּו‬ ְ ‫י־חת ָשׁ ָמּה ֻק ַבּר‬ ֵ ֵ‫אַב ָר ָהם ֵמ ֵאת ְבּנ‬ ְ ‫ר־קנָ ה‬ ָ ‫ַה ָשּׂ ֶדה ֲא ֶשׁ‬ The field that Abraham had purchased from the Hittites, there Abraham was buried, with his wife Sarah.

These examples are instances of the intricate, elaborate style (IES), with around a third (or slightly less) of all clauses of the unit in syntactic subordination, with as many long clauses (for instance, containing both explicit subject and explicit object), and with a noun pair in almost every clause (in the mean, around 80–90 percent of all clauses).29 The intricacies of the IES fit the typology of written language. The complexities of the syntactic constructions indicate an environment that facilitates detailed planning, rereading, and correction, while the frequency of long noun phrases indicates a predilection for exhaustive precision. These features place the IES in the milieu of the learned scribe, educated in the methods and norms of the official bureaucracy, who has acquired the know-how to plan contracts Oral Performance (New York: Columbia University Press, 1975), 135–239; W. Derks, The Feast of Storytelling: On Malay Oral Tradition (Jakarta: Pusat Pembinaan dan Pengembangan Bahasa, 1994), 583–613; W. S. Walker, A Turkish Folktale: The Art of Behçet Mahir (New York: Garland, 1996), xviii–xxvi. Thus in nonliterate societies, or in societies with restricted literacy, oral narrative is a form of art, not to be separated from the activity of the poet; against Gunkel, Genesis (see n. 5), xxx. 28  For a description of this method, see F. Polak, “Sociolinguistics, a Key to the Typology and the Social Background of Biblical Hebrew,” HS 47 (2006), 115–162, here 128–136, 141–151. 29  For statistical details, see F. P olak , “The Book of Samuel and the Deuteronomist: A Syntactic-Stylistic Analysis,” in Die Samuelbücher und die Deuteronomisten (ed. C. Schäfer-Lichtenberger; BWANT 188; Stuttgart: Kohlhammer, 2010), 34–73, here, 54–60; idem , “Sociolinguistics as Key” (see n. 28), 136–148.

Oral Platform and Language Usage in the Abraham Narrative

413

and governmental correspondence and to devise and to understand complex sentences. Significantly, this style appears in the epigraphic remains of ancient Hebrew, such as the short narrative of the Siloam Inscription and the ostraca of Kuntillet ʿAjrud, Lachish, and Arad.30 It dominates the Deuteronomic and Priestly corpora (the Judean corpus), as well as the texts that place themselves in the Persian era (the Achaemenid corpus). In the Abraham narrative, the following tales reveal an intricate style:31 the tales of Abrahamʼs separation from Lot (13:1–17), the war against the four kings (14:1–24), Isaacʼs birth (21:3–5) and the agreement with Abimelech (21:22–32); the opening of the tale of Abrahamʼs servant (24:1–21; vv. 22–67 represent the VoLB); and some Priestly passages (17:7–14, 23–27; 21:3–5; 23; 25:1–12). A second group of narratives is characterized by the low frequency of subordinate clauses (around a tenth of the text) and noun groups (mostly half of all clauses or less) and the high frequency of extremely short clauses in parataxis (predicate with implicit subject or with one complement), often amounting to a half of all clauses. In the following examples, the slash indicates the clause boundaries: Genesis 19:3 ‫אכלוּ‬ ֵ ֹ ‫ וַ יּ‬/ ‫אָפה‬ ָ ‫וּמצֹּות‬ ַ / ‫ וַ יַּ ַעשׂ ָל ֶהם ִמ ְשׁ ֶתּה‬/ ‫ל־בּיתֹו‬ ֵ ‫ וַ יָּ בֹאוּ ֶא‬/ ‫ וַ יָּ ֻסרוּ ֵא ָליו‬/ ‫ר־בּם ְמאֹד‬ ָ ‫וַ יִּ ְפ ַצ‬ But he urged them strongly, so they turned in to him, and came into his house; he made them a supper, and baked flat-cakes, and they ate.

This sequence contains five short clauses, consisting of predicate (with implied subject), ‫אכלּו‬ ֵ ֹ ‫וַ ּי‬, or of predicate with one complement, ‫וַ יָּ ֻסרוּ ֵא ָליו‬, ‫ל־בּיתֹו‬ ֵ ‫וַ יָּ בֹאוּ ֶא‬, ‫וּמצֹּות ָא ָפה‬. ַ One clause includes a pronominal and an adverbial phrase: ‫וַ יִּ ְפ ַצר־‬ ‫בּם ְמאֹד‬. ָ Only one clause is slightly longer: ‫וַ ּיַ ַעׂש ָל ֶהם ִמ ְׁש ֶּתה‬. An example of a sequence of short clauses in a divine promise is as follows: Genesis 12:2–3

‫ וֶ ְהיֵ ה ְבּ ָר ָכה‬/ ‫ וַ ֲאגַ ְדּ ָלה ְשׁ ֶמָך‬/ ‫ וַ ֲא ָב ֶר ְכָך‬/ ‫וְ ֶא ֶע ְשָׂך ְלגֹוי גָּ דֹול‬ ‫ וְ נִ ְב ְרכוּ ְבָך כֹּל ִמ ְשׁ ְפּחֹת ָה ֲא ָד ָמה‬/ ‫וּמ ַק ֶלּ ְלָך אָאֹר‬ ְ / ‫וַ ֲא ָב ֲר ָכה ְמ ָב ְר ֶכיָך‬ I will make of you a great nation, and I will bless you, and make your name great, and be a blessing. I will bless those who bless you, and the one who curses you I will curse; and in you all the families of the earth shall bless themselves.

This blessing contains six short clauses in parataxis, in which the subject is mostly implied in the verbal predicate (‫)וַ ֲאגַ ְדּ ָלה‬. The object is often expressed by 30  See F. P olak , “The Oral and the Written: Syntax, Stylistics and the Development of Biblical Prose Narrative,” JANES 26 (1998), 59–105, here 103–105. For the complex style of the petition from Yabneh Yam/Metzad Hashavyahu (KAI 200), see F. P olak , “Style is More than the Person: Sociolinguistics, Literary Culture and the Distinction between Written and Oral Narrative,” in Biblical Hebrew: Studies in Chronology and Typology (ed. I. Young; JSOTSup 369; London: T&T Clark, 2003), 38–103, here 47–48. 31  For more details, see Polak, “Abraham-Jacob Narrative” (see n. 1), 229–230.

414

Frank H. Polak

the object suffix (‫וְ ֶא ֶע ְשָׂך‬, ‫)וַ ֲא ָב ֶר ְכָך‬. The sequence is closed by a longer clause, with subject, ‫כֹּל ִמ ְשׁ ְפּחֹת ָה ֲא ָד ָמה‬, and indirect object, ‫ּבָך‬. ְ We note two noun groups: ‫ ְלגֹוי גָּ דֹול‬and ‫כֹּל ִמ ְשׁ ְפּחֹת ָה ֲא ָד ָמה‬, and two short subordinate clauses, ‫ ְמ ָב ְר ֶכיָך‬and ‫וּמ ַק ֶלּ ְלָך‬. ְ These passages exemplify the voiced, lean, brisk style (VoLB), which presents the information in short chunks and thus is close to the syntax of spontaneous spoken discourse. In its close resemblance to spoken language the VoLB shows its affinity to the oral arena. It is the style of the performance of the singer of tales. If this style is used in written narratives, the implication is that the writing authors recognize oral narrative as worthy literature and adhere to its stylistic habits and preferences rather than to the norms of the scribal chancery. Their narrative is “oral-derived.”32 The VoLB dominates the narratives of the patriarchs and the exodus, the Samuel-Saul-David narratives, and the prophetic cycle in 1 Kgs 17–2 Kgs 10 (13).33 The VoLB is found in the following episodes in the Abraham narrative: the tale of Abrahamʼs migration (12:1–9); the two tales of the endangerment of Sarah (Gen 12:10–20; 20:1–18); the story of the covenant between the pieces (15:1–12, 17–18); the tales of Hagarʼs flight and expulsion (ch. 16; 21:6–21); the stories of the terebinths at Mamre (18:1–15) and Abrahamʼs expostulation (vv. 16–33); the Sodom tale (19:1–22); large sections of the tale of Abrahamʼs servant (24:22–67; unlike vv. 1–21, IES); and the story of the divine promise to Abraham and Sarah (17:1–6,34 15–22; the sections on the circumcision represent IES). In the following sections of this study I will discuss further aspects of the oral-epic background of the Abraham narrative.

2 Epic Diction in the Abraham Tales 2.1 Residues of Epic Wording Biblical narrative contains many residues of poetic prosody.35 A case in point is the exposition of the Abraham tales:

 See J. M. Foley, Traditional-Oral Epic: The Odyssey, Beowulf and the Serbo-Croatian Return Song (Berkeley: University of California Press, 1990), 3–19. See also n. 77, below. 33  Further details are presented in Polak, “Abraham-Jacob Narrative” (see n. 1), 231–233. 34  On a pre-Priestly stratum in Gen 17:1–4a, 6, see, P. Weimar, “Gen 17 und die priesterschriftliche Abrahamgeschichte,” ZAW 100 (1988), 22–60, here 36–37. 35  See F. Polak, “Storytelling and Redaction: Varieties of Language Usage in the Exodus Narrative,” in the present volume, pp. 443–475, here 465–467, and C. A. Briggs, “The Hebrew Poem of the Creation,” Old Testament Student 3 (1883–1884), 273–288. 32

Oral Platform and Language Usage in the Abraham Narrative

415

Genesis 11:29–30 ‫ ֵׁשם ֵא ֶׁשת־‬/ ‫י־מ ְל ָכּה וַ ֲא ִבי יִ ְס ָכּה‬ ִ ‫ת־ה ָרן ֲא ִב‬ ָ ‫אַב ָרם וְ נָ חֹור ָל ֶהם נָ ִשׁים וְ ֵשׁם ֵא ֶשׁת־נָ חֹור ִמ ְל ָכּה ַבּ‬ ְ ‫וַ יִּ ַקּח‬ ‫ַא ְב ָרם ָׂש ָרי‬ ‫ ֵאין ָלּה וָ ָלד‬// ‫וַ ְּת ִהי ָׂש ַרי ֲע ָק ָרה‬ Abram and Nahor, now, took to themselves wives; the name of Abramʼs wife was Sarai, and the name of Nahorʼs wife was Milcah, the daughter of Haran the father of Milcah and Iscah. And Sarai remained barren, she had no child.

The introduction of the two women follows a well-known pattern, as evidenced by the opening of Samuelʼs birth tale: 1 Samuel 1:2

‫וְ לֹו ְשׁ ֵתּי נָ ִשׁים‬ ‫ וְ ֵשׁם ַה ֵשּׁנִ ית ְפּנִ נָּ ה‬// ‫אַחת ַחנָּ ה‬ ַ ‫ֵשׁם‬ ‫וּל ַחנָּ ה ֵאין יְ ָל ִדים‬ ְ // ‫וַ יְ ִהי ִל ְפנִ נָּ ה יְ ָל ִדים‬

He had two wives; the name of the one was Hannah, and the name of the other Peninnah. Peninnah, now, had children, but Hannah had no children.

In both cases the first clause mentions the husband(s) as antecedent for the introduction of two wives in the first bicolon,36 whereas the second bicolon specifies the status of the women in terms of offspring. The introduction of Sarai contains a number of noteworthy elements. The coupling of ‫ עקרה‬and ‫אין לה ולד‬ is based on an associative pair, ‫ עקרה‬// ‫ילד‬, such as ‫וְ ִא ְשׁתֹּו ֲע ָק ָרה וְ לֹא יָ ָל ָדה‬, “and his wife was childless, not having given birth” (Judg 13:2; and without negation, 1 Sam 2:5).37 An additional issue is the form ‫וָ ָלד‬, which preserves the initial w-, although in Northwest Semitic languages word initial w- generally shifts to y-.38 One also notes cases of repetitive parallelism:39 Genesis 12:10

‫אָרץ‬ ֶ ‫י־כ ֵבד ָה ָר ָעב ָבּ‬ ָ ‫ ִכּ‬// ‫אַב ָרם ִמ ְצ ַריְ ָמה ָלגוּר ָשׁם‬ ְ ‫ וַ יֵּ ֶרד‬// ‫אָרץ‬ ֶ ‫וַ יְ ִהי ָר ָעב ָבּ‬ Now a famine arose in the land, and Abram went down to Egypt to find shelter there, for the famine was severe in the land.

36  A similar pattern is used to introduce the fatherʼs daughters: Gen 29:16–17; 1 Sam 14:49, with a later variant in Job 42:13–14. 37  So also Judg 13:3; Isa 54:1; Job 24:21. 38  So also 2 Sam 6:23 (ketiv according to Madinḥa‍ʾe). See GVG 1.139 (§ 49.f); T. Nöldeke, Compendious Syriac Grammar (trans. J. A. Crighton; Eugene, OR: Wipf & Stock, 2003), 26 (§ 40A). In the Ugaritic grapheme wld (CTU 1.14 III:48), the survival of this waw may be explained by assimilation to following /u/: *wullad; see J. Tropper, Ugaritische Grammatik: Zweite, stark überarbeitete und erweiterte Auflage (AOAT 273; Münster: Ugarit-Verlag, 2012), 153, 564. 39  So also Gen 31:25, and similarly 37:34.

416

Frank H. Polak

Genesis 13:6

‫ וְ לֹא יָ ְכלוּ ָל ֶשׁ ֶבת יַ ְח ָדּו‬// ‫כוּשׁם ָרב‬ ָ ‫י־היָ ה ְר‬ ָ ‫ ִכּ‬// ‫אָרץ ָל ֶשׁ ֶבת יַ ְח ָדּו‬ ֶ ‫וְ לֹא־נָ ָשׂא א ָֹתם ָה‬ Thus the land could not support them to dwell together; for their possessions were great, and they were unable to dwell together. Genesis 13:12

‫ד־סד ֹם‬ ְ ‫ וַ יֶּ ֱא ַהל ַע‬/ ‫ וְ לֹוט יָ ַשׁב ְבּ ָע ֵרי ַה ִכּ ָכּר‬// ‫ץ־ּכנָ ַען‬ ְ ‫ַא ְב ָרם יָ ַׁשב ְּב ֶא ֶר‬ Abram dwelt in the land of Canaan, and Lot dwelt among the cities of the plain, and moved his tents as far as Sodom.

A case of repetitive parallelism between two lines is preserved in the tale of Abraham at the terebinths of Mamre: Genesis 18:7–8

‫ וַ יְ ַמ ֵהר ַל ֲעשֹׂות אֹתֹו‬/ ‫ל־הנַּ ַער‬ ַ ‫ וַ יִּ ֵתּן ֶא‬/ ‫ן־בּ ָקר ַרְך וָ טֹוב‬ ָ ‫וַ יִּ ַקּח ֶבּ‬ ‫אכלוּ‬ ֵ ֹ ‫ וַ יּ‬/ ‫יהם ַתּ ַחת ָה ֵעץ‬ ֶ ‫ וְ הוּא־ ע ֵֹמד ֲע ֵל‬/ ‫יהם‬ ֶ ֵ‫ וַ יִּ ֵתּן ִל ְפנ‬/ ‫ן־ה ָבּ ָקר ֲא ֶשׁר ָע ָשׂה‬ ַ ‫וּב‬ ֶ ‫וַ יִּ ַקּח ֶח ְמאָה וְ ָח ָלב‬ [Thus Abraham ran to the herd,] fetched a calf tender and choice, and gave it to the servant; and he hastened to prepare it. And he took curds and milk, and the calf which he had prepared, and set it before them; and he stood by them under the tree, and they did eat.

Since these cases, though not numerous, keep occurring in different narratives, they probably represent a traditional style. 2.2 Epic Formulae and Patterns An important feature that permeates the tales of the patriarchs is the use of frequently repeated epic formulae,40 traditional phrases that are opposed to simple expressions with similar meaning, are matched by or cognate with fixed phrases in Ugaritic/Akkadian epic poetry, and are related to the poetic structure of parallelism.41 A case in point is the phrase “raising the eyes and see,” ‫וישא עיניו‬ ‫וירא‬, in its various grammatical forms. This phrase is matched by the Ugaritic phrases wyšu ʿnh wyʿn, “he raised his eyes and saw,”or bnši ʿnh wyphn, “raising his eyes he saw.” In the Old Babylonian version of the Gilgamesh Epic one encounters the bicolon iššīma īnīšu, ītamar awīlam, “he – i.e. Enkidu – lifted 40  I define frequency as the number of occurrences of the phrase at hand, relative to the number of occurrences of the lexemes of which the formula consists. 41  See U. Cassuto, “Biblical and Canaanite Literatures,” in Biblical and Oriental Studies (trans. I. Abrahams; 2 vols.; Jerusalem: Magnes Press, 1973–1975), 2:16–59, here 16–26; Avishur, Studies in Biblical Narrative (see n. 14), 225–238; R. C. Culley, Oral Formulaic Language in the Biblical Psalms (Toronto: University of Toronto Press, 1967), 10; F. Polak, “Linguistic and Stylistic Aspects of Epic Formulae in Ancient Semitic Poetry and Biblical Narrative,” in Biblical Hebrew in Its Northwest Semitic Setting (ed. S. Fassberg and A. Hurvitz; Winona Lake, IN: Eisenbrauns, 2006), 285–304, with further references; and idem, “Formulaic Style,” in Encyclopedia of Hebrew Language and Linguistics (ed. G. Khan; 3 vols.; Leiden: Brill, 2013), 1:906–11. The relationship to parallelism means that the clause embodying the formula fills up a complete colon or that the formula is spread over two consecutive cola. On Van Setersʼs neglect of the data assembled by Cassuto, see below, n. 73.

Oral Platform and Language Usage in the Abraham Narrative

417

his eyes and saw the man” (Pennsylvania Tablet IV 137).42 A phrase of this kind is divided over two cola and thus forms a parallelistic structure, underlined by the semantic connection between “eye” and “see.” In the Abraham narrative we encounter this phrase with direct object: Genesis 22:4

‫ת־ה ָמּקֹום ֵמ ָרחֹק‬ ַ ‫ת־עינָ יו וַ יַּ ְרא ֶא‬ ֵ ‫אַב ָר ָהם ֶא‬ ְ ‫ישׁי וַ יִּ ָשּׂא‬ ִ ‫ַבּיֹּום ַה ְשּׁ ִל‬ On the third day Abraham lifted up his eyes, and saw the place from afar.

So also in 13:10, 14; 24:64. In other passages this formula is used with ‫הנה‬: Genesis 24:63

‫שׂוּח ַבּ ָשּׂ ֶדה ִל ְפנֹות ָע ֶרב וַ יִּ ָשּׂא ֵעינָ יו וַ יַּ ְרא וְ ִהנֵּ ה גְ ַמ ִלּים ָבּ ִאים‬ ַ ‫וַ יֵּ ֵצא יִ ְצ ָחק ָל‬ And Isaac went out to walk in the field toward evening and, he lifted up his eyes and saw, here, camels approaching.

So also in 18:2; 22:13. The frequency of this phrase is shown by comparison with the simple phrase and variants. The entire narrative includes twenty-three instances of rāʾâ (qal),43 of which the formulaic phrase forms almost a third (30.43 percent). When we count the cases in the chapters that contain instances of the formulaic phrases, the picture is even sharper. In chapter 13, we note two instances of the formula (13:10, 14) and one simple verb (13:15). The Mamre tale and the tale of Isaacʼs offering include the formulaic phrase only (18:2; 22:4, 13).44 The tale of Abrahamʼs servant contains two formulaic phrases (24:63, 64) and one example of the simple verb (22:30). In these chapters, then, the frequency of the formulaic phrase relative to rāʾâ is extremely significant. The verb nāśāʾ has thirteen instances, seven of them in the formulaic phrase at hand. This verb appears once in the meaning “to carry” (21:8) and a few times in figurative and thus irrelevant usage (13:6; 18:24, 26; 19:21). The only other instance is a second formulaic phrase, ְ‫“( וַ ִתּ ָשּׂא ֶאת־ק ָֹלהּ וַ ֵתּ ְבךּ‬she lifted her voice and wept,” 21:16). Disregarding the cases of figurative usage, one arrives at seven cases out of nine instances or more than three-quarters of the relevant cases (77.78 percent). Thus the phrase ‫ וישא עיניו וירא‬must indeed count as highly frequent.45 For the second formulaic phrase, ְ‫( וַ ִתּ ָשּׂא ֶאת־ק ָֹלהּ וַ ֵתּ ְבךּ‬21:16), frequency seems less well established. However, this is the only instance of bākâ, with the mean42  A. R. G eorge , The Babylonian Gilgamesh Epic: Introduction, Critical Edition and Cuneiform Texts (2 vols.; Oxford: Oxford University Press, 2003), 1:176. For details, see Polak, “Linguistic and Stylistic Aspects” (see n. 41), 286–288. 43  I have not counted the forms ‫( ר ִֹאי‬16:13), ‫“( יִ ְר ֶאה‬he will provide,” 22:8, 14). 44  ‫( ֵא ֲר ָדה־נָּ א וְ ֶא ְר ֶאה‬18:21) does not belong to the tale of Mamre. In 22:8, 13 the verb rāʾâ is used with the meaning “to provide,” “to see to.” 45  The noun ‫עין‬, in the meaning “eye” (21 instances) and not in figurative usage (16:4–6; 18:3; 19:8, 14, 19; 20:15, 16; 21:11–12), appears only in 23:11 and 18 and in the formulaic phrases.

418

Frank H. Polak

ing “to weep,” in the Abraham narrative.46 This passage is of particular interest since this tale includes the phrase ‫יה וַ ֵתּ ֶרא ְבּ ֵאר ָמיִם‬ ָ ֶ‫ת־עינ‬ ֵ ‫ֹלהים ֶא‬ ִ ‫וַ יִּ ְפ ַקח ֱא‬, “Then God opened her eyes and she saw a well of water” (21:19). This phrase represents a causative transformation of the formula, matched by the Old Aramaic phrase in the Sefire Treaty: ]. . .[ ‫פקחו עיניכם לחזיה עדי בר גאיה‬, “open your eyes to see the treaty of Bar Ga‍ʾyah [. . .]” (KAI 222 I A:13).47 In addition we note such compound phrases as ‫ ויאמר‬. . . ‫ויקרא‬, “call and say,”48 ‫ ויאמר‬. . . ‫ויען‬, “speak up and say,”49 ‫ ותלד‬. . . ‫ותהר‬, “conceive and give birth,”50 ‫ וילך‬. . . ‫ויקם‬, “rise up and go,”51 ‫ וילך‬. . . ‫ויקח‬, “take and go,”52 . . . ‫ויקח‬ ‫וישם‬, “take and put,”53 ‫ ויתן‬. . . ‫ויקח‬, “take and give,”54 ‫ וישתחו‬. . . ‫ויקד‬, “to kneel 46  In Gen 23:12, the verb appears with the meaning “to mourn.” In the Jacob tale we count three cases of bākâ, two of them in the formula ‫( וישא קלו ויבך‬27:38; 29:11), and one free (33:4). 47  See J. A. F itzmyer , The Aramaic Inscriptions of Sefire (rev. ed.; Biblica et Orientalia 19A; Roma: Pontifico Instituto Biblico, 1995), 43–44, 77–78. Note also the variant ‫וַ ִתּ ָפּ ַק ְחנָ ה‬ ‫יהם וַ יֵּ ְדעוּ‬ ֶ ֵ‫( ֵעינֵ י ְשׁנ‬Gen 3:7). 48  ‫ ויאמר‬. . . ‫ויקרא‬: Gen 12:18; 16:13; 19:5; 20:9; 21:17; 22:11, 15–16; 24:58 (eight instances). In Akkadian, the analogical phrase is šasû-zakāru. In Ugaritic the verb ṣḥ frequently serves to frame direct discourse. 49  ‫ ויאמר‬. . . ‫ויען‬: Gen 18:27 and 24:50. Vis-à-vis ʾāmar, this phrase is not significant at all, but the verb ʿānâ has only three additional instances: Gen 23:5, 10, 14 (always Ephron responding to Abraham), with ‫ ֵלאמֹר‬rather than a finite form of ʾāmar. In Ugaritic ʿn is the regular framing of spoken discourse (matched in Hebrew by Job 3:2 and passim). In Aramaic this phrase is found in two fables included in the Ahiqar proverbs (CAP 216, 100, 118–119). 50  ‫ ותלד‬. . . ‫ותהר‬: Gen 16:11; 19:36–37, 38; 21:2. In Akkadian, one notes the use of erû-alādu, and in Ugaritic the stretch bm.nšq.whr.bḥbq.ḥmḥmt.tqt[nṣn.w]tldn (“in kissing, conception; in embracing, mating/passion [?]; they crouch and give birth”; CTU 1.23:51, 56–58; 1.17 I:39–41); see T. J. Lewis, “The Birth of the Gracious Gods,” in Ugaritic Narrative Poetry (ed. S. B. Parker; SBLWAW 9; Atlanta: Scholars, 1997), 212; M. S. Smith, The Rituals and Myths of the Feast of the Goodly Gods of KTU/CAT 1.23 (Atlanta: Society of Biblical Literature, 2006), 96, 103–104. 51  ‫ וילך‬. . . ‫ויקם‬: Gen 13:17; 22:3, 19; 24:10, 61; five cases versus twenty-two instances of qûm in the qal (22.73 percent), and forty-five cases of hālak (11.11 percent). In Akkadian, one notes the collocation tebû-alāku, corresponding to Ugaritic tbʿ. If one adds the ten instances of ‫ וילך‬. . . ‫( ויקח‬see n. 52), one arrives at fifteen formulaic phrases, exactly a third of the number of occurrences of hālak. 52  ‫ וילך‬. . . ‫ויקח‬: Gen 12:19; 14:11, 12; 22:2, 6; 24:10, 51, 61; ‫ ויצא ללכת‬. . . ‫ויקח‬: Gen 11:31; 12:5 (ten cases). In Ugaritic one notes qḥ [. . .] lk (CTU 1.16 II:41–43), qḥ [. . .] ẓi (CTU 1.12 I:17–19), qḥ [. . .] idk pnk al ttn (CTU 1.5:6–13). The verb lāqaḥ occurs forty-eight times in the Abraham narrative, with ten cases in the present phrase (20.83 percent). In total fifteen instances of hālak are formulaic (33.33 percent; see n. 51). 53  ‫ וישם‬. . . ‫ויקח‬: Gen 21:14; 22:6; two cases versus seven nonfigurative occurrences of śym (25.57 percent; disregarding 13:16; 21:13, 18). In Ugaritic, we note the cognates tšu [. . .] tštnn and tqḥ [. . .] tštn (CTU 1.18 IV:27–28; 1.19 II:10; similarly Judg 9:48). 54  ‫ ויתן‬. . . ‫ויקח‬: Gen 15:10; 16:3; 18:7–8; 20:14; 21:14, 27; seven cases as against forty-one instances of nātan (17.07 percent), and forty-eight occurrences of lāqaḥ (14.58 percent). In

Oral Platform and Language Usage in the Abraham Narrative

419

and bow down,”55 ‫ וישתחו‬. . . ‫ויקם‬, “to rise up and bow down,”56 ‫וישתחו ארצה‬, “to bow down to the ground,”57 and ‫וישתחו אפים ארצה‬, “to bow down with the face to the ground.”58 Notably, hištaḥăwâ does not have a clear derivation in Hebrew but is matched by the Ugaritic verb hštḥwy, representing the reflective of the shaphel, which is a productive form in Ugaritic but not in Biblical Hebrew.59 In total the Abraham narrative includes thirteen different types of formulaic phrases with fifty-two instances.60 An analysis of the class of these phrases indicates the frequency of formulaic language relative to the size of a number total we count nineteen cases of bound phrases opening with the latter verb, or 39.58 percent of its forty-eight occurrences. This phrase is matched by the legal formula leqû-nadānu, used in the Akkadian contracts from Ugarit to indicate the transfer of property; see J. C. Greenfield, “Našû-Nadānu and Its Congeners,” in ʿAl Kanfei Yonah: Collected Studies of Jonas C. Greenfield on Semitic Philology (ed. S. M. Paul et al.; 2 vols.; Jerusalem: Magnes Press, 2001), 2:720–724; idem, “Aramaic HNṢL and Some Biblical Passages,” in ʿAl Kanfei Yonah: Collected Studies of Jonas C. Greenfield on Semitic Philology (ed. S. M. Paul et al.; 2 vols.; Jerusalem: Magnes Press, 2001), 1:214–216; and F. P olak , “The Lāqaḥ-Nātan Formula: Some Additional Comments,” Shnaton 7–8 (1984), xii–xiii, 179–186 (Hebrew with English summary). 55  ‫ וישתחו‬. . . ‫ויקד‬: Gen 24:26, 48 (so also 43:28; Exod 4:31; 12:27; 34:8; Num 22:31; 1 Sam 24:8; 28:14; 1 Kgs 1:16, 31; Neh 8:6; 1 Chr 29:20; 2 Chr 29:30; and see n. 58, below). This phrase is cognate with the Ugaritic formula lpʿn il thbr wtql / tštḥwy wtkbdh (CTU 1.4 IV:25–26; similarly 1.3 III:9–10; 1.4 VIII:26–29; 1.6 I:36–38; 1.17 IV:50–51; 1.1 II:15–16; III:2–3, 24–25; 1.2 III:5–6). In Akkadian, qadādu indicates bending down or prostration: appašu liqdud, ina erṣete lippalsiḫ, “let him bow his head/nose, squat on the ground,” in E. Ebeling, Leben und Tod nach den Vorstellungen der Babylonier, I: Texte (Berlin: de Gruyter, 1931), 71–72 (text 18 ii:13). 56  ‫ וישתחו‬. . . ‫ויקם‬: Gen 19:1; 23:7 (so also Exod 33:10; 1 Sam 25:41; 1 Kgs 2:19). The corresponding phrase in Ugaritic is lpnnh.ydd.wyqm, lpʿnh.ykrʿ.wyql (CTU 1.10 II:18–19), and in Akkadian, izzizū iknušū (Enuma Elish V 88). 57  ‫וישתחו ארצה‬: Gen 18:2; 24:52 (so also 33:3; 37:10; 43:26; 2 Kgs 2:15). 58  ‫וישתחו אפים ארצה‬: Gen 19:1 (so also 42:6; 25:41; Neh 8:6; 1 Chr 21:21); and with ‫כרע‬ or ‫ קדד‬only: 2 Chr 7:3; 20:18. 59  D. Pardee , review of Ugaritische Grammatik, by J. Tropper, AfO 50 (2003–2004), 294; C. Cohen , “The Saga of a Unique Verb in Biblical Hebrew and Ugaritic,” in Textures and Meanings: Thirty Years of Judaic Studies at University of Massachusetts, Amherst (ed. L. H. Ehrich et al.; Amherst: University of Massachusetts, 2004), 323–342; M. S. S mith , The Ugaritic Baʿal Cycle, Vol. I: Introduction with Text and Commentary of KTU 1.1–1.2 (VTSup 55; Leiden: Brill, 1994), 168, n. 95; contra J. A. E merton , “The Etymology of hištaḥawāh,” OTS 20 (Leiden: Brill, 1977), 41–55; S. Kreuzer, “Zur Bedeutung und Etymo­ logie von hištaḥawāh/yštḥwy,” VT 35 (1985), 39–70. 60  A slightly different result is presented in F. P olak , “Epic Formulas in Biblical Narrative: Frequency and Distribution,” in Les actes du second colloque internationale Bible et Informatique: Méthodes, outils, résultats (Jerusalem, 9–13 Juin 1988) (Geneva: Champion-Slatkine), 435–488, here 453–459. In this paper I included the phrases ‫ וישת‬. . . ‫ויאכל‬ (in the Abraham narrative only Gen 24:54), and with ‫ שמע‬vocative as opening formula (four instances: 23:6, 11, 13, 15).

420

Frank H. Polak

of pericopes.61 In chapters 12, 13, 16, 21, and 24, the frequency hovers around 3–4 percent, whereas the incidence of such phrases is lower in chapters 19–20 (1.5–2 percent) and far lower in chapters 14–15, 17, and 23. In contrast, in 18:1–15 and 22:1–19 the frequency of formulaic phrases is far higher (around 7–8 percent).62 For the sake of comparison, in the Jacob tale (Gen 25:29–34; 27:1–33:17) we find fifteen types of formulaic phrases with forty instances.63 In Judges we encounter eleven phrase types with thirty-three instances and in the Samuel-Saul-David narratives seventeen different phrases with 130 examples (1 Sam 1–1 Kgs 2), but in view of the size of those texts these figures are far less impressive than they may seem.64 In addition, oral poetry uses thematic patterns, as “compositional themes,”65 such as the call pattern. The initiator utters the name (or title) of the addressee, who responds with “here I am.”66 In the narrative of Isaacʼs offering, which as discussed above is rich in formulaic language, this pattern appears twice: Genesis 22:1

‫ ִהנֵּ נִ י‬/ ‫אמר‬ ֶ ֹ ‫ וַ יּ‬/ ‫אַב ָר ָהם‬ ְ / ‫אמר ֵא ָליו‬ ֶ ֹ ‫וַ יּ‬ He said to him, “Abraham!” And he said, “Here I am.” Genesis 22:7

‫ ִהנֶּ נִּ י ְבנִ י‬/ ‫אמר‬ ֶ ֹ ‫ וַ יּ‬/ ‫אָבי‬ ִ / ‫אמר‬ ֶ ֹ ‫ וַ יּ‬/ ‫אָביו‬ ִ ‫ל־אַב ָר ָהם‬ ְ ‫אמר יִ ְצ ָחק ֶא‬ ֶ ֹ ‫וַ יּ‬ Isaac said to his father Abraham, “Father!” And he said, “Here I am, my son.”

61  The size of a pericope is established by the number of nouns, verbs, adjectives (excluding numerals), and adverbs (content words, excluding particles, prepositions and pronouns). 62  If one includes recurring phrases that do not have parallels in Ugaritic or Akkadian texts, the figures would be even higher: ‫ ויאכל‬. . . ‫( ויתן‬18:8), ‫ בבקר‬. . . ‫( וישכם‬19:27; 20:8; 21:14; 22:3); ‫( וישלח ידו ויקח‬22:10). In view of the limited size of the biblical narrative corpus and our scant knowledge of Ugaritic or Akkadian narrative poetry, it is possible that the actual incidence of formulaic language in patriarchal narrative is far more significant, but only new discoveries in ancient Near Eastern literature can enlighten us. 63  See n. 60, above. For the Jacob tale, I have added the phrase ‫] וישק‬. . .[ ‫ויחבק‬, “embrace and kiss” (29:13; 33:4; 48:10), cognate with the Ugaritic formula bm.nšq.whr. / bḥbq.ḥmḥmt. / tqt[nṣn.w]tldn (see n. 50). With 2,596 content words, the frequency of formulaic phraseology stands at 1.5 percent. In the first section of Exodus (Exod 1:1–6:1; 7:8–11:10; 12:29–39) we encounter seventeen instances (six types) for 2,852 content words, with frequency 0.60 percent, and thirteen (eight types) in the second section (14:1–31; 15:22–20:21; 24:1–14; 32–34; eight types) for 2,950 content words, and a frequency of 0.44 percent. 64  The book of Judges contains 4,750 content words; thus, the overall frequency of formulaic phrases is 0.69 percent. For the book of Samuel we find 19,884 content words, with a formulaic frequency of 0.65 percent. 65  See A. B. Lord, The Singer of Tales (Cambridge, MA: Harvard University Press, 1960), 68–92; M. P. Coote, “The Singerʼs Themes in Serbocroatian Heroic Song,” California Slavic Studies 11 (1980), 201–236, here 201–202. 66  So also Gen 46:2; Exod 3:4–5; and with variants: 1 Sam 3:4, 8, 10 (all with doubling of the name).

Oral Platform and Language Usage in the Abraham Narrative

421

These data provide ample justification for the conclusion that the Abraham narrative contains residues of a stratum that is close to the formulaic tradition of the Northwest Semitic epic diction, as represented by the Ugaritic poetic narratives. Actually, the Ugaritic texts, in particular the Baʿal Cycle and the tale of Danʾilu and ʾAqhat, are far richer in formulaic language.67 This issue is of great weight in view of the research of Milman Parry and Albert Lord, who found that Serbo-Croatian oral poetry is almost entirely built of traditional formulae and thus argued that high incidence of formulaic language, such as in the Homeric Iliad and Odyssey, or early medieval Old English poetry, indicates oral composition by the singer of tales.68 In accordance with this approach, Frank Cross and Kenneth Aitken conclude that Ugaritic narrative poetry originates in the oral performance.69 Admittedly, the issue of formulaic language is problematic. First, Ruth Finnegan has shown that not all oral literature is based on traditional formulae.70 Second, formulaic language is also found in written discourse, such as Old English translations of Latin texts and even in Vergilʼs Aeneid.71 In view

 K. T. A itken , “Oral Formulaic Composition and Themes in the Aqhat Narrative,” UF 21 (1989), 1–16. 68  See M. Parry , The Making of Homeric Verse: The Collected Papers of Milman Parry (ed. A. Parry; Oxford: Clarendon Press, 1971); A. B. L ord , The Singer of Tales (see n. 65); G. S. K irk , “The Making of the Iliad: Preliminary Considerations,” in The Iliad: A Commentary (ed. G. S. Kirk; 6 vols.; Cambridge: Cambridge University Press, 1985–1993), 1:1–37; idem , “Typical Motifs and Themes,” in The Iliad: A Commentary (ed. G. S. Kirk; 6 vols.; Cambridge: Cambridge University Press, 1985–1993), 2:15–27; B. H ainsworth , “Introduction,” in The Iliad: A Commentary (ed. G. S. Kirk; 6 vols.; Cambridge: Cambridge University Press, 1985–1993), 3:1–53; R. J anko , “The Origins and Evolution of the Epic Diction,” in The Iliad: A Commentary (ed. G. S. Kirk; 6 vols.; Cambridge: Cambridge University Press, 1985–1993), 4:8–19; M. W. Edwards, “Composition by Theme,” in The Iliad: A Commentary (ed. G. S. Kirk; 6 vols.; Cambridge: Cambridge University Press, 1985–1993), 5:11–23; J. Russo, “The Formula,” in A New Companion to Homer (ed. I. Morris and B. Powell; Leiden: Brill, 1997), 238–260. For the view the Homeric text came into being in writing within a formulaic, oral framework, see M. L. West, The Making of the Iliad: Disquisition and Analytical Commentary (Oxford: Oxford University Press, 2011), 3–7, 10–14, 28–37, 48–51, 67–70; J. Latacz, “Formelhaftigkeit und Mündlichkeit,” in Homers Ilias: Gesamtkommentar; Prolegomena (ed. J. Latacz; München: Saur, 2000), 39–60. 69  See Cross, “Epic Traditions of Ancient Israel” (see n. 11), 14–16; Aitken, “Oral Formulaic Composition” (see n. 67). 70  R. F innegan , Oral Poetry: Its Nature, Significance and Social Context (Cambridge: Cambridge University Press, 1977), 73–87. 71  L. D. B enson , “The Literary Character of Anglo-Saxon Formulaic Poetry,” PMLA 81 (1966), 334–341; M. Sale, “Virgilʼs Formularity and Pius Aeneas,” in Signs of Orality: The Oral Tradition and Its Influence in the Greek and Roman World (ed. E. A. Mackay; Mnemosyne Supplementum 188; Leiden: Brill, 1999), 199–220. 67

422

Frank H. Polak

of this criticism, Patricia Kirkpatrick and John Van Seters reject Crossʼs theory regarding the oral-epic background of pentateuchal narrative.72 However, many scholars have recognized that the rejection of all connection between orality and formulaic language goes too far.73 When Rosalind Thomas argues that formulaic language is traditional in a more general sense,74 one may ask where to posit the origins of the formulaic tradition. The assumption that this phraseology originates in the written tradition crumbles in the face of a severe problem: no ancient Near Eastern text contains formulaic phraseology with the frequency encountered in the Ugaritic texts analyzed by Aitken. In this respect, then, the formulaic Ugaritic texts are highly exceptional. Thus, it seems best to accept the hypothesis that the formulaic character of the Ugaritic texts connects those texts to the oral tradition.75 The force behind the tradition is the oral background of the ancient Northwest Semitic epic tradition. Hence, the frequent use of formulaic language in patriarchal narrative indicates an underlying oral, poetic tradition. The text as transmitted in writing was composed by the sōpēr, but this text is heavily indebted to the artists of oral literature whose norms, habitus, and linguistic register served as examples. A text of this type is a tertium quid rather than a written artefact.76 In the words of John Foley and Albert Lord,  J. Van Seters, In Search of History: Historiography in the Ancient World and the Origins of Biblical History (New Haven, CT: Yale University Press, 1983), 19–22, 30, 220–226; P. G. K irkpatrick , The Old Testament and Folklore Study (JSOTSup 62; Sheffield: JSOT Press, 1988), 53–64; R. D. Miller, Oral Tradition in Ancient Israel (Eugene, OR: Cascade Books, 2010), 13–39. However, these discussions consider neither the formulaic phrases themselves nor their distribution in Ugaritic and biblical texts. For Cassutoʼs contribution, Van Seters (In Search of History, 226, n. 62) refers the reader to the discussion of C. Conroy, “Hebrew Epic: Historical Notes and Critical Reflections,” Bib 61 (1980), 1–30, here 9–11; the latter mentions the stylistic features that biblical narrative shares with the Ugaritic texts but does not discuss their import. Thus, Van Seters is able to refer to the epic formulae as “folkloristic” elements, although they are attested en masse in Ugaritic written texts (227). 73  M. S. Jensen, Writing Homer (Copenhagen: The Royal Danish Academy of Sciences and Letters, 2011), 194–256; M. Sale, “The Oral-Formulaic Theory Today,” in Speaking Volumes: Orality and Literacy in the Greek and Roman World (ed. J. Watson; Mnemosyne Supplementum 218; Leiden: Brill, 2001), 53–80; R. T homas , Literacy and Orality in Ancient Greece (Cambridge: Cambridge University Press, 1992), 35–36, 41; M. Finkelberg, “Formulaic and Nonformulaic Elements in Homer,” Classical Philology 84 (1989), 179–197. 74  Thomas, Literacy and Orality (see n. 73), 43–44. The argument that formulaic language is used by Virgil is countered by important deviations in the use of the Homeric register by the poets in the post-Homeric tradition, such as Virgil and Quintus Smyrnaeus; see S ale , “Oral-Formulaic Theory Today” (see n. 73), 71–73, 78–80; idem, “Virgilʼs Formularity” (see n. 71), 218–220. 75  The oral background of Ilimilkuʼs literary work is highlighted by Pardee, Ugaritic Texts (see n. 9), 47–49, in view of the (partially preserved) colophon of RS 92.2016. 76  P. Z umthor , “The Text and the Voice,” New Literary History 16 (1984), 67–92; F. Bäuml, “Medieval Texts and the Two Theories of Oral-Formulaic Composition: A Proposal for a Third Theory,” New Literary History 16 (1984), 31–49; idem, “Autorität und Performanz: 72

Oral Platform and Language Usage in the Abraham Narrative

423

it is an oral-derived, transitional text, a view that is in keeping with the analysis of Rosalind Thomas.77 In the final section of this study I will point to some of the literary effects that are particular to the oral background of patriarchal narrative and classical biblical narrative in general. But first I must discuss some problems pertaining to the historical background of the Abraham narrative. 2.3 The Formula in Exilic/Persian-Era Prophetic Literature Can we presume that the formulaic phraseology of the Abraham narrative, and of patriarchal narrative in general, derives from other branches of biblical literature? A theory along these lines is not supported by the state of biblical literature as it is known to us. First, one must note the variety in the use of these formulae in the Abraham tales. Formulaic phraseology appears in many different contexts, representing different themes and different expressivity. The phrase ‫וישא עיניו‬ ‫וירא‬, “to lift the eyes and see,” is used for such diverse events as Abrahamʼs perception of the travelers approaching (Gen 18:2), of Mount Moriah (22:4) and of the ram (22:13), Jacobʼs account of his dream vision (31:10), Jacob seeing Esau approaching, and Esau viewing Jacobʼs camp (33:1, 5).78 The occurrence of the very same idiom in so many different contexts must raise the question which conditions motivate different narrators to use this phrase. Of course, one could answer that this is just the way to formulate a visual-perception event. However, outside patriarchal narrative this idiom is only encountered a few times in the Joseph narrative (Gen 37:23; 43:29),79 the Balaam tale (Num 24:2), the Deuteronomic speech (Deut 3:27; 4:19), Joshuaʼs meeting with the angel (Josh 5:13), the Gibeah tale (Judg 19:17), and the Samuel-Saul-David tales (1 Sam 6:13; 2 Sam 13:34; 18:24).80 Gesehene Leser, gehörte Bilder, geschriebener Text,” in Verschriftung und Verschriftlichung: Aspekte des Medienwechsels in verschiedenen Kulturen und Epochen (ed. C. Ehler and U. Schaefer; Scriptoralia 94; Tübingen: Narr, 1998), 248–273, here 248–249; U. Schaefer, Vokalität: Altenglische Dichting zwischen Mündlichkeit und Schriftlichkeit (Scriptoralia 53; Tübingen: Narr, 1993). 77  A. B. Lord, The Singer Resumes the Tale (ed. M. L. Lord; Ithaca, NY: Cornell University Press, 1995), 222–237; F oley , Traditional-Oral Epic (see n. 32), 3–19; T homas , Literacy and Orality (see n. 73), 45–50. For biblical narrative, this framework has been adopted by S. Niditch, Oral World and Written Word: Ancient Israelite Literature (Louisville, KY: John Knox Press, 1996), 9–13, 113–120; R. C. C ulley , “Orality and Writtenness,” in Writings and Speech in Israelite and Ancient Near Eastern Prophecy (ed. E. Ben Zvi and M. H. Floyd; Atlanta: Society of Biblical Literature, 2000), 45–64, here 55–56. 78  For the statistics see pp. 416–421, above. 79  Two instances in the Joseph tale, among forty-four cases of rāʾâ (4.54 percent), in sharp contrast with the 30.43 percent of uses of this verb in the Abraham tales (see p. 417, above). 80  Three instances in the Samuel-Saul-David tales, among 118 cases of rāʾâ (2.54 percent). In addition, we have to mention the variant in Job 2:12, in which the phrase ‫ וְ לֹא ִה ִכּ ֻירהוּ‬replaces

424

Frank H. Polak

Should we posit that the source of this phraseology be found in prophetic texts such as the framing of Zechariahʼs visions (Zech 2:1, 5; 5:1, 5, 9; 6:1; so also Dan 8:3; 10:5), Jeremiah (3:2; 13:20), and Deutero- and Trito-Isaiah (Isa 40:26; 49:18; 52:8; 60:4)? These data may suggest that this formula originated in prophetic discourse rather than in epic narrative. But the way this formula is used indicates fundamental differences. First, in Zechariah and Daniel its use is limited to the framing of the vision. This limitation contrasts sharply with the variety of usage in biblical narrative, where the dream vision (Gen 31:10, 12) is only one of the contexts in which the formula is attested. Thus it seems far preferable to explain the use of the formula in this specific context as a special case, dependent on its use in biblical narrative. Second, in the prophetic texts the formula is phrased in a different way. In narrative, this formula is characterized by its two-clause structure. The first clause indicates the inception of the perception event, and mostly the viewing subject, and the second clause indicates the action itself and the object of the perception: Genesis 18:2

‫ֹלשׁה ֲאנָ ִשׁים נִ ָצּ ִבים ָע ָליו‬ ָ ‫וַ יִּ ָשּׂא ֵעינָ יו וַ יַּ ְרא וְ ִהנֵּ ה ְשׁ‬ And he lifted up his eyes and looked, and, here, three men stood near him.

The first clause, ‫וַ יִּ ָשּׂא ֵעינָ יו‬, indicates that Abraham “activated” his eyes and thus started perceiving something, whereas the following clauses, ‫ֹלשׁה‬ ָ ‫וַ יַּ ְרא וְ ִהנֵּ ה ְשׁ‬ ‫אנָ ִשׁים‬, ֲ indicate that he saw three wayfarers. In this verse Abrahamʼs name is not mentioned, but his name is stated in the second instance: Genesis 22:13

‫אַחר נֶ ֱא ַחז ַבּ ְסּ ַבְך‬ ַ ‫ת־עינָ יו וַ יַּ ְרא וְ ִהנֵּ ה־אַיִ ל‬ ֵ ‫אַב ָר ָהם ֶא‬ ְ ‫וַ יִּ ָשּׂא‬ And Abraham lifted up his eyes, and looked, and here, a ram (‫)איל אחר‬, caught in the thicket.

These cases are structurally different from the examples in Deuteronomy and Isaiah. Deuteronomy 4:19

‫כֹּוכ ִבים‬ ָ ‫ת־ה‬ ַ ‫ת־היָּ ֵר ַח וְ ֶא‬ ַ ‫ת־ה ֶשּׁ ֶמשׁ וְ ֶא‬ ַ ‫ית ֶא‬ ָ ‫ן־תּ ָשּׂא ֵעינֶ יָך ַה ָשּׁ ַמיְ ָמה וְ ָר ִא‬ ִ ‫וּפ‬ ֶ Lest you lift up your eyes unto heaven, and will see the sun and the moon and the stars.

In this case, in addition to the “activation of the eyes,” the first clause mentions the direction, ‫ה ָשּׁ ַמיְ ָמה‬, ַ which does not appear in this position in Genesis–Numbers. In the divine instruction to Abraham the direction is indicated in the second clause, in the position of the object perceived:

the rāʾâ clause; for the next clause pair uses the second nāśāʾ formula: ‫קֹולם וַ יִּ ְבכּוּ‬ ָ ‫וַ יִּ ְשׂאוּ‬. Two cases involve textual variants: Exod 14:10 (Samaritan Pentateuch/LXX); 2 Sam 24:16 (4QSama/4Q51 and 1 Chr 21:16).

Oral Platform and Language Usage in the Abraham Narrative

425

Genesis 13:14

‫ר־אַתּה ָשׁם ָצפֹנָ ה וָ נֶ גְ ָבּה וָ ֵק ְד ָמה וָ יָ ָמּה‬ ָ ‫ן־ה ָמּקֹום ֲא ֶשׁ‬ ַ ‫ָשׂא נָ א ֵעינֶ יָך ְוּר ֵאה ִמ‬ Lift up your eyes and look out from where you are, to the north and south, to the east and west.

This distinction is not only structural. When the indication of the direction is linked to nāśāʾ, this verb indicates the maneuvering of the face in a certain direction and thus assumes a full semantic meaning, much like the phrase ‫וַ יִּ ָשּׂא‬ ‫ל־ה ַחלֹּון‬ ַ ‫פנָ יו ֶא‬, ָ “He looked up toward the window” (2 Kgs 9:32). By contrast, in the formula as used in Genesis–Numbers (and in the Ugaritic texts), the meaning of the verb is bleached. It loses the sense of causative upward movement and is reduced to the meaning “activation.” A similar structure appears in the divine command to Moses: Deuteronomy 3:27

‫וּמזְ ָר ָחה ְוּר ֵאה ְב ֵעינֶ יָך‬ ִ ‫ימנָ ה‬ ָ ‫ֲע ֵלה רֹאשׁ ַה ִפּ ְסגָּ ה וְ ָשׂא ֵעינֶ יָך יָ ָמּה וְ ָצפֹנָ ה וְ ֵת‬ Go up unto the top of Pisgah, and lift up your eyes westward, and northward, and southward, and eastward, and look with your eyes.

Here the first clause indicates the direction, ‫וּמזְ ָר ָחה‬ ִ ‫ימנָ ה‬ ָ ‫יָמּה וְ ָצפֹנָ ה וְ ֵת‬. ָ whereas the second clause contains the instrument of perception, ‫ב ֵעינֶ יָך‬, ְ which in Genesis– Numbers appears in the first clause only. The repetition indicates that the term eye has its status reduced to that of an element in a fixed structure.81 Similar constellations are found in prophetic literature: Jeremiah 3:2

)‫(שׁ ַכּ ְב ְתּ‬ ֻ ‫ל־שׁ ָפיִ ם ְוּר ִאי ֵאיפֹה לֹא שגלת‬ ְ ‫י־עינַ יִ ְך ַע‬ ֵ ‫ְשׂ ִא‬ Lift up your eyes to the bare heights, and see: Where have they not lain with you?

The direction, ‫ל־שׁ ָפיִ ם‬ ְ ‫ע‬, ַ is indicated in the opening clause. The verb ‫ ְוּר ִאי‬does not have an object attached to it: the question ‫ ֵאיפֹה לֹא שגלת‬forms an independent clause.82 The same structure is revealed in an Isaianic saying: Isaiah 40:26 Lift up your eyes and see: Who created these?

‫י־ב ָרא ֵא ֶלּה‬ ָ ‫יכם ְוּראוּ ִמ‬ ֶ ֵ‫אוּ־מרֹום ֵעינ‬ ָ ‫ְשׂ‬

The direction, ‫מרֹום‬, ָ is indicated in the opening clause. The second clause lacks a proper object. The question ‫י־ב ָרא ֵא ֶלּה‬ ָ ‫ ִמ‬forms an independent clause: “Lift up your eyes on high and see: Who created these?” (NRSV).83 The same use of rāʾâ appears in the second instance:  See also Num 6:26 (‫)יִ ָשּׂא יְ הוָ ה ָפּנָ יו ֵא ֶליָך‬. In the phrase ‫ל־יֹוסף‬ ֵ ‫יה ֶא‬ ָ ֶ‫ת־עינ‬ ֵ ‫ת־אד ֹנָ יו ֶא‬ ֲ ‫וַ ִתּ ָשּׂא ֵא ֶשׁ‬ (Gen 39:7) the term ‫יה‬ ָ ֶ‫ ֵעינ‬occupies the position of ‫ פנים‬in the former idiom. 82  However, the Vulgate (ubi non prostrata sis, in the conjunctive) considers this question as a subordinate clause, in indirect discourse. 83  Tg. Neb. creates a connection by adding a verb: ‫ברא ִא ֵלין‬ ָ ‫דחל ֳק ָדם ַמן (!) ִד‬ ַ ‫וַ חזֹו ְל ִמ‬. The Peshitta construes ‫ מי‬as the direct object of ‫( וראו‬similarly Kimhi). 81

426

Frank H. Polak

Isaiah 49:1884

‫אוּ־לְך‬ ָ ‫י־ס ִביב ֵעינַ יִ ְך ְוּר ִאי ֻכּ ָלּם נִ ְק ְבּצוּ ָב‬ ָ ‫ְשׂ ִא‬ Lift up your eyes around and see: They are all assembled, have come to you!

Once again, the stretch following rāʾâ is not dependent on the verb.85 The first clause indicates where to look and thus reveals the same structure as Isa 40:26. Interestingly, this section includes a phrase that is similar to the formula and contains a direct object. But in this instance the place of rāʾâ is taken by hibbîṭ: Isaiah 51:6

‫אָרץ ִמ ַתּ ַחת‬ ֶ ‫ל־ה‬ ָ ‫יכם וְ ַה ִבּיטוּ ֶא‬ ֶ ֵ‫ְשׂאוּ ַל ָשּׁ ַמיִ ם ֵעינ‬ Raise your eyes to the heavens, And look upon the earth beneath.

The only saying that adheres to the formulaic structure of patriarchal narrative is found in the Jeremianic diatribe:86 Jeremiah 13:20 Lift up your eyes and see those coming from the north.

‫יכם ְוּראוּ ַה ָבּ ִאים ִמ ָצּפֹון‬ ֶ ֵ‫ְשׂאוּ ֵעינ‬

A group of formulae that is matched by recurrent phrases in Isaiah 40–66 contains the verb hištaḥăwâ (see 2.2, above), with eight instances in Deutero- and Trito-Isaiah. In some cases the verb serves as a predicate in an independent clause (44:15; 45:14) or in a final clause (66:23). Two passages contain a twoclause pattern: Isaiah 44:17

‫וּשׁ ֵא ִריתֹו ְל ֵאל ָע ָשׂה ְל ִפ ְסלֹו יִ ְסגָּ ד־לֹו וְ יִ ְשׁ ַתּחוּ‬ ְ The rest of it he makes into a god, his idol, bows down to it and worships. Isaiah 46:6 He makes it into a god; they bow down and worship.

‫וְ יַ ֲע ֵשׂהוּ ֵאל יִ ְסגְּ דוּ אַף־יִ ְשׁ ַתּ ֲחוּוּ‬

Here the verb preceding hištaḥăwâ is sāgad, a loanword borrowed from Aramaic. This phrase, then, is quite unlike the epic formulae of patriarchal narrative. In other places the prophetic logos uses parallelism where the narrative has a consecutive formula: Isaiah 44:15

‫ד־למֹו‬ ָ ָ‫ל־אל וַ יִּ ְשׁ ָתּחוּ ָע ָשׂהוּ ֶפ ֶסל וַ יִּ ְסגּ‬ ֵ ‫אַף־יִ ְפ ַע‬ Then he makes a god of it and worships it, Fashions an idol and bows down to it!  This stretch appears again in Isa 60:4.  In the LXX πάντας / ‫ ֻכּ ָלּם‬is attached to καὶ ἴδε and separated from the next clause (ἰδοὺ συνήχθησαν καὶ ἤλθοσαν πρὸς σέ), whereas the Peshitta subordinates the clause to ‫וּר ִאי‬. ְ The Tg. Neb. and the Vulgate follow the MT. The LXX to 60:4 reveals a similar structuring in a different wording. 86  The LXX reading of Jer 13:20 implies the plus ‫ירושלם‬, in the vocative, as found in Jer 13:27; 4:14; 6:8; 15:5. 84 85

Oral Platform and Language Usage in the Abraham Narrative

427

Thus we note the parallel with qûm, as in the formula ‫ השתחוה‬. . . ‫קום‬: Isaiah 49:7

‫ְמ ָל ִכים יִ ְראוּ וָ ָקמוּ ָשׂ ִרים וְ יִ ְשׁ ַתּ ֲחוּוּ‬ Kings shall see and stand up; nobles, and they shall bow down.

In this case, ‫ וְ יִ ְשׁ ַתּ ֲחוּוּ‬parallels ‫וָ ָקמוּ‬, unlike the formulaic pattern in which it continues the first verb (Gen 19:1; 23:7; 1 Sam 25:41). One also encounters a fully formulaic colon in parallelism with another formulaic phrase, known from poetic contexts:87 Isaiah 49:23

‫אַפּיִ ם ֶא ֶרץ יִ ְשׁ ַתּ ֲחווּ ָלְך‬ ַ ‫וַ ֲע ַפר ַרגְ ַליִ ְך יְ ַל ֵחכוּ‬ With their faces to the ground they shall bow down to you, and the dust of your feet they will lick.

The first colon exhibits the formula ‫( וישתחו אפים ארצה‬Gen 19:1; 42:6).88 It should be noted, however, that the Isaianic phrase differs from the narrative formula regarding morphology and word order. In the prophetic saying, ‫ארץ‬ ‫ אפים‬opens the colon, whereas the formulaic phrase always opens with the verb. The phrase ‫ ארצה אפים‬is marked by the directional suffix -h, which is lacking in the prophetic saying.89 Thus the Isaianic phrase derives from the formulaic register but deviates from its usage. The deviation is noteworthy, since a significant decline in the use of the directional -h is one of the characteristics of Late Biblical Hebrew.90 One also notes a striking case of parallelism: Isaiah 60:14

‫ל־מנַ ֲא ָציִ ְך‬ ְ ‫ל־כּפֹּות ַרגְ ַליִ ְך ָכּ‬ ַ ‫חֹוח ְבּנֵ י ְמ ַענַּ יִ ְך וְ ִה ְשׁ ַתּ ֲחווּ ַע‬ ַ ‫וְ ָה ְלכוּ ֵא ַליִ ְך ְשׁ‬ Bending low before you, they shall come, the children of those who tormented you; prostrate at the soles of your feet they shall be, all those who reviled you.

The phrase ‫ל־כּפֹּות ַרגְ ַליִ ְך‬ ַ ‫ וְ ִה ְשׁ ַתּ ֲחווּ ַע‬is preceded by the parallel colon ‫וְ ָה ְלכוּ ֵא ַליִ ְך‬ ‫חֹוח‬ ַ ‫( ְשׁ‬paronomasia). Notably, the phrase ‫ ַכּפֹּות ַרגְ ַליִ ְך‬bears a certain affinity to the Ugaritic formula lpʿn il thbr wtql (“at the feet of Ilu she bowed down and  See Mic 7:17; Ps 72:9.  See nn. 55–58, above. 89  The phrase without h is also found in 1 Kgs 1:31, but the LXX has ἐπὶ πρόσωπον ἐπὶ τὴν γῆν. 90  In Isa 40–66, one notes the directional -h in 43:14 (‫;)ּב ֶב ָלה‬ ָ 55:10; 65:9 (‫שׁ ָמּה‬, ָ twice). The decline in the use of this form, followed by its virtual disappearance in Qumran and Rabbinic Hebrew, is analyzed by J. Joosten, “The Distinction between Classical and Late Biblical Hebrew as Reflected in Syntax,” HS 46 (2005), 327–339, here 337–338; see also E. Y. Kutscher, The Language and Linguistic Background of the Isaiah Scroll (1Q Isaa) (Leiden: Brill, 1974), 414; I. Young et al., Linguistic Dating of Biblical Texts: An Introduction to Approaches and Problems (2 vols.; London: Equinox, 2008), 1:42, 273. Young et al. (80–81) demur that this is “a matter of statistical divergence or accumulation,” but this divergence is exactly the point: frequency is of interest because of its implications for structure. 87 88

428

Frank H. Polak

fell,” CTU 1.4 IV:25–26) and a related Hebrew phrase, ‫ל־רגְ ָליו‬ ַ ‫“( וַ ִתּפֹּל ַע‬she fell at his feet”; 1 Sam 25:24; 2 Kgs 4:37). But the prophetic saying includes the ballast variant ‫ ַכּפֹּות‬not used in the narrative formula. The prophetic use of formulaic language, then, is based on the traditional formulaic register as used by patriarchal narrative. The Abraham narrative thus shares many features with the Northwest Semitic epic tradition as represented by the texts found in Ugarit. In other words, the Abraham narrative stands in the tradition stream of the Northwest Semitic epic. According to its stylistic-syntactic profile, its substratum is at home in the oral arena. Thus the sociocultural context in which this narrative came into being is significantly different from the circumstances in which the Priestly and Deuteronomic literature was created, since these corpora bear the imprint of the scribal desk. 2.4 Sociocultural Background and Periodization It is often assumed that the scribal style of the Deuteronomic, Priestly, and Persian-era corpora always coexisted side by side with the lean, brisk style of the oral-written corpus. Susan Niditch posits an oral-written continuum in which the writing author may imitate the style of oral literature,91 whereas Ian Young, Robert Rezetko, and Martin Ehrensvärd describe the distinction between various styles as a matter of stylistic choice.92 In a different vein, David Carr views orality as an element in the scribal education, and presumes that texts were composed in writing and were promulgated among the nonliterate population by word of mouth. In his view, literary composition in the ancient Near Eastern context  S. Niditch, “Epic and History in the Hebrew Bible: Definitions, ‘Ethnic Genres’, and the Challenges of Cultural Identity in the Biblical Book of Judges,” in Epic and History (ed. D. Konstan and K. A. Raaflaub; Oxford: Blackwell, 2010), 86–102, here 91. 92  Young et al., Linguistic Dating (see n. 90), 1:96; 2:81–83. Some of their criticism seems misleading. I am incorrectly represented as arguing against the thesis of an oral-written symbiosis. However, in my view the problem is how to explain the fact that the sōpərîm of a certain corpus adopt a style that is close to orality, whereas the literati responsible for three other corpuses do not adopt a style of this kind. Thus, the problem is how to explain why a certain symbiosis is not found in the Judean/Achaemenid corpora. Another element of the discussion is the transition from written to oral, which they posit (1:96–97) in view of a similar transition in Somali. Unfortunately, their description of Somali evidence omits a few crucial points. Analysis of the newspapers during the third stage of the transition period (1987–1989) revealed a certain decline in the production of written texts, as newspapers became shorter and less professional (including the printing itself), because of the difficult socioeconomic conditions during the civil war; see D. B iber and M. H ared , “Linguistic Correlates of the Transition to Literacy in Somali: Language Adaptation in Six Press Registers,” in Sociolinguistic Perspectives on Register (ed. D. Biber and E. Finegan; New York: Oxford University Press, 1994), 182–216, here 193–194. Since this process occurred during a transition period, because of specific conditions, it does not imply universal reversibility of stylistic transitions. 91

Oral Platform and Language Usage in the Abraham Narrative

429

always centers on the scribal school; the oral channel is subservient to the scribal institutions.93 These presumptions, however, are too strong. Particular attention is due to the Hurro-Hittite Song of Release, a large-scale epic narrative with connections to a historical context: the seventeenth-century conquest of Ebla by the Hittites. The colophons of two tablets of this text refer to a singer (NAR).94 Since the colophon is the place where the copyist establishes the authority of the transmitted text, the mention of the singer is highly significant, even though both colophons are incomplete. Here, then, we have a rare indication of the oral roots of an ancient Near Eastern written text. I have already pointed to the oral-formulaic background of the Ugaritic texts of Baʿal and ʾAqhat. Thus, the oral arena can hold its own and is not of necessity subservient to the scribal desk. How, then, to describe the relation between these two orbits? In my view the crucial point is the social institution. The social context of the scribal IES is the royal chancery, with its status, prestige, and authority95. The intricate, elaborate scribal style highlights the scribeʼs habitus and expertise and, thereby, his position within the royal bureaucracy and the society it governs96. 93  D. M. C arr , Writing on the Tablet of the Heart: Origins of Scripture and Literature (Oxford: Oxford University Press, 2005), 9–20. 94  E. Neu, Das Hurritische Epos der Freilassung, I (SBOT 32; Wiesbaden: Harrassowitz, 1996), 227, 272. An evaluation of this text is offered by M. R. B achvarova , “The Eastern Mediterranean Epic Tradition from Bilgames and Akka to the Song of Release to Homerʼs Iliad,” Greek, Roman, and Byzantine Studies 45 (2005), 131–153, here 132–138; see also A. Archi, “Transmission of Recitative Literature by the Hittites,” AoF 34 (2007), 185–203; idem , “Orality, Direct Speech and the Kumarbi Cycle,” AoF 36 (2009), 209–229. 95  See, in general, K. van der Toorn , Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007), 51–82; and in postbiblical context, C. A. R ollston , “Ben Sira 38:24–39:11 and the ‘Egyptian Satire of the Trades’: A Reconsideration,” JBL 120 (2001), 131–139. The status of the chief scribe at Emar is indicated by his prominent participation in the state cult and the meat cuts he received; see Y. C ohen , The Scribes and Scholars of Emar: Ancient Scribal Education in a Late Bronze Age City (HSS 59; Winona Lake, IN: Eisenbrauns, 2009), 41–42, and in general, 26–42, 89–91, 117–119, 147–158. Cohen emphasizes that the role of the scribes was not limited to the royal administration: they worked in various social contexts (90, 118). The role of the scribes in Ugarit is studied by I. Márquez Rowe, The Royal Deeds of Ugarit: A Study of Ancient Near Eastern Diplomatics (AOAT 335; Münster: Ugarit Verlag, 2006), 99–137. The status of (or rather, claimed by) the royal scribes in the Neo-Assyrian empire has been established by S. Parpola, “The Assyrian Cabinet,” in Vom Alten Orient zum Alten Testament: Festschrift für Wolfram Freiherrn von Soden zum 85. Geburtstag (ed. M. Dietrich and O. Loretz; AOAT 240; Kevelaer: Butzon & Bercker; Neukirchen-Vluyn: Neukirchener Verlag, 1995), 379–401. On literacy and bureaucracy in the Persian Empire, see J. Ray, “Literacy and Language in Egypt in the Late and Persian Periods,” in Literacy and Power in the Ancient World (ed. A. K. Bowman and G. Woolf; Cambridge: Cambridge University Press, 1994), 51–66. 96  The role of scribal culture in Deuteronomy has been highlighted again by van der Toorn , Scribal Culture (see n. 95), 143–149, 162–172, 182–194; J.-P. S onnet , The Book within the Book: Writing in Deuteronomy (BibInt 14; Leiden: Brill, 1997).

430

Frank H. Polak

By contrast, in its close resemblance to spoken discourse, the lean, brisk style implies quite a different social context and social position. The social context is the arena of the singer of tales. Use of the VoLB positions the writing author in the context of the oral performance. The author thereby recognizes the prestige of a branch of literature for which not literacy and scribal know-how but the artistry of the performer, whether literate or not, are determinant. This feature is in keeping with the literary motifs: formal steps such as an international agreement, the acquisition of real estate, and marriage are performed by such gestures as a common meal and the presentation of gifts rather than written contract (Gen 21:22–34; 24:53–54; 26:26–31; 29:18–29; 31:44–54).97 The reluctance to represent the honored ancestors as writing points to a difference in social context rather than merely in literary taste. This is a society in which the grandees and magnates are not necessarily fully literate and in which the written text does not enjoy the prestige it has in a society that is dominated by a developed bureaucracy. In contrast, in such a context the oral literacy of the kingʼs men is of greatest weight: as is shown by large sections of the royal correspondence from Mari, Zimri-Limʼs leading envoys and army commanders formulate their reports of contact with other leaders, such as Hammurabi, as narrative, even including dialogues. In my view this is the social background presupposed by the oral-written corpus98. These considerations are in keeping with the magic-mythic, dynamistic, and sometimes almost animistic themes found in many of the tales concerning Abraham and Jacob, such as the tale of the oaks at Mamre (Gen 18:1–15),99 the oath performed by Abrahamʼs servant (24:2–3, 9), the confirmation of Isaacʼs blessing in spite of Jacobʼs trickery (27:25–35), and Jacobʼs agreement with Laban (31:46–51, 53).100 Many of these motifs are central to the divine promises to the patriarchs, which thus contain clear residues of the ancient domestic cult. These considerations form ample justification for the thesis that the oral-written corpus, and in particular the  S. Gandz, “Oral Tradition in the Bible,” in Jewish Studies in Memory of George A. Kohut (ed. S. W. Baron and A. Marx; New York: The Alexander Kohut Memorial, 1953), 249–269, here 249–250. On the early medieval habit of using symbolic objects as proof for a title or a deed, see M. T. Clanchy, From Memory to Written Record: England, 1066–1307 (2nd ed.; Oxford: Blackwell, 1993), 35–42; idem , “Reading the Signs at Durham Cathedral,” in Literacy and Society (ed. K. Schousboe and M. T. Larsen; Copenhagen: Akademisk Verlag, 1989), 171–182, here 174–178, 181. 98  See P olak , “Sociolinguistics as Key” (see n. 28), 156–159; idem , “Negotiating with Ḫammu-rāpi: A Case Study,” in Birkat Shalom: Studies in the Bible, Ancient Near Eastern Literature and Postbiblical Judaism Presented to Shalom M. Paul (ed. C. Cohen et al.; Winona Lake, IN: Eisenbrauns, 2008), 643–665, here 652–661, 665. 99  See n. 14 above. 100  Important comments are offered by R. A lbertz , Religionsgeschichte Israels in alttestamentlicher Zeit (2 vols.; Göttingen: Vandenhoeck & Ruprecht, 1992), 1:52–54. For additional details, see Polak , “Style Is More Than the Person” (see n. 33), 67–79; idem , “Abraham-Jacob Narrative” (see n. 1), 234. 97

Oral Platform and Language Usage in the Abraham Narrative

431

main strata of patriarchal narrative, came into being before the formation of the Judean corpus.

3 Questions concerning the Power of Linguistic Data 3.1 Linguistic Analysis and Textual Transmission Many scholars have expressed doubts concerning the validity of the use of linguistic features as indications for literary-historical periodization.101 Carr argues that the transmission process of the text naturally entails the steady intrusion of elements that reflect the language state of the period of copying rather than that of primary composition.102 However, this argument, though valid for textual detail, fails to account for the large-scale differentiation between the various corpora as a whole. A case in point is the use of the object suffix. An analysis of the uses of this morpheme attached to the finite verb (e.g., ‫)וַ יַּ נִּ ֵחהוּ‬, as opposed to the object particle with suffix (e.g., ‫)א ָֹתם‬, indicates systematic contrasts between large corpora. In the Priestly texts the object particle prevails (2:1 ratio), but in the Achaemenid corpus this form is extremely rare (1:9 ratio).103 A similarly low ratio is found for a number of long texts from the Judean Desert. Thus the findings for Qumran Hebrew dovetail with Late Biblical Hebrew rather than with the Priestly tradition. This correspondence does not suffer explanation by textual transmission, which does not affect the Qumran texts in the same way that it affects the biblical text. Thus the vagaries of textual transmission do not obliterate the systematic contrast between the different syntactic-stylistic profiles of the various corpora.104 101  Y oung et al., Linguistic Dating (see n. 90), 1:95–102; 2:81–83; E. B en Z vi , “The Communicative Message of Some Linguistic Choices,” in A Palimpsest: Rhetoric, Ideology, Stylistics and Language Relating to Persian Israel (ed. E. Ben-Zvi et al.; Piscataway, NJ: Gorgias Press, 2009), 269–290, here 289–290; D. H. Aaron, “Reflections on a Cognitive Theory of Culture and a Theory of Formalized Language for Late Biblical Studies,” in Remembering Biblical Figures in the Late Persian and Early Hellenistic Periods: Social Memory and Imagination (ed. E. Ben-Zvi and D. V. Edelman; Oxford: Oxford University Press, 2013), 451–473. However, the recognition of Classical Biblical Hebrew as “cultural capital,” or as shared representation of cultural memory/identity, relates to high-level abstraction, whereas language usage, even when it is fed by the aspirations involved in this“ cultural capital,” is affected no less by the concrete social situation, which in the present case means the sociological status of the chancery, the power it imparts, and the language usage it promotes. These matters are glossed over by Ben-Zvi, “Communicative Message,” 279, and Aaron, “Reflections on a Cognitive Theory,” 465–467 (in spite of his references to social conditions, 463–465). 102  D. M. C arr , The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2009), 126–127. 103  See below, pp. 432–433. 104  See Polak, “Samuel and the Deuteronomists” (see n. 29), 69–70.

432

Frank H. Polak

3.2 The Validity of Linguistic Distinctions Young, Rezetko, and Ehrensvärd argue that the distinction between Classical Hebrew (EBH) and the language of the texts composed in the late Babylonian/ Persian era (LBH) is merely a matter of stylistic choice.105 However, their discussion of linguistic differentiation is not unproblematic.106 A prooftext for their view is Pesher Habakkuk (1QpHab). This text includes a number of terms that are definitely tied to texts from the Persian era and later, such as ‫גמר הקץ‬, “the end of the period” (1QpHab 7:2). The noun gəmār, “end,” common in Aramaic, rabbinic Hebrew, and Qumran Hebrew, is unattested in Biblical Hebrew (unlike the verb gāmar, Ps 12:2; 77:9). The second noun is qēṣ, which in the Qumran texts has the meaning “aeon, period.”107 However, since this usage is not found in characteristic LBH texts (in spite of ‫קץ ַה ְפּ ָלאֹות‬,ֵ “the period of horrors,” Dan 12:6), Young does not view the phrase ‫ גמר הקץ‬as a postclassical feature.108 Likewise, no count is taken of the phrase ‫( עושי התורה‬1QpHab 7:11; 8:1; 12:5).109 This phrase is not biblical, but the use of tôrâ as object of ʿāśâ is balanced by perfect LBH texts (Neh 9:34; 2 Chr 14:3; and in a different sense, Num 5:30). Other details in the revisionist thesis are similarly problematic. For instance, since forms of the object particle with the suffix (OPS) are far more frequent than object suffixes in Ezekiel, whereas in LBH object suffixes with the verb (VOS) are frequent, the high frequency of object suffixes has been viewed as a hallmark of LBH.110 The revisionist analysis adopts this view and counts the high frequency of the object suffix everywhere as a LBH feature.111 However, the preference for the VOS is not limited to LBH. It is true that in the Achaemenid corpus the ratio is 9:1, with similar figures for two of the lengthy Qumran texts,112 105  I. Young, “Late Biblical Hebrew and the Qumran Pesher Habakkuk,” Journal of Hebrew Scriptures 8 (2008), http://jhsonline.org/Articles/article_102.pdf (accessed 12/29/14); Young et al., Linguistic Dating (see n. 90), 1:139, 141. 106  For methodological criticism, see J. Joosten, review of Linguistic Dating of Biblical Texts, by I. Young and R. Rezetko, with the assistance of M. Ehrensvärd, Babel und Bible 6 (ed. L. Kogan; Winona Lake, IN: Eisenbrauns, 2012), 535–542; A. Hornkohl, Ancient Hebrew Periodization and the Language of the Book of Jeremiah: The Case for a Sixth-Century Date of Composition (Studies in Semitic Languages and Linguistics 74; Leiden: Brill, 2014), 27–50. 107  See J. J. Collins, A Commentary on the Book of Daniel (Hermeneia 27; Minneapolis: Fortress, 1993), 338. 108  1QpHab 5:7–8; 7:7–8, 12–13; see Young, “Pesher Habakkuk” (see n. 105), 36; Young et al., Linguistic Dating (see n. 90), 1:260–261. 109  So also 4Q171 f1 2ii:14, 22; compare 4Q470 f1 4. Similarly: ‫ועשו ;לעשות דבר מן התורה‬ ‫התורה‬ ֯ ‫לעשות כפרוש התורה ;את כול‬. 110  Young et al., Linguistic Dating (see n. 90), 1:261. 111  Y oung et al., Linguistic Dating (see n. 90), 1:134, n. 59, 1:135, n. 71, 1:262. This preference is also classified as LBH in such poetic texts as Ps 18 (1:135, n. 74). 112  I have checked the data for the Community Rule (1QS; forty-one VOS / two OPS; with the overlapping parallel texts, 4Q255–264: thirty-three VOS / no OPS), the Damascus Covenant (forty-five VOS / seven OPS), with the overlapping parallel texts 4Q266–273 (sixty-eight

Oral Platform and Language Usage in the Abraham Narrative

433

but object suffixes are the norm in the Ugaritic texts and in biblical poetry.113 In epigraphic Hebrew texts we encounter twenty-seven instances, in contrast to six cases of the OPS (ratio 4.5:1). In narrative prose in the oral/written corpus (and the Deuteronomic legislation), the ratio of the VOS compared to the OPS ranges from 3:1 to 4:1. For Deuteronomistic historiography, I note ratios of 2:1 and 1:1, whereas for the Priestly narrative and legislation the ratio is 1:2.5. These complex findings need careful analysis. Indeed, our understanding of the different varieties of Biblical Hebrew will be advanced by heedful analysis of the complex data rather than easy rejection of the distinction between the various strata. 3.3 Deutero-Isaiah and the Transition to the Style of the Persian Era Thus I want to state once again that global differences in language usage indicate large-scale differences in the social context in which language is used and thus in the society to which the language users belong.114 The development of a dominant bureaucracy has a profound impact on the syntactic-stylistic profile of the writings of the Deuteronomic/Deuteronomistic school and the Priestly circles (the Judean corpus). The Babylonian conquest and Persian domination involve the use of Aramaic as the official language and thus as the language of scribal education (and of all contacts with the administration), resulting in a bilingual society in which Hebrew writing, as found in the Achaemenid corpus, is profoundly influenced by Aramaic and by the Hebrew vernacular. The new conditions already shine through in the language used in Lamentations,115 which thus marks the inception of the transition to the literary language of the Babylonian/Persian era. Do we have the right to posit a period in which the knowledge of the classical language was still strong enough to enable the narrator to adhere to this language stratum? The possibility that a classical language stratum maintained itself, even among the exiles in Babylonia, seems viable in light of the stylistic nature of the prophecy of Deutero-Isaiah, in which the classical usage still seems VOS / nine OPS), the War Scroll (1QM; thirteen VOS / two OPS), with the overlapping texts 4Q285, 4Q491, 4Q492, 4Q493, 4Q494, 4Q495, 4Q496, 11Q14 (eight VOS / one OPS), and the Temple Scroll (11Q19–21; seventy-one VOS / twenty-six OPS; the data for this scroll are to a large extent conditioned by the quotations from pentateuchal literature). 113  The OPS is not found in archaic poetry (with twenty-one cases of the object particle with noun, apart from Num 23:10), whereas VOS is frequent For reasons of space, the statistical details will be published elsewhere. 114  I mean the supraregional and literary spheres of language usage rather than regional/ local language. 115  See F. W. Dobbs-Allsopp, “Linguistic Evidence for the Date of Lamentations,” JANES 26 (1998), 1–36, and P olak , “Sociolinguistics and the Judean Speech Community in the Achaemenid Empire,” in Judah and Judeans in the Persian Period (ed. O. Lipschits and M. Oeming; Winona Lake, IN: Eisenbrauns, 2006), 589–628, here 596–606, 614–617.

434

Frank H. Polak

strongly developed. Nevertheless one also notes a certain weakening of this tradition. Shalom Paul points to a number of Aramaisms and deviations ִ in the meaning “tax” (Isa from classical usage, such as, for example, ‫מ ָּדה‬, 45:14), ‫“( וְ ִה ְתא ָֹשׁשׁוּ‬be strong,” Isa 46:8), ‫“( סבל‬to carry,” Isa 46:4, 7; 53:4, 11; see Qoh 12:5; Lam 5:7), ‫“( סגד‬to bow down,” Isa 44:15, 17, 19; 46:6; see above), ‫ בחר‬in the meaning “to test” (Isa 48:10; see Sir 4:17 and Tg Neb. to Judg 7:4; Jer 17:10), ‫“( מערב‬west,” Isa 43:5; 45:6, as against classical ‫)ים‬, the plural ‫“( עולמים‬eternity,” Isa 45:17; 51:9; Ps 145:13; Qoh 1:10; and in Aramaic, Dan 2:4), and ‫“( ָׁש ָרב‬parched ground,” Isa 49:12).116 Deutero-Isaiah, then, represents a transition from Classical to Late Biblical Hebrew. Seoung Yun-Sin reaches a similar conclusion for the books of Haggai, Zechariah, and Malachi.117 A transition is also suggested by the use of verbs indicating motion. A study of the use of these verbs in the books of Haggai–Malachi shows a decrease in the use of the verbs hālak (qal) and lāqaḥ and an increasing frequency of yāṣāʾ (qal) and bôʾ (qal and hifil), all as in narrative from the Persian era.118 A similar tendency can be noticed in Isa 40–48; 49–55. In Isa 40–48, hālak does not appear, and there are a mere two instances in Isa 49, in contrast to a significant number of occurrences of yāṣāʾ and bôʾ. For lāqaḥ one notes a similar decrease in Isa 40–48 but not in chapters 49–55. These figures indicate that the diction of Isa 40–55 stands midway between the classical style and Persian-era language usage.119 Hence I conclude that a discussion of various doubts concerning the value of linguistic analysis only underscores the validity of this approach. Additional problems relate to the sociohistorical background presupposed by the Abraham narratives. This will be the subject of the ensuing paragraphs.

116  See S. M. Paul , “Signs of Late Biblical Hebrew in Isaiah 40–66,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Winona Lake, IN: Eisenbrauns, 2012), 293–299. 117  S. Yun-Sin, “A Lexical Study on the Language of Haggai–Zechariah–Malachi and its Place in the History of Biblical Hebrew” (PhD diss., Hebrew University of Jerusalem, 2007). 118  See F. P olak , “Verbs of Motion in Biblical Hebrew: Lexical Shifts and Syntactic Structure,” in A Palimpsest: Rhetoric, Ideology, Stylistics and Language Relating to Persian Israel (ed. E. Ben-Zvi et al.; Piscataway, NJ: Gorgias Press, 2009), 161–197, here 168–174. The sample of the preexilic prophetic texts includes Amos; Hosea; Isa 1–11; 14; 17–22; 28–32; Micah; Zephaniah; Nahum; Habakkuk. 119  For reasons of space, the statistical details for Isa 40–55; 56–66 will be published elsewhere. Let me only note that the language of Isa 56–66 is even closer to the style of Persian era prophetic texts.

Oral Platform and Language Usage in the Abraham Narrative

435

4 The Abraham Tradition in Exilic Prophetic Literature The Abraham narrative is not mentioned in prophetic literature before two prophetic speeches from the Babylonian or early Persian era.120 This fact has convinced John Van Seters, and many scholars with him,121 that the Abraham narrative was not known before the Babylonian period. However, this inference is highly problematic: narrators or poets do not introduce themes and characters in order to demonstrate their acquaintance with a given theme or character but because it fits their tale or poem, and only then.122 Hence we have to ask ourselves what the great anonymous author achieves by means of the allusions to the Abraham narrative: “But you, Israel, my servant, Jacob, whom I have chosen, the offspring of Abraham, my friend; you whom I took (‫)א ֶשׁר ֶה ֱחזַ ְק ִתּיָך‬ ֲ from the ends of the earth, and called from its farthest corners, saying to you, ‘You are my servant, I have chosen you and not cast you off’ ” (Isa 41:8–9).123 The phrase that presents the exiled Judeans as “the offspring of Abraham, my friend” is continued by an allusion to Abrahamʼs call “from the ends of the earth, from its farthest corners.” The phraseology has been explained in different ways. The most common exegesis views the “ends of the earth” as an allusion to Haran, in the far north of Mesoptamia, the place where the tradition located Abrahamʼs call. In contrast, Kimhi regards the entire relative clause (v. 9) as a prophetic description of the divine call to the exiles, who will be brought from the ends of the earth (Isa 43:6; 49:6). He relates the suffixes of ‫ ֶה ֱחזַ ְק ִתּיָך‬and the other verbs to the community addressed, interpreting the qatal forms as prophetic perfect. An interpretation along these lines (possibly as a performative) points to a fusion of the past call to Abraham and the present call to the exiles, much like the overlay in the address to “worm Jacob, men of Israel” in the continuation of this logos (Isa 41:14) and the fusion of the trek from Egypt and the flight from 120  I disregard Mic 7:20 (a late addition); Jer 33:20 (not represented in the LXX); and Isa 29:22. 121  Van S eters , Abraham (see n. 5), 310–311; idem , Prologue to History (see n. 12), 215–216, 239–242; M. Köckert, “Die Geschichte der Abrahamüberlieferung,” in Congress Volume: Leiden, 2004 (ed. A. Lemaire; VTSup 109; Leiden: Brill, 2006), 103–128, here 123–128 (with an ancient kernel of the Abraham narrative in Gen 13; 16; 18–19; see pp. 120–122). Less unequivocal is E. Blum, Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 358–359. This view is often based on the analysis of K. Galling, Die Erwählungstraditionen Israels (BZAW 48; Gießen: Töpelmann, 1928), 52–53, 56. But one should not disregard Gallingʼs discussion of the prophetic rejection of the ancestor tradition, an attitude that was broken by the exile and the separation of the leading parts of the people from its land (67). 122  R. Wellek and A. Warren, Theory of Literature (3rd ed.; Harmondsworth: Penguin, 1963), 258–259. Imagine the number of allusions there would be in modern literature if narrators/poets were obliged to allude to a significant number of known characters or works known to them. 123  See Köckert, “Die Geschichte der Abrahamüberlieferung” (see n. 121), 103–113.

436

Frank H. Polak

Babylon in the vision of the new exodus (Isa 41:18–19; 42:15–26; 43:19–20). A fusion of this kind would be extremely enigmatic if the narrative were not wellknown to the community. Its use in prophetic discourse implies the assumption that the audience would readily understand it and, moreover, respond to it well. Another case in point is the allusion to Abraham and Sarah: “Look to Abraham your father and to Sarah who bore you; for he was but one when I called him, but I blessed him and made him many” (Isa 51:2). This overt allusion is the introductory note of a speech that is meant to convince the exiles of the coming salvation: “The triumph I grant is near, the success I give has gone forth. My arms shall provide for the peoples; the coastlands shall trust in me, they shall look to my arm” (Isa 51:5, NJPS). The reference to Abraham and Sarah is a rhetorical argument for the truth of the prophetʼs message. Thus the audience must be acquainted with the story, and, what is more, the argument must be strong enough to move the listeners from utter despair to hope and confidence. Therefore, the narrative alluded to cannot be new. If it is to overcome despair and distrust it has to be sufficiently well established in order to fulfill its task.124 Moreover, the mention of “Sarah who brought you forth” is only effective if the audience is aware of the story of the promise to the childless couple.125 The point of the prophetic reference is the overturning of Sarahʼs despair and her giving birth against all odds (a theme that is closely allied to the complaint “quoted” in Isa 49:21). The prophet does not explain the reference to make it work. This way of speaking is only possible if the tale is well-known and highly regarded. Notably, the phrase “Raise your eyes to the heavens, And look upon the earth beneath” (Isa 51: 6) is based on the formula, ‫וישא עיניו וירא‬, “raising the eyes and see.” The imperative is used in the encouragement to Abraham: “Raise your eyes now, and look (‫)שׂא נָ א ֵעינֶ יָך ְוּר ֵאה‬ ָ [. . .] northward and southward and eastward and westward” (Gen 13:14; and see section 2.1). Our pericope, then, has definite associations with the Abraham narrative, in a wording not unlike the present text.

124  Similarly Blum, Vätergeschichte (see n. 121), 296. Blum (297) finds the roots for the primary promise stratum in Judea between 721 and 586. 125  See D. Rom-Shiloni, Exclusive Inclusivity: Identity Conflicts between the Exiles and the People Who Remained (6th–5th Centuries BCE) (LHBOTS 543; New York: Bloomsbury T&T Clark, 2013), 118–119; J. M uilenburg , “The Book of Isaiah 40–66,” IB 5:404, 591; B. Duhm, Das Buch Jesaia übersetzt und erklärt (3rd ed.; HKAT 3/1; Göttingen: Vandenhoeck & Ruprecht, 1914), 354. This point is duly acknowledged by T. Römer, “Abraham Traditions in the Hebrew Bible outside the Book of Genesis,” in The Book of Genesis: Composition, Reception and Interpretation (ed. C. A. Evans et al.; Leiden: Brill, 2012), 159–180, here 163– 167. For a redaction-historical discussion, see O. H. Steck, “Zions Tröstung: Beobachtungen und Fragen zu Jesaja 51:1–11,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Fest­schrift fur Rolf Rendtorff zum 65. Geburtstag (ed. E. Blum and C. Macholz; Neukir­chenVluyn: Neukirchener Verlag, 1990), 257–276, here 271–275.

Oral Platform and Language Usage in the Abraham Narrative

437

The second passage that demands our attention is the allusion to Abraham that Ezekiel put in the mouth of the people who remained in Judea: “Abraham was only one man, yet he got possession of the land; but we are many; the land is surely given us to possess” (Ezek 33:24).126 The allusion is all the more important since Ezekiel imagines its use by the remaining Judeans. The Abraham tradition is strong enough to allow Ezekiel to suppose that it is sufficiently well-known in Judea to underscore claims to real estate. This argument would enable the imagined Judeans to support their argument against the exiles, who previously were partners in the ownership of the land. There is a second presupposition as well: the tale must be sufficiently known among the exiles addressed, so as to make them feel the sting. Not only must the Abraham tale have been well-known and esteemed among the exiles and in the Judean fatherland, but it must also have been well established in the Judean tradition before the first exile, that is to say, before 597. If that were not the case, the prophet would hardly have been able to use it as an argument concerning real estate. Another point that demands our attention concerns the number of prophetic allusions to patriarchal narrative. In this context it would be a mistake to count every instance of the name Jacob, for this is no more than a synonym for Israel and thus would not have implications for the attestation of the Jacob narrative. We need references that point specifically to the narrative. Such specific referָ ‫ָּכ‬ ences are found in two texts. First, we have the Jeremianic allusion ‫ל־אח ָעֹקוב‬ ‫( יַ ְעקֹב‬Jer 9:3, “for every brother will utterly supplant,” ASV), and second the Hoseanic pointers to Jacobʼs flight to the “field of Aram,” his service there for the sake of a wife, and his struggle with the divine being (Hos 12:4, 13). When such allusions are that scarce, the “law of small numbers” must apply:127 it is only a matter of chance that other texts are not alluded to. Thus the fact that we do not find allusions to Abraham in prophetic texts other than the books of Ezekiel and Deutero-Isaiah does not justify the conclusion that the Abraham narrative was unknown before the exilic period. On the contrary, these texts indicate that in the exilic period this narrative was familiar to such an extent that even veiled references would be comprehensible to the exilic community. The Abraham narrative was part and parcel of the cultural memory of both the exilic and the Judean community.128 126  See Römer, “Abraham Traditions” (see n. 125), 164, 167–168; Rom-Shiloni, Exclusive Inclusivity (see n. 125), 146–148. 127  A. Tversky and D. Kahneman, “Belief in the Law of Small Numbers,” Psychological Bulletin 76 (1971), 105–110. 128  R ömer , “Abraham Traditions” (see n. 125), theorizes that these allusions may hark back to oral tradition. However, the narrative alluded to must have included the migration theme, the inheritance of the land, the theme of the childless couple, the divine promise and the overcoming of childlessness, and some traditional wording: in short, the quintessence of the Abraham narrative as it is transmitted in the Pentateuch.

438

Frank H. Polak

5 The Art of Biblical Narrative and the Oral Performance These considerations return me to the oral-epic platform of the Abraham narrative. Many features of biblical narrative have to be viewed in light of the performerʼs art. In the oral-derived corpus, the dialogue covers more than 40 percent of the text,129 only slightly less than in Homeric poetry, where dialogue is said to range between 40 and 50 percent.130 When we think in terms of performance, the dialogue is of particular weight for the narrator-performer, since it enables him to give his figures different voices, according to their roles, characters, and moods. In her classic book Oral Literature in Africa, Ruth Finnegan comments more than once on the histrionic capabilities of oral performers.131 In biblical narrative, sensitivity to the role of the voice has been demonstrated by Robert Alter, who points to a feature of particular importance: the appearance of two different characters using two contrasting speaking styles.132 For example, Isaac uses uncomplicated syntax in his question, “Here are the fire and the wood; but where is the sheep for the burnt offering?” (Gen 22:7, ‫ִהנֵּ ה ָה ֵאשׁ וְ ָה ֵע ִצים וְ אַיֵּ ה ַה ֶשּׂה‬ ‫)לע ָֹלה‬. ְ By contrast, Abrahamʼs sophisticated but evasive answer is formulated in an elaborate sentence pattern: “God will see to the sheep for His burnt offering, my son” (v. 8, ‫ֹלהים יִ ְר ֶאה־לֹּו ַה ֶשּׂה ְלע ָֹלה ְבּנִ י‬ ִ ‫)א‬. ֱ In an oral performance the narrator can use intonation, mimic, and poise to bring out the innocent wondering and the troubled but grave answer. Thus in the oral vista, the dialogue in biblical narrative is to be viewed as an opportunity for the performer and even as a residue of the performerʼs voice. In a written text it is an inheritance of the performerʼs art. But then, biblical narrative also confronts us with some severe difficulties that would not be felt at all in oral performance. The tale of Abrahamʼs complaint preceding the covenant between the pieces provides many opportunities for the performing artist. The divine assurance concerning his innumerable offspring is underscored by the countless stars: “He brought him outside and said (‫יֹּוצא‬ ֵ ַ‫ו‬ ‫אמר‬ ֶ ֹ ‫חוּצה וַ יּ‬ ָ ‫)אֹתֹו ַה‬, ‘Look toward heaven and count the stars, if you are able to ֶ ֹ ‫)וַ יּ‬, ‘So shall your posterity be.’ ” The count them.’ Then he said to him (‫אמר לֹו‬ divine suggestion to go out and count the stars is followed by the promise “so shall your offspring be.” What happened between the divine suggestion to count 129  A. J. C. Verheij, Verbs and Numbers: A Study of the Frequencies of the Hebrew Verbal Tense Forms in the Books of Samuel, Kings, and Chronicles (SSN 28; Assen: Van Gorcum, 1990), 32–36; Y. T. R adday and H. S hore , Genesis: An Authorship Study in ComputerAssisted Statistical Linguistics (AnOr 103; Rome: Biblical Institute Press, 1985), 24–25. 130  I. J. F. de J ong , “Convention versus Realism in the Homeric Epics,” Mnemosyne 58 (2005), 1–22, here 12. 131  Finnegan, Oral Literature in Africa (see n. 27), 382–388, 501–505; see also Walker, Turkish Folktale (see n. 27), xviii–xix. 132  R. Alter, The Art of Biblical Narrative (New York: Basic Books, 1981), 72–74: Jacob and Esau in Gen 25:30–31; Joseph and Potipharʼs wife in Gen 39; Saul and David in 1 Sam 24.

Oral Platform and Language Usage in the Abraham Narrative

439

the stars and the promise is not told, but the singer of tales can show by gesture and poise how Abraham went out, looked at the sky, and wondered. Thus the dialogue involves performance. Some performance is also implied when the utterance of a given speaker is immediately followed by a second utterance that is attributed to the same speaker. Such cases imply a pause and even a silent reaction on the part of the addressee.133 This is the case in Abrahamʼs questions that occasioned the divine response (vv. 2a, 3): “But Abram said (‫אַב ָרם‬ ְ ‫אמר‬ ֶ ֹ ‫)וַ יּ‬, ‘O Lord God, what will you give me? For I am to die childless.’ [. . .] And Abram said (‫אַב ָרם‬ ְ ‫אמר‬ ֶ ֹ ‫)וַ יּ‬, ‘You have given me no offspring, and so a slave born in my house is to be my heir.’ ” Abramʼs second argument, though often viewed as a doublet,134 is more than a mere reiteration.135 The saying “you have given me no offspring” returns to the theme of giving in the first saying, applies it to the notion of dying childless (‫)ע ִר ִירי‬, ֲ and thus augments the previous complaint. The repetition of the quotation formula indicates a pause between the two complaints and, even more than that, a lack of response. The divine answer is triggered by the second complaint. In the text, this constellation is only indicated by the technicality of the repetition of the quotation formula, which is to be opened up by interpretation.136 But the storyteller can signify what happened by means of tone, poise, gesture, and facial expression. By the same token, a number of passages present a change in speaking or acting subject without any indication on the part of the narrator.137 A case in point is the final verse of this episode: 133  M. S hiloah , “And He Said . . . and He Said,” in Sefer Korngrin (ed. A. Weiser and B. Luria; Tel Aviv: Niv and The Society for Bible Research, 1964), 251–276; C. L. Miller, “Silence as a Response in Biblical Hebrew Narrative: Strategies of Speakers and Narrators,” JNSL 32 (2006), 23–43; L. J. de Regt, Participants in Old Testament Texts and the Translator: Reference Devices and Their Rhetorical Impact (Assen: Van Gorcum, 1999), 13. Pauses between different stages in the argumentation are common in human intercourse, as shown by, e.g., I. H utchby and R. W ooffitt , Conversation Analysis: Principles, Practices and Applications (Cambridge: Polity Press, 1998), 81. For the doubling of the speaking formula in texts from Mari see M. A nbar , “Formule dʼintroduction du discours direct au mileu du discours à Mari et dans la Bible,” VT 47 (1997), 530–536. 134  C. Westermann, Genesis 12–36: A Commentary (trans. J. J. Scullion; CC; Minneapolis: Augsburg, 1985), 219–220; J. C. Gertz, “Abraham, Mose und der Exodus: Beobachtungen zur Redaktionsgeschichte von Gen 15,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 67–81, here 70. 135  Köckert, “Gen 15” (see n. 24), 38. Wenham notes that the interchange of ‫ הן‬and ‫הנה‬ does not fit the hypothesis that v. 3 reiterates v. 2 as a gloss: G. J. W enham , Genesis 1–15 (WBC 1; Waco, TX: Word, 1987), 327. 136  See also Gen 20:9–10; 29:14–15; 30:27–28; 37:21–22. 137  J. Wellhausen, Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871), 22–23; R. E. L ongacre , Joseph: A Story of Divine Providence; A Text Theoretical

440

Frank H. Polak

Genesis 15:6

‫וְ ֶה ֱא ִמן ַבּיהוָ ה וַ יַּ ְח ְשׁ ֶב ָה לֹּו ְצ ָד ָקה‬ And he believed in YHWH; and he reckoned it to him for righteousness.

The interpretation that the second clause indicates the divine response to Abrahamʼs trust in his overlord is commonplace.138 Nevertheless, the early medieval commentator R. Josef Bechor Shor construes ‫ וַ יַּ ְח ְשׁ ֶב ָה‬as Abrahamʼs act.139 Once again, it is the performing artist who can clarify the situation by means of tone and gesture. An additional issue, well-known since the work of Meir Weiss, Robert Alter, and Meir Sternberg, is the repetition-deviation pattern, according to which (a) one of the characters formulates a plan, gives advice, or issues an order worded in certain terms, whereas (b) its execution is worded in slightly different terms in the narratorʼs domain; or an event (a) is described in the narratorʼs domain and (b) is afterwards recounted by one of the characters in different wording. In literary studies such deviations are considered significant since they reflect the narratorʼs stance or the point of view of the speaker or addressee. But it takes all oneʼs semiotic, semantic, and syntactic insight to reveal the meaning of such repetition-deviation; and in many cases, lack of deviation is no less meaningful. Can we presume that the ancient reader is that conscious of the text and the complex games the narrator is playing? When we think of the performer, this issue looks totally different. The narrator can use intonation, gesture, and face in order to make the audience experience these nuances.140 In a living performance, such modulations come to life and turn into a game between narrator and audience. Let us consider Abrahamʼs proposal to his guests: “Let a little water be brought, and wash your feet, and rest yourselves under the tree. Let me bring and Textlinguistic Analysis of Genesis 37 and 39–48 (Winona Lake, IN: Eisenbrauns, 1989), 144–148, 155–169, 174–183; de R egt , Participants (see n. 133), 28–84; H. R eckendorf , Die syntaktischen Verhältnisse des Arabischen (Leiden: Brill, 1898), 371–372. I discuss some pragmatic-rhetoric aspects in F. Polak, “Speaker, Addressee and Positioning: Dialogue Structure and Pragmatics in Biblical Narrative,” in Interested Readers: Essays on the Bible in Honor of David J. A. Clines (ed. J. K. Aitken et al.; Atlanta: Society of Biblical Literature, 2013), 359–372. 138  Following the Mek. dʼRabbi Ishma‍ʾel, Bešallaḥ 6; see Mechilta dʼRabbi Ismael (ed. H. S. Horowitz and I. A. Rabin; Frankfurt am Main: J. Kauffmann, 1931), 114–115, Rashi, Kimhi, and Nahmanides; and see Westermann, Genesis (see n. 134), 222–223. If this clause had been introduced at a later stage as a redactional explanation, the interpolator would have clarified who reckoned what to whom, as do the NRSV (“and the Lord reckoned it to him as righteousness”), the NJPS (“And because he put his trust in the Lord, He reckoned it to his merit”), the LXX (καὶ ἐλογίσθη αὐτῷ εἰς δικαιοσύνην, cf. Ps 106:31), the Vulgate, Tg. Neof., the Peshitta, and Luther. 139  In the Abraham tale, one encounters a similar problem in Gen 18:28–32; 24:32–33. 140  The differences between the art of the performer and that of the writing author are touched upon by Y. Zakovitch, “From Oral Tale to Written Narrative in the Bible,” Jerusalem Studies in Jewish Folklore 1 (1981), 9–43, here 33–34 (Hebrew, with summary in English).

Oral Platform and Language Usage in the Abraham Narrative

441

a little bread, that you may refresh yourselves, and after that you may pass on – since you have come to your servant” (Gen 18:4–5, NRSV). This modest suggestion is followed by a rich regaling: “Then he took curds and milk and the calf that he had prepared, and set it before them; and he stood by them under the tree while they ate” (18:8). The contrast between modesty and rich entertainment could be underlined by appropriate gestures and a happy smile. It is not the scribal desk that forms the background for the literary milieu of biblical narrative. It is the three-dimensional art of the performing singer of tales.141 The written text is a reflection of the narratorʼs art of voice and gesture in the two dimensions of ink and papyrus.

 For recent attention to the performance, see, e.g., T. Giles and W. J. Doan, Twice Used Songs: Performance Criticism of the Songs of Ancient Israel (Peabody, MA: Hendrickson, 2009); R. D. M iller , “Orality and Performance in Ancient Israel,” RevScRel 86 (2012), 181–192; and with a focus on theater, S. Levy, The Bible as Theatre (Brighton: Sussex Academic Press, 2002); H. Utzschneider, “Is There a Universal Genre of ‘Drama?,’ ” in Literary Fiction and the Construction of Identity in Ancient Literatures (ed. H. Liss and M. Oeming; Winona Lake, IN: Eisenbrauns, 2010), 63–79; one notes the comments by F innegan , Oral Literature in Africa (see n. 27), 501–505. 141

Storytelling and Redaction Varieties of Language Usage in the Exodus Narrative Frank H. Polak In the present study, I want to advance two interconnected theses. My first claim is that the book of Exodus harbors two distinct linguistic registers that allow for easy identification and quantification. The first register, mainly found in the Priestly and Deuteronomic strata, is characterized by a number of features that represent the cross-linguistic, cross-cultural profile of written language. A second register reveals quite a different character and is in many respects close to spontaneous spoken discourse. Narratives in this style form the backbone of the narrative in Exodus. My second thesis is that this second layer preserves an underlying oral-epic substratum, an overarching platform, that is situated within the tradition stream of Northwest Semitic epic narrative poetry and in which the narrative in its present, written form is anchored.1 This platform comprises the narrative of the exodus and the episode of the conclusion of the covenant at Mount Sinai in one overarching framework, the Exodus-Sinai narrative (ESN). Anthropological and ethnopoetic fieldwork provides definite and conclusive counterevidence to Gunkelʼs thesis regarding the extreme brevity of the oral folktale. Fieldwork also indicates that oral narrators and poets, the singers of tales, are accomplished artists of oral literature, and philological analysis shows that Ugaritic epic texts reveal many features that are to be associated with oral poetry.2 Both claims are diametrically opposed to the views that dominate recent study of the narrative of the exodus and the conclusion of the covenant at Mount Sinai. According to these views, the ESN is to be attributed to the Deuteronomic, post-Deuteronomic, Priestly, and post-Priestly authorial and redactional strata.  On the oral-epic platform of patriarchal narrative, see F. Polak, “Oral Substratum, Stylistic-Syntactic Profile and Thematic Flow in the Abraham-Jacob Narrative,” in Contextualizing Israelʼs Sacred Writings: Ancient Literacy, Orality, and Literary Production (ed. B. Schmidt; Atlanta: Society of Biblical Literature, 2015), 217–238; and idem, “Oral Platform and Language Usage in the Abraham Narrative,” in the present volume, pp. 405–441. 2  K. T. A itken , “Oral Formulaic Composition and Themes in the Aqhat Narrative,” UF 21 (1989), 1–16; and see Polak, “Oral Platform” (see n. 1), pp. 406–409, 416–428, and the references there. 1

444

Frank H. Polak

An emerging consensus separates these narratives from the tales of the patriarchs, with which they were united by a post-Priestly redaction stratum. This consensus has the advantage of positing a rather synchronic view of the book of Exodus, according to which the author-redactor combines the task of editor and creative author, whose activity consists, in the words of Jean Louis Ska, “of collecting, rearranging, re-elaborating, and reshaping older material.”3 The redactional approach harbors an important truth but also fails to do justice to the variety in language usage and the sharp linguistic distinctions between the diverse strata of the exodus narrative. Therefore, the first part of the present study will establish these stylistic distinctions and the extent of the oral-epic strand of the narrative (sections 1–2). In the second part, I will discuss methods of establishing redactional intervention and ways of reading (section 3). The third part will be dedicated to structural aspects of the ESN (sections 4–5). It is a basic tenet of this study that language usage is to be viewed as part of social praxis and thus has important implications for the sociocultural background of the texts at hand.

1 From the Oral to the Written: Stylistic-Syntactic Patterns 1.1 Syntactic-Stylistic Analysis The principal argument for the existence of an oral background to the exodus narrative in its present form is based on a syntactic-stylistic analysis of language usage, by means of the parameters introduced in my study “Oral Platform and Language Usage in the Abraham Narrative.”4 Analysis by this method points to a distinction between two basic styles. On the one hand, one discerns an intricate elaborate style (IES) characterized by the frequency of long, elaborate clauses, clauses in syntactic subordination, and often quite long noun groups, such as:

3  J. L. S ka , “A Plea on Behalf of the Biblical Redactors,” Studia Theologica 59 (2005), 4–18, here 4; E. O tto , “Das postdeuteronomistische Deuteronomium als integrierender Schluß­stein der Tora,” in Die deuteronomistischen Geschichtswerke: Redaktions- und reli­ gions­geschichtliche Perspektiven zur “Deuteronomismus”-Diskussion in Tora und Vor­deren Pro­pheten (ed. M. Witte et al.; BZAW 365; Berlin: de Gruyter, 2006), 71–102, here 72–78, 100–102. 4  See above, pp. 412–414. This method (very much a project in progress) is developed in detail in my previous studies: see F. P olak , “Sociolinguistics, A Key to the Typology and the Social Background of Biblical Hebrew,” HS 47 (2006), 115–162, here 128–136, 141–51; idem , “The Book of Samuel and the Deuteronomist: A Syntactic-Stylistic Analysis,” in Die Samuelbücher und die Deuteronomisten (ed. C. Schäfer-Lichtenberger; BWANT 188; Stuttgart: Kohlhammer, 2010), 34–73, here 38–54.

Storytelling and Redaction

445

Exodus 35:4–5

‫( ֵלאמֹר‬3) ‫ל־ע ַדת ְבנֵ י־יִ ְש ָר ֵאל‬ ֲ ‫ל־כ‬ ָ ‫( ֶא‬2) ‫( מ ֶֹשה‬1) ‫אמר‬ ֶ ֹ ‫וַ י‬ ‫ר־צוָּ ה יְ הוָ ה ֵלאמֹר‬ ִ ‫( זֶ ה ַה ָדּ ָבר ֲא ֶשׁ‬1) ‫( ַליהוָ ה‬3) ‫רּומה‬ ָ ‫( ְת‬2) ‫( ֵמ ִא ְת ֶכם‬1) ‫ְקחּו‬ ‫( זָ ָהב וָ ֶכ ֶסף ּונְ ח ֶֹשת‬3) ‫רּומת יְ הוָ ה‬ ַ ‫( ֵאת ְת‬2) ‫יא ָה‬ ֶ ‫( כֹל נְ ִדיב ִלּבֹו ִיְב‬1) Thus Moses said to the whole community of Israelites, saying: This is the thing which YHWH has commanded, saying: Take from among you gifts to YHWH; everyone whose heart so moves him shall bring them – gifts for YHWH: gold, silver,  and copper.5

In this excerpt I have given numbers to the various constituents that are dependent on the predicate, such as, in the last clause, (1) subject (‫)כֹּל נְ ִדיב ִלבֹּו‬, (2) direct object (‫רוּמת יְ הוָ ה‬ ַ ‫)את ְתּ‬, ֵ and (3) second object (‫)זָ ָהב וָ ֶכ ֶסף וּנְ ח ֶֹשׁת‬. One notes a number of noun groups (‫רוּמת יְ הוָ ה ;זָ ָהב וָ ֶכ ֶסף וּנְ ח ֶֹשׁת‬ ַ ‫)תּ‬ ְ and a subordinate clause (‫)לאמֹר‬. ֵ A second hypotactic clause (‫ר־צוָּ ה יְ הוָ ה ֵלאמֹר‬ ִ ‫)א ֶשׁ‬ ֲ includes a clause subordinate to it, ‫לאמֹר‬,ֵ in complex hypotaxis. The IES, which in its complexity reveals scribal expertise and education, and thereby also a well-established scribal mileu,6 is characteristic of the narrative and homiletic texts in Deuteronomy, Joshua, and 1–2 Kings (1 Kgs 3–15; 2 Kgs 11–12; 14–25), the Jeremiah vita (Jer 26; 36–43), and large sections attributed to the Priestly corpus. Since many texts in this group position themselves in the late Judean monarchy, I speak of the Judean corpus. A similar style is revealed by the narratives placed, by their very content, in the Persian (or early Hellenistic) era, the Achaemenid corpus, characterized as it is by features of Late Biblical Hebrew, including Aramaic borrowings and calques.7 On the other hand, the voiced, lean, brisk style (VoLB) is characterized by a high frequency of extremely short clauses, consisting of predicate only or predicate with at most one additional constituent (around 50 percent of all clauses), as against the low frequency of subordinate clauses (around 15 percent

 The listing is continued in 35:6–9.  For cross-cultural and cross-linguistic characterizations of written language, see J. M. Miller and R. W einert, Spontaneous Spoken Language: Syntax and Discourse (Oxford: Clarendon Press, 1998), 4–21, 80–94, 105–125, 133–149, 209–220; M. A. K. Halliday, Spoken and Written Language (2nd ed.; Oxford: Oxford University Press, 1989), 73–75, 87, 98–100; F. Polak, “Oral Platform” (see n. 1), n. 33; and idem, “Language Variation, Discourse Typology, and the Socio-Cultural Background of Biblical Narrative,” in Diachrony in Biblical Hebrew (ed. C. Miller-Naudé and Z. Zevit; Winona Lake, IN: Eisenbrauns, 2012), 301–338, here 316, nn. 21–23, and the references there. 7  On Late Biblical Hebrew, see now A. H urvitz , A Concise Lexicon of Late Biblical Hebrew (VTSup 160; Leiden: Brill, 2014), 1–11; A. H ornkohl , “Characteristically Late Spellings in the Hebrew Bible: With Special Reference to the Plene Spelling of the O-Vowel in the Qal Infinitive Construct,” JAOS 134 (2014), 643–671; and further, the references in Polak, “Oral Platform” (see n. 1), pp. 431–434. 5 6

446

Frank H. Polak

of all clauses) and noun groups (around 45 percent of all clauses), such as, for example: Exodus 4:1–2

/ (0) ‫אמר‬ ֶ ֹ ‫ וַ יּ‬/ ‫( מ ֶֹשׁה‬1) ‫וַ יַּ ַען‬ (0) ‫אמרוּ‬ ְ ֹ ‫ ִכּי י‬/ ‫( ְבּק ִֹלי‬1) ‫ וְ לֹא יִ ְשׁ ְמעוּ‬/ ‫( ִלי‬1) ‫וְ ֵהן לֹא־יַ ֲא ִמינוּ‬ ‫( יְ הוָ ה‬2) ‫( ֵא ֶליָך‬1) ‫לֹא־נִ ְראָה‬ (0) ‫ ַמ ֶטּה‬/ (0) ‫אמר‬ ֶ ֹ ‫ וַ יּ‬/ ‫( ְביָ ֶדָך‬1) ‫ ַמה זֶּ ה‬/ ‫( יְ הוָ ה‬2) ‫( ֵא ָליו‬1) ‫אמר‬ ֶ ֹ ‫וַ יּ‬ But Moses spoke up and said, “What if they do not believe me and do not listen to me, but say: YHWH did not appear to you?” So YHWH said to him, “What is that in your hand?” And he said, “A rod.”

In this excerpt we note a series of clauses with predicate only (marked as 0), or with one additional constituent (marked as 1, such as ‫;בּק ִֹלי‬ ְ ‫)לי‬, ִ and the absence of long noun groups and subordinate clauses. A general characteristic of the VoLB is the sparsity of the information that is spread over a number of small chunks (chunking).8 Some features contribute to such sparsity: the frequent use of two-clause constructions, with preclause (verb of motion) and core clause; the use of paratactic circumstantial clauses; and the preference for implicit indication of subject and object, when already known from previous sections, by means of pre- and affixes of the verbal form, by object suffix, or by ellipsis. Indications concerning place and time are often presented in separate clauses, in particular in opening clauses introduced by ‫ויהי‬.9 Such sparsity is in keeping with the characteristics of spoken discourse, which is delivered on the spot, in face-to-face communication, with full knowledge of the surrounding circumstances. As such, it is aided by intonation, facial expression, gesture, and poise. Some of the basic features of the VoLB fit the conditions of spoken discourse: the short paratactic clauses do not overload short-term memory; elaborate specification by noun phrases is cumbersome and superfluous, since the circumstances are clear to the speaking parties; and for clarification of their intentions the speakers may use gesture, intonation, and facial expression. Thus the VoLB reveals a close proximity to spoken discourse and thereby to oral narrative. This style reflects the work of writing authors who are well acquainted with the performance of the oral narrator, the singer of tales, accept the art of oral narrative as prestigious, and adhere to its ways and norms. The close proximity to the art of the oral performer stands out in a number of stylistic 8  Miller and Weinert, Spontaneous Spoken Language (see n. 6), 14–15, 22–23, 58–71; H alliday , Spoken and Written Language (see n. 6), 30–45, 61–67, 79–84, 92–101; and see P olak , “Sociolinguistics and Social Background” (see n. 4), 149, nn. 95–97, and the references there; idem, “Orality: Biblical Hebrew,” in Encyclopedia of Hebrew Language and Linguistics (ed. G. Khan; 3 vols.; Leiden: Brill, 2013), 2:930–937. 9  J. Joosten, “Diachronic Aspects of Narrative wayhi in Biblical Hebrew,” JNSL 35 (2009), 43–61, here 48–55.

Storytelling and Redaction

447

features that are rare in the Judean corpus and Persian-era narrative, such as the tendency not to indicate the speaking characters when their identity is clear in the context, even in case of interchanging of the speaking person. The oral performer can suggest the change of identity by change of tone. By the same token, the narrative may prefer not to indicate a silent response to a preceding utterance and thus continue the utterance of a given speaker by a second utterance of the same participant (Gen 15:5). Gesture and facial expression form sufficient indications of the silent reaction. The particular way in which biblical narrative uses the circumstantial clause shows how seemingly “simple” features of spoken discourse, such as paratactic clause linking, turn into an art form, the rhythmic and meaningful interchange of sequentiality and nonsequentiality.10 The VoLB dominates patriarchal narrative, the Samuel-Saul-David history (1 Sam 1–4; 1 Sam 8–2 Sam 5; 9–20; 1 Kgs 1–2); the tales of Elijah, Elisha, and other prophets from the Northern Kingdom (roughly 1 Kgs 17–2 Kgs 10); parts of the tales of the heroic saviors and congeners (Judg 3–9; 13–19); and the scroll of Ruth. This group of texts constitutes the oral-written corpus.11 1.2 Two Styles and Their Sociocultural Backgrounds The contrast between IES and VoLB is more than radical. One may compare the opening of the so-called Book of the Covenant with the closure of the Deuteronomic legislation: Exodus 21:1 These are the laws that you shall set before them.

‫יהם‬ ֶ ֵ‫וְ ֵא ֶלּה ַה ִמּ ְשׁ ָפּ ִטים ֲא ֶשׁר ָתּ ִשׂים ִל ְפנ‬

This opening includes one single noun, with a relative clause that comprises predicate and addressee (‫יהם‬ ֶ ֵ‫)ל ְפנ‬ ִ only. By contrast, the colophon of the Deuteronomic series of blessings and curses includes long noun phrases and complex subordination: Deuteronomy 28:69 ‫מֹואב ִמ ְל ַבד ַה ְב ִרית ֲא ֶשר־‬ ָ ‫ת־בנֵ י יִ ְש ָר ֵאל ְב ֶא ֶרץ‬ ְ ‫ר־צוָ ה יְ הוָ ה ֶאת־מ ֶֹשה ִל ְכר ֹת ֶא‬ ִ ‫ֵא ֶלה ִד ְב ֵרי ַה ְב ִרית ֲא ֶש‬ ‫ָכ ַרת ִא ָתם ְבח ֵֹרב‬ These are the terms of the covenant which YHWH commanded Moses to conclude with the Israelites in the land of Moab, in addition to the covenant which he had made with them at Horeb.

 See Polak, “Oral Platform” (see n. 1), pp. 438–441; idem, “The Circumstantial Clause as Trigger: From Syntax to Discourse and Plot Structure in Biblical Narrative,” in Strategies in Clause Linking in Semitic Languages (ed. B. Isaksson and M. Persson; Wiesbaden: Harrassowitz, 2014), 191–203. 11  In previous studies I used the terms classical corpus or medial corpus. See P olak “Language Variation” (see n. 6), 303. 10

448

Frank H. Polak

For the purpose of the present study I distinguish between two types of the VoLB style: (1) cases in which at least 50 percent of the text consists of extremely short clauses (predicate only, or predicate with one additional constituent, such as subject, object, or adverb; VoLB-1), and (2) cases in which such clauses cover between 40 and 50 percent of all clauses (VoLB-2). Let me clarify the distinction between these two types and the IES with the aid of the following table:12 Comparison13 Unit Class

Exodus VoLB-1 Median

Mean ±Standard Deviation

Exodus VoLB-2 Median

Mean ±Standard Deviation

Deut 1–4; 8–10; 34 Narrative Median

Mean ±Standard Deviation

Short Clauses

57.14

57.30 ±5.58

42.86

43.31 ±2.55

34.09

31.69 ±9.37

All Hypotaxis

14.00

15.58 ±7.47

20.00

19.59 ±4.61

32.81

36.10 ±12.74

Noun Phrases (MNP)

48.68

48.38 ±20.70

43.45

44.11 ±7.20

63.73

68.70 ±21.17

Intricate

11.43

11.27 ±5.35

18.00

18.41 ±6.44

31.67

34.33 ±11.26

Thus a fundamental and radical difference sets the IES apart from the VoLB. This differentiation has implications beyond stylistics and language usage. Since the intricate style reveals the habitus and expertise of the scribal chancery, the narratives, legal provisions, and theological homilies couched in this style can only originate in a context that is dominated by the scribal chancery. This style, then, presupposes a well-developed and self-conscious royal bureaucracy that  A note on analytic terminology: MNP: mean noun pair (the number of grouped nouns represented as noun pairs). The column for “intricate” clauses includes both long clauses (predicate with three complements) and clauses in complex hypotaxis (syntactic subordination to an independent clause, a long subordinate clause, or a subordinate clause including a noun group; see Polak, “Language Variation” [see n. 6], 305–307). The median is the middle value of the data set. In the series 10, 17, 22, 27, and 29, the median is 22. 13  The significance of these distinctions was tested by means of the chi-square test. For all categories the distinction between the VoLB-2 units in Exodus and the narrative sections in Deuteronomy was found to be significant or even extremely significant. An analysis of the distinction between the midrange data of the various categories revealed highly significant figures for the MNP and the intricate clauses. For “all hypotaxis” the distinctions are significant. For “short clauses” the distinction turned out nonsignificant, but a similar test of the distinction between Deuteronomic narrative and the VoLB-1 texts resulted in a highly significant distinction. 12

Storytelling and Redaction

449

provides the education to impart the skills required for handling texts in this style. The Hebrew epigraphic texts, most of them from the Judean monarchy, all reveal the characteristics of the IES.14 It is true that the IES is likewise in use in the Persian era, but the texts originating in this period reveal many features that indicate the dominating role of the Aramaic writing chancery of the Achaemenid administration, in which the scribal education focused on Aramaic. By contrast, as discussed above, the VoLB, though used in writing, is close to the oral arena. This style represents the author-sōpēr who is attuned to the art of the oral performance and adheres to its style and language usage. The sociocultural implications of this style point to a context in which the oral narrator enjoys high prestige. If the IES represents the power, status, and linguistic preferences of the scribal chancery, the VoLB indicates a context in which the writing author can prefer the capabilities of the oral arena over the norms and powers prevailing in and imparted by the chancery. In other words, the sociocultural situation is one in which the royal administration is less dominant, less developed, and less powerful than in the later stages of the Judean monarchy. In terms of periodization, these considerations place the Judean corpus in the seventh and early sixth century, whereas the oral-written corpus precedes this period (although overlapping is not to be excluded). Notably, tales in the VoLB include many signs of a magic-mythic worldview, which is far less developed in IES tales, even though some of them do contain magical details. The attenuation and rationalization of these themes in the Deuteronomic versions of the Sinai narratives was noticed and commented upon by de Wette himself.15 1.3 Late Imitation of the Early Style? An argument that has been raised against the distinction between the various strata in Biblical Hebrew is the contention that in modern times many bilingual authors are able to write in a language that is not theirs by birth.16 As I have pointed out previously, this argument is not borne out by the examples adduced. Stylistic research indicates that the English used by Joseph Conrad deviates from the English literary style(s) in vogue in his generation and actually contains many expressions that are based on his native Polish or on French, the language 14  See C. A. R ollston , Writing and Literacy in the World of Ancient Israel: Epigraphic Evidence from the Iron Age (Atlanta: Society of Biblical Literature, 2010), 85–109, 128–135; F. Polak, “Sociolinguistics and Social Background” (see n. 4), 137–138; and idem, “The Oral and the Written: Syntax, Stylistics and the Development of Biblical Prose Narrative,” JANES 26 (1998), 59–105, here 103. 15  W. M. L. de Wette, Beiträge zur Einleitung in das Alte Testament (2 vols.; Halle: Schimmelpfennig, 1806–1807), 1:275–280. 16  H. M. B arstad , “Can Prophetic Texts be Dated? Amos 1–2 as an Example,” in Ahab Agonistes (ed. L. L. Grabbe; LHBOTS 421; London: T&T Clark, 2007), 21–39, here 23.

450

Frank H. Polak

in which he was taught navigation.17 By the same token, Samuel Beckettʼs lean style in French and English is quite different from his earlier baroque English style.18 The Latin of John Milton is described as superb, but some details in lexicon and metrical scansion reveal important deviations from authentic classical Latin;19 his English style is found to lean heavily on Latin syntax.20 These examples show that the style used by bilingual authors is subject to interference by the alternative language. Such is the psychology of language usage.21 Moreover, the sociology of language indicates a radical difference between second-language acquisition in the modern era and such processes in preclassical antiquity. In the preclassical period we cannot postulate the existence of systematic grammars of classical Hebrew, of comprehensive lexica, and of compendia listing differences between the classical language and Late Biblical or Qumran Hebrew. Such instruments are not indicated by Ben Sira, nor have they been found among the texts from the Judean Desert. By contrast, systematic grammars, lexica, and elaborate schooling are well-known in the early modern era, both in Latin and

17  M. Lucas, “Conradʼs Adjectival Eccentricity,” Style 25 (1991), 123–151; idem, Aspects of Conradʼs Literary Language (Boulder, CO: Social Science Monographs; Lublin: Maria Curie-Sklodowska University, 2000); M. Morzinski, Linguistic Influence of Polish on Joseph Conradʼs Style (New York: Columbia University Press; Lublin: Marie Curie-Sklodowska University, 1994). Conradʼs stylistic use of such phrases is discussed by E. Kujawska-Lis, “(Pseudo)Polonisms in Joseph Conradʼs Amy Foster and Prince Roman and Their Polish Translations,” Acta Neophilologica 14 (2012), 5–17. 18  R. N. Coe, “Beckettʼs English,” in Samuel Beckett: Humanistic Perspectives (ed. M. Beja et al.; Columbus: Ohio State University Press, 1983), 36–57, here 41–45; on Beckettʼs early language education and his bilingualism, see M. Kager, “Comment Dire: A Neurolinguistic Approach to Beckettʼs Bilingual Writings,” L2 Journal 7 (2015), 63–83, https://escholarship. org/uc/item/9wx9s230 (accessed 4/21/15); A. Beer, “Beckettʼs Bilingualism,” in The Cam­ bridge Companion to Beckett (ed. J. Pilling; Cambridge: Cambridge University Press, 1994), 209–221. 19  K. H ale , Miltonʼs Languages: The Impact of Multilingualism on Style (Cambridge: Cambridge University Press, 1997), 30–31; on Miltonʼs use of impersonal licet as a personal verb (and his neologisms), see idem, “Notes on Miltonʼs Latin Word-Formation in the ‘Poemata’ of 1645,” Humanistica Lovaniensia 43 (1994), 405–410, here 406–407. 20  Sensitive distinctions are proposed by Hale, Miltonʼs Languages (see n. 19), 104–120, 157–179; E. Haan, “Both English and Latin: Miltonʼs Bilingual Muse,” Renaissance Studies 21 (2007), 679–700. 21  B. Köpke, “Neurolinguistic Aspects of Attrition,” Journal of Neurolinguistics 17 (2004), 3–30; M. Wei et al., “How Age of Acquisition influences Brain Architecture in Bilinguals,” Journal of Neurolinguistics 36 (2015), 35–55; M. S. Schmidt, Language Attrition (Cambridge: Cambridge University Press, 2011); K. B ardovi -H arlig and D. S tringer , “Variables in Second Language Attrition: Advancing the State of the Art,” Studies in Second Language Acquisition 32 (2010), 1–45; and the references in P olak , “Sociolinguistics and Social Background” (see n. 4), 118 (nn. 9, 11, 13), 122–125 (nn. 34, 44, 47, 49); idem, “Language Variation” (see n. 6), 326.

Storytelling and Redaction

451

in modern languages.22 Milton could learn Latin systematically at the grammar school; Becket held a university degree in French and Italian; Conrad was given the education of Polish nobility. A developed infrastructure of this kind did not exist in the Judean monarchy or the period of Babylonian-Persian domination, and thus the sociocultural circumstances of language usage of this area and period are entirely different from the linguistic ecology of modern literature. A similar difference sets the ancient Israelite sōpēr apart from the Roman and Greek, mostly aristocratic, literati, who had their schooling in the highly developed educational system of the Hellenistic-Roman period, 23 which enabled them to adopt a Greek cultural identity.24 Thus an author like Lucian, who was active around the middle of the second century CE,25 writing in Greek as his second language, could endeavor to use a language that was as close 22  R. Black, Humanism and Education in Medieval and Renaissance Italy: Tradition and Innovation in Latin Schools from the Twelfth to the Fifteenth Century (Cambridge: Cambridge University Press, 2001), 46–167, 331–365; I. Green, Humanism and Protestantism in Early Modern English Education (Surrey: Ashgate, 2009), 55–265. Of course, an extremely high level of education and a rich literary, grammatical, and lexicographic inheritance is also attested in medieval Islam; see, e.g., M. Chamberlain, Knowledge and Social Practice in Medieval Damascus, 1190–1350 (Cambridge: Cambridge University Press, 1994), 70–107; M. Marin, “The ʿulamāʾ,” in The Western Islamic World: Eleventh to Eighteenth Centuries, vol. 2 of The New Cambridge History of Islam (ed. M. Fierro; Cambridge: Cambridge University Press, 2010), 679–704. The sociocultural context is not taken into account in the comparison between ancient Hebrew and medieval Arabic in I. Young et al., Linguistic Dating of Biblical Texts: An Introduction to Approaches and Problems (2 vols.; London: Equinox, 2008), 1:48; and see J. Blau, “The Structure of Biblical and Dead Sea Scrolls Hebrew in Light of Arabic Diglossia and Middle Arabic,” Leshonenu 60 (1996), 21–32, here 27–28 (Hebrew, with summary in English). 23  See H. I. Marrou, A History of Education in Antiquity (trans. G. Lamb; New York: New American Library, 1964), 223–308, 381–418; R. C ribiori , Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt (Princeton: Princeton University Press, 2001), 220–252; G. Anderson, The Second Sophistic: A Cultural Phenomenon in the Roman Empire (London: Routledge, 1993), 10–40; W. Schmid, Der Atticismus in seinen Hauptvertretern von Dionysius von Halikarnassus bis auf den Zweiten Philostratus (5 vols.; Stuttgart: Kohlhammer, 1887–1897), 1:21–32; G. L. Kim, “The Literary Heritage as Language: Atticism and the Second Sophistic,” in A Companion to the Ancient Greek Language (ed. E. J. Bakker; Chichester: Wiley-Blackwell, 2010), 468–482, here 469–496. 24  S. Goldhill, “Setting an Agenda: ‘Everything Is Greece to the Wise,’ ” in Being Greek under Rome: Cultural Identity, the Second Sophistic, and the Development of Empire (ed. S. Goldhill; Cambridge: Cambridge University Press, 2001), 1–25. The status of the prominent Greek and Eastern aristocracies in their dependence on Roman authority is studied by E. L. Bowie, “Greeks and Their Past in the Second Sophistic,” Past and Present 46 (1970), 3–41. 25  See H. H elm , “Lukianos,” PW 26:1725–1777, here 1726–1728. Lucian (around 120/125–180 CE), who calls himself a Syrian, tells us that he was born of lowly parents (in sharp contrast with the aristocratic descent of most famous rhetors) in Samosata (Commagene).

452

Frank H. Polak

as possible to classical Attic.26 Such efforts were buttressed by references to classical authors worth studying and emulating and by linguistic manuals.27 Lucianʼs impressive linguistic capabilities were created by his tertiary Greek education in Ephesus and Athens28 and his intensive reading of the great authors of classical Athens and of Athenian inscriptions. Thus a study of Lucianʼs morphology can conclude that his language usage is extremely close to classical Attic, and a general review by Wilhelm Schmid notes that his style is very close to the best Attic texts.29 Nevertheless, studies of Lucianʼs syntactic usage indicate a number of features that follow the Greek koinè of the period rather than Attic usage.30 A significant part of his lexical repertoire

 See Schmid, Der Atticismus (see n. 23), 1:216–223; S. Chabert, Lʼatticisme de Lucien (Paris: Société Française dʼimprimerie et de librairie, 1897), 204–205; A nderson , Second Sophistic (see n. 23), 86–99; Kim, “Literary Heritage” (see n. 23), 476–478; and see n. 27, below. 27  The imitation of classical Attic from Plato, Xenophon, Thucydides, Demosthenes, and poetic texts is discussed by Schmid, Der Atticismus (see n. 23), 1:223–225, 250–303, 308–311; Chabert, Lʼatticisme (see n. 26), 229–234. 28  Higher education in Ephesus (and its relationship to Roman authority) is discussed by S. D mitriev , City Government in Hellenistic and Roman Asia Minor (Oxford: Oxford University Press, 2005), 283–286; E. L. B owie , “The Geography of the Second Sophistic: Cultural Variations,” in Paideia: The World of the Second Sophistic (ed. B. E. Borg; Berlin: de Gruyter, 2004), 65–83, here 68. 29  R. J. Deferrari, Lucianʼs Atticism: The Morphology of the Verb (Princeton: Princeton University Press, 1916), 80–82; S chmid , Der Atticismus (see n. 23), 1:223–225, 428–432; But Schmid (432) also notes that no less than 12.5 percent (one eighth) of Lucianʼs lexical repertoire is postclassical (see 352–404). Deferrari (Lucianʼs Atticism, 9–13) notes instances of nonclassical usage of the augment; for other non-Attic features, see 20–23 (verbal endings), 36–39 (γινώσκω / classical γιγνώσκω and, less frequently, γίνομαι / classical γίγνομαι), 61–66 (verbs -μι class); C habert , Lʼatticisme (see n. 26), 102–116. D eferrari (Lucianʼs Atticism, 80) concludes that “[a]lthough Lucian is not absolutely accurate in his use of early and late futures, aorists, and perfects, he is nevertheless more strict than his fellow Atticists in this regard.” In his view, Lucianʼs deviations from authentic Attic serve rhetorical purposes (such as the wish to avoid pedantry), whereas many koine features were introduced in the manuscript tradition. Chabert (Lʼatticisme, 102, 108, 204–205) judges that “Lucien diffère sensiblement, de lʼattique dʼabord, de la κοινὴ διάλεκτος, ensuite” but that “cependant les formes attiques sont de beaucoup les plus nombreuses.” The stylistic/rhetorical coordinates of Lucianʼs language usage are studied by J. Bompaire, Lucien écrivain: Imitation et création (Paris: de Boccard, 1958), 132–154, 379–404, 443–469, 564–572, 622–640. 30  Such features as the nonclassical use of the cases are mentioned in the careful discussion of A. du M esnil , “The Grammatical Manner which Lucian Followed in Comparison with the Manner of the Ancient Attics,” in Programm des städtischen Gymnasium zu Stolp für das Schuljahr 1866–1867 (ed. H. Schütz; Stolp: Feige, 1867), 1–58 (Latin), here 8–10, 27–28 (infinitive constructions), 30–40 (prepositions). He also notes deviations in the use of the conjunctive and the optative (15–26) and the negations οὐ/μή (40–48); see also C habert , Lʼatticisme (see n. 26), 205. 26

Storytelling and Redaction

453

is characteristic of his own period.31 Thus Lucian, and the Atticistic authors of the second century BCE, well educated and learned though they were, did not escape the imperfections inherent in any writing in a language that is not in living usage. The literature of this period cannot serve to underpin the assumption that the syntactic-stylistic profile of the oral-derived corpus could be created by all scribes in any period. Moreover, the extant corpus in the VoLB is extremely small, unlike, for instance, classical Greek literature. Such authors as Lucian who endeavored to imitate classical Attic could benefit from the vast literary corpus of fifth- and fourth-century Athens. The extent of this corpus enabled them to detect the hidden rules of the literary Attic dialect. Thus we must ask ourselves how the extremely small corpus of biblical narrative could have served to detect the hidden rules of the classical language.32 Moreover, the argument from Atticistic rhetoric suffers from a grave internal contradiction. The assumption that pentateuchal narrative was largely created by imitation on the part of learned scribes from the Persian era raises the question which corpus served them as an example. And if one presumes that such a corpus did exist, we must ask how it is possible that this corpus was not preserved by the scribes who used it as a venerated example. Hence, the imitation hypothesis can explain the interpolation of certain passages, but it fails to explain the creation of the oral-written corpus as such. Sometimes one expresses the feeling that Judeans in exile might have preserved the ancient language perfectly, like the more conservative strata among the Pennsylvania Dutch. However, sociolinguistic research indicates strong attrition by the dominant English language33 and the loss of many features of the German dialect.34 Moreover, Persian business documents from the Nippur 31  See S chmid , Der Atticismus (see n. 23), 1:352–404, 432; C habert , Lʼatticisme (see n. 26), 119–123. 32  One must take into account that according to the imitation hypothesis this small corpus would not have included the emulations themselves and thus would have been even smaller. Or are we to adopt the heavy presumption that the emulations effaced the venerated texts they imitated? 33  Lexical attrition is already documented in the nineteenth century, by S. S. Haldeman, Pennsylvania Dutch: A Dialect of South German with an Infusion of English (London: Trübner, 1872), 28–33. 34  M. L. Huffines, “Pennsylvania German: Convergence and Change as Strategies of Discourse,” in First Language Attrition (ed. H. W. Seliger and R. M. Vago; Cambridge University Press, 1991), 125–137. Huffines notes the loss of the dative form in nouns (128) and the weakening of strict word-order rules (133–134). On the sociocultural status of Pennsylvania Dutch, see idem , “Pennsylvania German: Maintenance and Shift,” International Journal of the Sociology of Language 25 (1980), 43–57; L. V. N ess , “The Pressure of English on the Pennsylvania German Spoken in Two West Virginia Communities,” American Speech 67 (1992), 71–82; T. Adkins, “ ‘The English Effect’ on Amish Language and Literacy Practices,” Community Literacy Journal 5 (2010–2011), 25–45.

454

Frank H. Polak

region, in which exiles from Judaea appear as principals and witnesses, indicate intensive contacts with the Babylonian environment, including the use of Babylonian personal names within families that also used Hebrew names.35

2 Stylistic Distinctions in the Book of Exodus 2.1 The Complex, Intricate Style in Exodus Both the VoLB and the IES play an important role in the book of Exodus. The IES characterizes most of the passages attributed to the Priestly source.36 This stratum includes the second tale of Mosesʼs call (Exod 6:2–13; 6:26–7:6);37 parts of the tale of the splitting of the sea (14:1–4, 8–10*, 15–18, 22–23, 27*–29);38 parts of the tale of the manna (16:1–3, 6–12, 15b–21a, 22–26, 32–33);39 the section on the radiant face of Moses (34:29–35);40

 L. E. Pearce and C. Wunsch, Documents of Judean Exiles and West Semites in Babylonia in the Collection of David Sofer (CUSAS 28; Bethesda, MD: CDL Press, 2014); Y. Bloch, “Judeans in Sippar and Susa during the First Century of the Babylonian Exile: Assimilation and Perseverance under Neo-Babylonian and Achaemenid Rule,” JANEH 1 (2014), 119–172, here 124–142; L. E. P earce , “Continuity and Normality in Sources Relating to the Judean Exile,” JANEH 1 (2014), 163–184, here 165–177; K. Abraham, “West Semitic and Judean Brides in Cuneiform Sources from the Sixth Century B.C.E.: New Evidence from a Marriage Contract from Al-Yahudu,” AfO 51 (2005–2006), 198–219. 36  The terminology used in the following footnotes is explained in n. 12, above. The term ELC (explicit lexicalized constituent) refers to the additional constituents, such as subject, object, adverb; if the clause consists of predicate with implicit subject only, I speak of 0 ELC. By 0–1 ELC, I mean extremely short clauses. 37  Exod 6:2–13; 6:26–7:6 (seventy-eight clauses): 0–1 ELC 29.49 percent; all hypotaxis 25.64 percent; MNP, 62.18 percent; complex hypotaxis 18.46 percent; 3+ ELC 10.26 percent. The attribution of 6:6–8 to a post-Priestly expansion has been defended by J. C. Gertz, Tradi­ tion und Redaktion in der Exoduserzählung: Untersuchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000), 244–248. 38  Exod 14:1–4, 8–10*, 15–18, 22–23, 27*–29 (fifty clauses): 0–1 ELC 34.00 percent; all hypotaxis 12.00 percent; MNP, 70.00 percent; complex hypotaxis 4.00 percent; 3+ ELC 8.00 percent. For the attribution of 14:10abα to J and bβ to P, see H. H olzinger , Exodus (KHAT 2; Tübingen: Mohr Siebeck, 1900), 54. 39  Exod 16:1–3, 6–12, 15b–21a, 22–26, 32–33 (109 clauses): 0–1 ELC 33.03 percent; all hypotaxis 33.03 percent; MNP, 41.28 percent; complex hypotaxis 17.43 percent; 3+ ELC 12.84 percent. Verse 34 has not been taken into account because of its textual difficulties; v. 35 is considered a gloss. The scholarly differences of opinion concerning various half-verses are presented by T. B. Dozeman, Exodus (Grand Rapids, MI: Eerdmans, 2009), 376. A number of difficulties are discussed by H olzinger , Exodus (see n. 38), 54. On the JE/redactional sections, see below, n. 75. 40  Exod 34:29–35 (thirty-one clauses): 0–1 ELC 12.90 percent; all hypotaxis 41.94 percent; MNP, 50.00 percent; complex hypotaxis 25.81 percent; 3+ ELC 9.68 percent. 35

Storytelling and Redaction

455

the opening sections of the building of the tabernacle (35:1–36:7);41 and, for comparison, the opening of the instructions concerning the Passover ritual (12:1–14).42 In the narrative of the plagues, we note the sign of the serpents (7:8–13)43 and the plague of boils (9:8–12).44

In addition, we note a number of passages that are not attributed to the Priestly or Deuteronomic corpus but are couched in the intricate style. This stratum includes the opening of the march out of Egypt (Exod 12:42; 13:3–10, 17–19, 21–22);45 the tale of the meeting with Jethro (18:1–12,46 13–27);47 the Fortschreibung within the Sinai tale (19:9b–10a,48 20–25);49 the last section of the legal and cultic instructions in the book of the covenant (23:1–19);50 and the second revelation to Moses on Mount Sinai (34:1–4, 27–28).51 This style likewise characterizes the following sections of the plagues narrative: the 41  Exod 35:1–36:7 (ninety-three clauses): 0–1 ELC 15.05 percent; all hypotaxis 49.46 percent; MNP, 113.44 percent (for MNP, a figure larger than 100 percent indicates that in the mean every clause contains a noun pair, and some clauses contain more than one noun pair); complex hypotaxis 38.71 percent; 3+ ELC 13.98 percent. 42  Exod 12:1–14 (forty-seven clauses): 0–1 ELC 29.79 percent; all hypotaxis 21.28 percent; MNP, 81.91 percent; complex hypotaxis 6.39 percent; 3+ ELC 4.25 percent. 43  Exod 7:8–13 (twenty-two clauses, too small for analysis): 0–1 ELC 36.36 percent; all hypotaxis 22.73 percent; MNP, 86.36 percent; complex hypotaxis 9.09 percent; 3+ ELC 13.64 percent. 44  Exod 9:8–12 (seventeen clauses, too small for analysis): 0–1 ELC 29.41 percent; all hypotaxis 23.53 percent; MNP, 88.24 percent; complex hypotaxis 5.88 percent; 3+ ELC 23.53 percent. On the connection with 7:8–13, see C. Berner , Die Exoduserzählung: Das Literarische Werden einer Ursprungslegende Israels (FAT 73; Tübingen: Mohr Siebeck, 2010), 191–193. 45  Exod 12:42; 13:3–10, 17–19, 21–22 (forty-eight clauses): 0–1 ELC 12.50 percent; all hypotaxis 27.08 percent; MNP, 81.25 percent; complex hypotaxis 16.67 percent; 3+ ELC 31.25 percent. 46  Exod 18:1–12 (thirty-nine clauses): 0–1 ELC 33.33 percent; all hypotaxis 38.46 percent; MNP, 79.49 percent; complex hypotaxis 28.21 percent; 3+ ELC 12.82 percent. 47  Exod 18:13–27 (fifty-eight clauses): 0–1 ELC 29.31 percent; all hypotaxis 20.69 percent; MNP, 94.83 percent; complex hypotaxis 5.17 percent; 3+ ELC 13.79 percent. 48  The repetitive resumption of vv. 9a/10a was already noted by the French commentator R. Shmuel b. Meir (Rashbam, 1080/5–1174). In v. 9 one notes the elaborate clause structure with 4 ELCs and the intricacies of the complex subordination (‫)בּ ֲעבוּר יִ ְשׁ ַמע ָה ָעם ְבּ ַד ְבּ ִרי ִע ָמְּך‬. ַ Hence the style of this passage differs significantly from its context. Since these features are joined by the epanalepsis of vv. 9a/10a, the issue seems settled. 49  Exod 19:20–25 (thirty-one clauses): 0–1 ELC 25.81 percent; all hypotaxis 29.03 percent; MNP, 24.19 percent; complex hypotaxis 16.13 percent; 3+ ELC 9.68 percent. These findings fit the positions of W. B eyerlin , Herkunft und Geschichte der ältesten Sinaitraditionen (Tübingen: Mohr Siebeck, 1961), 12–13; E. Blum, Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 48–49. 50  Exod 23:1–19 (fifty-seven clauses): 0–1 ELC 33.33 percent; all hypotaxis 22.81 percent; MNP, 61.40 percent; complex hypotaxis 12.28 percent; 3+ ELC 7.02 percent. 51  Exod 34:1–4, 11–28 (seventy-two clauses): 0–1 ELC 29.17 percent; all hypotaxis 18.06 percent; MNP 67.36 percent; complex hypotaxis 15.28 percent; 3+ ELC 11.11 percent. On vv. 5–10 (VoLB-2), see nn. 81, 141 below.

456

Frank H. Polak

pestilence of the livestock (9:1–7);52 the first stages of the plague of hail (9:13–21,53 22–26);54 and the introduction to the plague of the firstborn (11:1–10).55 These sections, then, represent later stages and redactional strata of the ESN.

2.2 The Lean Brisk Style in Exodus The first type of the VoLB style, with more than 50 percent short paratactic clauses (VoLB-1) is found in many passages: VoLB-1: the tales of Mosesʼs birth and flight (Exod 2:1–10,56 11–2557); the second part of the call tale (4:1–17,58 18–2359); the plague of lice (8:12–15);60 the closure of the plague of hail (9:27–35);61 the inception (10:3–7)62 and the continuation (10:8–20)63 of the plague of locusts;

 Exod 9:1–7 (twenty-two clauses, too small for analysis): 0–1 ELC 36.36 percent; all hypotaxis 18.18 percent; MNP 84.09 percent; complex hypotaxis 13.64 percent; 3+ ELC 18.18 percent. For a post-J attribution of this episode, not matched by Ps 105, see M. Noth, Exodus (OTL; Philadelphia: Westminster, 1962), 79; for post-Priestly attribution, see Berner, Die Exoduserzählung (see n. 44), 202–203. 53  Exod 9:13–21 (thirty-one clauses): 0–1 ELC 32.26 percent; all hypotaxis 35.48 percent; MNP 61.29 percent; complex hypotaxis 25.81 percent; 3+ ELC 12.90 percent. H olzinger (Exodus [see n. 38], 27) discusses the post-JE character of vv. 14–17, 19–21. 54  Exod 9:22–26: (seventeen clauses, too small for analysis): 0–1 ELC 23.53 percent; all hypotaxis 17.65 percent; MNP 97.06 percent; complex hypotaxis 11.76 percent; 3+ ELC 17.65 percent. See Berner, Die Exoduserzählung (see n. 44), 221–224. 55  Exod 11:1–10 (thirty-one clauses): 0–1 ELC 25.00 percent; all hypotaxis 25.00 percent; MNP 121.43 percent; complex hypotaxis 14.29 percent; 3+ ELC 17.86 percent. Redactional stratification is suggested by Noth, Exodus (see n. 52), 92–93. 56  Exod 2:1–10 (fifty clauses): 0–1 ELC 52 percent; all hypotaxis 14 percent; MNP 34 percent; complex hypotaxis 4.00 percent; 3+ ELC 2.00 percent. 57  Exod 2:11–25 (sixty-eight clauses): 0–1 ELC 61.76 percent; all hypotaxis 11.76 percent; MNP 37.50 percent; complex hypotaxis; 3+ ELC negligible. 58  Exod 4:1–17 (seventy-nine clauses): 0–1 ELC 67.09 percent; all hypotaxis 11.39 percent; MNP 27.84 percent; complex hypotaxis and 3+ ELC negligible. The scene of Exod 3:1–16 reveals a type-2 style, whereas 3:17–22 combine VoLB and IES features. 59  Exod 4:18–23 (thirty-five clauses): 0–1 ELC 57.14 percent; all hypotaxis 17.14 percent; MNP 22.86 percent; complex hypotaxis 5.71 percent; 3+ ELC 5.71 percent. For the scene of the nightly attack (4:24–26; twelve clauses) we have 0–1 ELC 75.00 percent; all hypotaxis 8.33 percent; MNP 37.50 percent; complex hypotaxis and 3+ ELC negligible. 60  Exod 8:12–15 (nineteen clauses, too small for analysis): 0–1 ELC 52.63 percent; all hypotaxis 10.53 percent; MNP 52.63 percent; 3+ ELC 10.53 percent. 61  Exod 9:27–35 (thirty-eight clauses): 0–1 ELC 52.63 percent; all hypotaxis 36.84 percent; MNP 48.68 percent; complex hypotaxis 26.32 percent; 3+ ELC 6.38 percent. These data do not fit the analysis of this episode (and following) as redactional, as proposed by, e.g., Berner, Die Exoduserzählung (see n. 44), 225–241. 62  Exod 10:3–7 (twenty-nine clauses): 0–1 ELC 51.72 percent; all hypotaxis 31.03 percent; MNP 53.45 percent; complex hypotaxis 13.79 percent; 3+ ELC 3.45 percent; see n. 61. 63  Exod 10:8–20 (fifty-three clauses): 0–1 ELC 58.49 percent; all hypotaxis 9.43 percent; MNP 81.13 percent; complex hypotaxis 3.77 percent; 3+ ELC 9.43 percent; see n. 61. 52

Storytelling and Redaction

457

and the tale of the ascent to the divine abode at Mount Sinai (24:1–11).64 One could also add the basis of the tale of the Sea of Reeds (14:5–7, 10a, 13–14, 19–20, 24–25).65

These sections represent, in my view, a rather ancient stage of the ESN, with clear connections to oral performance. Most narratives, however, contain slightly fewer short clauses (between 40 and 50 percent), and thus seem closer to the IES. This is the second type of the VoLB: VoLB-2: the tale of Israelʼs enslavement (Exod 1:1–22);66 the tale of Mosesʼs call (3:1–16);67 the divine command announcing the blood plague (7:14–18);68 the plagues of frogs (7:25–8:11),69 swarms (8:16–28),70 and darkness (10:21–29);71 the final plague (12:21–23, 25, 29–34);72 the tale of the defeat of Egypt at the Sea of Reeds (14:5–7, 10abc, 11–14, 19–20, 21bc, 24–25, 27b, 30–31);73 the Marah tale (15:22–27);74 parts of the manna tale (16:4–5, 13–15a, 21b, 27–31,  Exod 24:1–11 (forty clauses): 0–1 ELC 62.50 percent; all hypotaxis 7.50 percent; MNP 77.50 percent; complex hypotaxis 2.50 percent; 3+ ELC 5.00 percent. 65  Exod 14:5–7, 10a, 13–14, 19–20, 24–25 (forty-two clauses): 0–1 ELC 52.38 percent; all hypotaxis 14.29 percent; MNP 51.19 percent; complex hypotaxis 7.14 percent; 3+ ELC 9.52 percent. In this analysis (see H olzinger , Exodus [see n. 38], 43–44; B erner , Die Exoduserzählung [see n. 44], 345–348) I have excluded the themes of the complaint (vv. 11–12), the sea (vv. 21bc, 27b), and the continuation in vv. 30–31. If one includes these pericopes, one obtains a tale in the type-2 style (see n. 73, below). 66  Exod 1:1–22 (sixty-seven clauses): 0–1 ELC 41.79 percent; all hypotaxis 29.85 percent; MNP 50.00 percent; complex hypotaxis 11.94 percent; 3+ ELC 5.97 percent. 67  Exod 3:1–16 (seventy-eight clauses): 0–1 ELC 44.87 percent; all hypotaxis 19.23 percent; MNP 57.69 percent; complex hypotaxis 8.97 percent; 3+ ELC 6.41 percent. The syntactic-stylistic profile is not significantly affected by the references to the patriarchs and the six peoples of Canaan in vv. 6, 8, and 15. The continuation in 3:17–22 combines VoLB and IES features, whereas the scene in 4:1–17 reveals a VoLB-1 style. 68  Exod 7:14–18 (twenty-four clauses, too small for analysis): 0–1 ELC 42.67 percent; all hypotaxis 20.83 percent; MNP 27.08 percent; complex hypotaxis 8.33 percent; 3+ ELC 4.17 percent. On the continuation in vv. 19–24, see n. 84, below. 69  Exod 7:25–8:11 (fifty-seven clauses): 0–1 ELC 42.11 percent; all hypotaxis 12.28 percent; MNP 48.25 percent; complex hypotaxis 7.02 percent; 3+ ELC 12.28 percent. 70  Exod 8:16–28 (fifty-five clauses): 0–1 ELC 40.00 percent; all hypotaxis 20.00 percent; MNP 45.45 percent; complex hypotaxis 10.91 percent; 3+ ELC 9.09 percent. 71  Exod 10:21–29 (thirty-nine clauses): 0–1 ELC 41.03 percent; all hypotaxis 17.94 percent; MNP 33.33 percent; complex hypotaxis 2.56 percent; 3+ ELC 7.69 percent. 72  Exod 12:21–23, 25, 29–36, 38–39 (fifty-seven clauses): 0–1 ELC 42.10 percent; all hypotaxis 19.30 percent; MNP 67.54 percent; complex hypotaxis 12.28 percent; 3+ ELC 10.53 percent. 73  Exod 14:5–7, 10abc, 11–14, 19–20, 21bc, 24–25, 27b, 30–31 (sixty-nine clauses): 0–1 ELC 39.13 percent; all hypotaxis 21.74 percent; MNP 42.75 percent; complex hypotaxis 11.59 percent; 3+ ELC 17.39 percent. However, if we take our cue from the low figure for the MNP and limit the analysis to the basis of the tale (14:5–7, 10a, 13–14, 19–20, 24–25), excluding the themes of the complaint (vv. 11–12) and the sea (vv. 21bc, 27b) and the continuation in vv. 30–31, we obtain a tale in the type-1 style, with forty-two clauses, as noted above (n. 65). 74  Exod 15:22–27 (twenty-nine clauses): 0–1 ELC 48.28 percent; all hypotaxis 24.14 percent; MNP 37.93 percent; complex hypotaxis 10.34 percent; 3+ ELC 13.79 percent. 64

458

Frank H. Polak

34a);75 the tale of the water from the rock and the war at Rephidim (17:1–7,76 8–1677); the Sinai tale (19:3–19; 20:18–21);78 the tales of the golden calf and Mosesʼs intercession (32:1–8, 15–30;79 33:1–2280); and the second revelation to Moses (34:5–10).81

This stratum is later than the first type of the VoLB tales (although some overlapping is possible) but is not associated with the developed administration implied by the IES. A number of passages, however, mark the transition to the chancery style: the second part of the tale of Mosesʼs call (3:17–22);82 the tale of the confrontation with Pharaoh (5:1–6:1);83 and the blood plague (7:19–24).84

75  Exod 16:4–5, 13–15a, 21b, 27–31; 34a (partly JE, partly redactional/Dtr; forty-five clauses): 0–1 ELC 44.44 percent; all hypotaxis 20.00 percent; MNP 45.46 percent; complex hypotaxis 13.33 percent; 3+ ELC 13.33 percent. The establishment of this narrative sequence, which is a mere illustration, follows Holzinger, Exodus (see n. 38), 53–54. On the sections attributable to P, see above, n. 39. A different analysis is proposed by B. Baentsch, Exodus, Leviticus und Numeri (HKAT 1/2; Göttingen: Vandenhoeck & Ruprecht, 1903), 145–156; Noth, Exodus (see n. 52), 130–132. 76  Exod 17:1–7 (thirty-eight clauses): 0–1 ELC 39.47 percent; all hypotaxis 18.42 percent; MNP 32.89 percent; complex hypotaxis 13.16 percent; 3+ ELC 15.79 percent. 77  Exod 17:8–16 (thirty-five clauses): 0–1 ELC 42.86 percent; all hypotaxis 14.29 percent; MNP 37.14 percent; complex hypotaxis and 3+ ELC negligible. 78  Exod 19:3–8, 10–19; 20:18–21 (eighty-four clauses): 0–1 ELC 45.23 percent; all hypotaxis 14.29 percent; MNP 37.50 percent; complex hypotaxis 5.95 percent; 3+ ELC 10.71 percent (Exod 19:20–25: IES). 79  Exod 32:1–8, 15–30 (125 clauses): 0–1 ELC 44.80 percent; all hypotaxis 20.80 percent; MNP 36.29 percent; complex hypotaxis 7.20 percent; 3+ ELC 6.40 percent. 80  Exod 33:1–22 (one hundred clauses): 0–1 ELC 46 percent; all hypotaxis 21.00 percent; MNP 45.00 percent; complex hypotaxis 9.00 percent; 3+ ELC 9.00 percent. 81  Exod 34:5–10 (twenty-seven clauses, too small for analysis): 0–1 ELC 48.15 percent; all hypotaxis 25.93 percent; MNP 66.67 percent; complex hypotaxis 22.22 percent; 3+ ELC 7.41 percent. The figures for MNP and complex hypotaxis are on the high side (possibly because of vv. 6–7) and may reveal redactional intervention, to be viewed in connection with the surrounding vv. 1–5, 11–28 (IES, see nn. 51, 141). 82  Exod 3:17–22 (twenty-four clauses, too small for analysis): 0–1 ELC 45.83 percent; all hypotaxis 20.83 percent; MNP 102.08 percent; complex hypotaxis 12.50 percent; 3+ ELC 4.17 percent. Although the basic syntax of this unit is close to the VoLB-2 style of 3:1–16, the high incidence of noun phrases (vv. 17, 18, 22) indicates IES elements (4:1–17: VoLB-1). 83  Exod 5:1–6:1 (106 clauses): 0–1 ELC 36.79 percent; all hypotaxis 27.36 percent; MNP 34.91 percent; complex hypotaxis 15.09 percent; 3+ ELC 7.55 percent. This tale is close to the IES, but the figures for the MNP fit the VoLB-2 style. A possible LBH feature is the use of ‫( אשר‬rather than ‫ )כי‬in Pharaohʼs question: ‫( ִמי יְ הוָ ה ֲא ֶשׁר ֶא ְשׁ ַמע ְבּקֹלֹו‬5:3). 84  Exod 7:19–24 (twenty-seven clauses, too small for analysis): 0–1 ELC 44.44 percent; all hypotaxis 18.52 percent; MNP 144.44 percent; complex hypotaxis 7.41 percent; 3+ ELC 7.41 percent. This episode is close to the VoLB, but the figures for the MNP fit the intricate style. However, this result may be connected to the exceptionally long noun phrase in v. 19 (attributed to P, together with v. 20a, because of the role of Aaron, which is not in keeping with v. 20b, ‫ ;)וירם‬see Baentsch, Exodus (see n. 75), 61–62; Holzinger, Exodus (see n. 38),

Storytelling and Redaction

459

The claim that the VoLB represents an ancient layer of the exodus narrative is buttressed by some mythic-magical features, apart from the plagues cycle.85 The scene at the nightly encampment stands out in its mythic-magic associations, as a divine attack on Moses is staved off by means of Zipporahʼs magic (Exod 4:24–26).86 Mythic-magic features are also in evidence in the notion of the staff endowed with divine power, the divine maṭṭê (“staff” or “rod”) ַ 4:20; 17:5, 9), which is paralleled by the idea given to Moses (‫ֹלהים‬ ִ ‫מ ֵטּה ָה ֱא‬, of Hadadʼs divine weapons that were granted to Zimri-Lim as a gift.87 One also notes the carving on the “throne of YHWH” (17:16, ‫ל־כּס יָ הּ‬ ֵ ‫ )יָ ד ַע‬and the divine writing: “The tablets were Godʼs work, and the writing was Godʼs writing, incised upon the tablets” (32:16; see also v. 15; 31:18). Significantly, the second revelation (IES) mentions tablets carved and inscribed by Moses (34:1, 4, 27, “like the first”).88 The stylistic differentiation has far-reaching implications for the understanding of the stratification and redaction of the book of Exodus.89 But a discussion of this subject requires a discussion of the way in which a text is to be read.

22–23. Discarding vv. 19–20a, one arrives at a VoLB-2 style: 0–1 ELC 42.11; all hypotaxis 21.05; MNP 52.63. On 7:14–18 (VoLB-2), see n. 68, above. 85  On the magical aspects of the plague cycle, see most recently G. A. Rendsburg, “Moses the Magician,” in Israelʼs Exodus in Transdisciplinary Perspective: Text, Archaeology, Culture, and Geoscience (ed. T. E. Levy et al.; Berlin: Springer, 2015), 243–258. 86  See in particular Rashbam on Gen 32:29; Y. Avishur , Studies in Biblical Narrative: Style, Structure, and the Ancient Near Eastern Literary Background (Tel Aviv, 1999), 137–158; B. Embry, “The Endangerment of Moses: Towards a New Reading of Exodus 4:24–26,” VT 60 (2010), 177–196; S. A. Geller, “The Struggle at the Jabbok,” JANES 14 (1982), 37–60, here 57–58; and F. P olak , “Theophany and Mediator: The Unfolding of a Theme in the Book of Exodus,” in Studies in the Book of Exodus: Redaction-Reception-Interpretation (ed. M. Vervenne; BETL 126, Leuven: Leuven University Press, 1996), 117–147, here 124–126. 87  On A.1968 (rev. 2′–4′), see J.-M. Durand, “Le mythologème du combat entre le Dieu de lʼorage et la mer en Mésopotamie,” MARI 7 (1993) 41–61, here 45. Allusions to this myth in Ebla have been detected by P. Fronzarolio, “Les Combats de Hadda dans les Textes dʼÉbla,” MARI 8 (1997), 283–290. The divine maṭṭê is viewed as the residue of a divine weapon, like mi-ṭi-šu (“his staff”) of Marduk (Enuma Elish IV 130), by S. E. Loewenstamm, The Evolution of the Exodus Tradition (trans. B. J. Schwartz; Jerusalem: Magnes, 1992), 147–154. 88  As against the mention of human chiseling (34:27), the first command to Moses preserves the notion of divine writing (v. 1), unless ‫ וְ ָכ ַת ְב ִתּי‬is comparable to ‫( וַ ֶיִּבן ְשֹׁלמֹה‬1 Kgs 6:14). 89  This is not the place to discuss the legal sections (largely VoLB-1) and covenant sections (VoLB-2) of Exod 21–23. On the parallels between the covenant sections and the texts from Mari, see below, p. 468.

460

Frank H. Polak

3 Ways of Reading 3.1 Literary-Genetic Criticism and the Aesthetics of Neoclassicism In modern redaction history, as in all diachronic-genetic study of the Hebrew Bible,90 strata are set apart by the criteria of literary criticism, which are presented as historical and objective. But what are the norms on which these scholarly perceptions are based? How does one arrive at norms that are valid for biblical literature? Literary criticism is not concerned with this question, and the Gunkel tradition is to a large extent based on untenable a priori considerations. Consequently, the presupposed norms are mainly those of the more conservative strands of Western literary taste. As John Barton puts it, “All literary studies must assume that even quite remote cultures have some affinities with our own.”91 This kind of reasoning disregards the sociocultural coordinates by which scholarly judgment is conditioned and the intuitive, unconscious adherence to the norms imposed by (neo)classicism or post-Renaissance literary culture in general.92 The classicistic stance of the diachronic-genetic approach is revealed by the terminology used by Jean Astruc in his description of repetition: “Can one point to some similar example and would one dare to impute, without proof, to Moses a mistake which no writer has ever committed?”93 Richard Simon argues that “[i]t is probable that if onely one Authour had composed this Work he would have explained himself in fewer words, especially in a History.”94 Thus, the historical, objective facade of the literary-critical approach conceals a  According to the picture sketched by Ska, “Plea” (see n. 3), and Otto, “Das postdeuteronomistische Deuteronomium” (see n. 3), the perspective of redaction history is entirely different from that of source analysis. However, the analytical procedures have not changed; see J. Jeon, The Call of Moses and the Exodus Story: A Redactional-Critical Study in Exodus 3–4 and 5–13 (FAT 60; Tübingen: Mohr Siebeck, 2013), 34–36, 43, 64–65. An exception is the cautious treatment of Dozeman, Exodus (see n. 39). 91  J. B arton , Reading the Old Testament: Method in Biblical Study (London: Darton, Longman and Todd, 1984), 28–29. 92  E. Vinaver, The Rise of Romance (Oxford: Clarendon, 1971), 70–76, 85 (on brevity). 93  J. Astruc, Conjectures sur les mémoires originaux dont il paroit que Moyse sʼest servi pour composer le Livre de la Genèse (Brussels: Fricx, 1753), 13: “A-t-on quelque example pareil à citer, & ose-t-on bien sans preuve imputer à Moyse une faute quʼaucun Écrivain nʼa jamais commise?” Other terms used are “de pareilles méprises,” “inexcusable,” “moins excusable,” “choquantes,” and “negligence” (16, 300–301, 359, 381). Transitions are regarded as “trop eloignées & trop brusques” (21). In his view, one of the advantages of the Documentary Hypothesis is that “it acquits Moses of the neglect and even mistakes that one dares to impute him, and which one bethinks to find in Genesis” (431). 94  R. Simon, A Critical History of the Old Testament (London: Davis, 1682), 37. Eichhornʼs methodology is not that different; see, for instance, J. G. E ichhorn , Einleitung in das Alte Testament (5 vols.; 4th ed.; Göttingen: Carl Eduard Rosenbusch, 1823–1824), 3:40, 48–49, 93, 113. Works by modern critics contain many expressions that au fond are evaluative rather than historical. 90

Storytelling and Redaction

461

classicistic stance that disregards the literary ways of a different culture. Since literary norms are culturally conditioned,95 our vista must include other cultures and other periods, particularly ancient Near Eastern literary sources in all their diversity. In the following considerations, I intend to discuss some aspects of the ESP in light of ancient stylistic patterning. Although I have followed the redaction-critical discussion rather closely, I will refrain from frequent references, for the sake of space. 3.2 Repetition Patterns and Delay The (neo)classicistic condemnation of repetition disregards the obvious fact that in Akkadian, Ugaritic, and Sumerian literary texts (and royal inscriptions) repetition is a frequent rhetorical figure.96 The failure to recognize a three-four repetition scheme has led to erroneous splitting up of the description of the flood in Gen 7:17–20.97 In the tale of Mosesʼs call, the splitting of repetition patterns causes grave difficulties. The doubling of the divine announcement in Exod 3:7, 9, both indicating the divine perception of Israelʼs suffering in Egypt, has been classified as redundant. Accordingly, in the redaction-critical analysis, the divine command to Moses to turn to the elders of Israel (v. 16) is often viewed as the direct continuation of the declaration of divine salvation (v. 8).98 However, contrary to this analysis, v. 16 does not form a perfect continuation of v. 8. The divine plan as unfolded at first (3:7–8) concerns the rescue of “my people in Egypt,” but the name of this people is not mentioned.99 And second, the identity of the 95  See in particular S. B. Heath, Ways with Words: Language, Life, and Work in Communities and Classrooms (Cambridge: Cambridge University Press, 1983), 18–57, 157–184. 96  U. C assuto , Biblical and Oriental Studies (trans. I. Abrahams; 2 vols.; Jerusalem: Magnes Press, 1973–1975), 2:29–33; V. A. Hurowitz, “The Priestly Account of the Building of the Tabernacle,” JAOS 105 (1985), 21–30, here 26–28; and for Sumerian, A. B erlin , Enmerkar and Ensuḫkešdanna: A Sumerian Narrative Poem (Philadelphia: University Museum, 1979), 18–25; idem, “Shared Rhetorical Figures in Biblical and Sumerian Literature,” JANES 10 (1978), 35–42; H. L. J. Vanstiphout, “Repetition and Structure in The Aratta Cycle: Their Relevance for the Orality Debate,” in Mesopotamian Epic Literature: Oral or Aural? (ed. M. E. Vogelzang and H. L. J. Vanstiphout; Lewiston: Mellen, 1992), 247–264. 97  B. W. Anderson, “From Analysis to Synthesis: The Interpretation of Genesis 1–11,” JBL 97 (1978), 23–39, here 35–36; R. W. L. Moberly, At the Mountain of God: Story and Theology in Exodus 32–34 (JSOTSup 22; Sheffield: JSOT Press, 1983), 29–31. 98  B aentsch , Exodus (see n. 75), 21, 23; P. W eimar , Die Berufung des Mose: Literaturwissenschaftliche Analyse von Exodus 2,23–5,5 (OBO 32; Freiburg: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht, 1980), 124–125; M. Greenberg, Understanding Exodus (New York: Behrman, 1969), 101–102; Gertz, Tradition und Redaktion (see n. 37), 282–290. 99  G. Fischer, Jahwe unser Gott: Sprache, Aufbau und Erzähltechnik in der Berufung des Mose (Ex 3–4) (OBO 91; Freiburg: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht,

462

Frank H. Polak

deity addressing Moses is only given as “the God of your father, the God of Abraham, the God of Isaac, and the God of Jacob” (v. 6). In v. 16 these partial identifications are replaced by the full name: “YHWH, the God of your fathers [. . .] has appeared to me.” This proclamation cannot be viewed as the simple continuation of the sayings of vv. 6–8, in which neither the speaker nor the target group are fully identified.100 The redaction-critical reconstruction, then, creates more problems than it solves. The lacking information is provided in the sayings that follow the announcement of the divine decision in vv. 6–8 and that mention, in addition to “my people” (v. 7), “the Israelites” (v. 9) and “my people, the Israelites” (v. 10) and explain the identity of the divine speaker (vv. 14–15). In both cases, the problem is the method of identification. Nowhere does the divine speech state that “my people in Egypt” (‫ע ִמּי ֲא ֶשׁר ְבּ ִמ ְצ ָריִ ם‬, ַ v. 7) is identical with “the Israelites” (‫בּנֵ י־יִ ְשׂ ָר ֵאל‬, ְ v. 9), but both terms are combined in the divine instruction to confront Pharaoh on behalf of “my people the Israelites” (‫ַע ִמּי ְבנֵ י־‬ ‫יִ ְשׂ ָר ֵאל‬, v. 10). Thus the identity of the people to be delivered is developed step by step, culminating finally in full recognition and the establishment of a close relationship to the speaking deity as “my people.” This scheme represents the literary method of delayed identification, which has been identified in biblical literature101 and is well-known in Akkadian and Ugaritic, and in particular in Sumerian, literature. For instance, the Exaltation of Inanna opens with a series of praises addressed to “the Lady of all the ‘divine powers,’ resplendent light, righteous woman clothed in radiance, beloved of An and Uraš!” (lines 1–2), mentioning the goddessʼs name at the end of lines 11–12: “a flood descending from its mountain, foremost one of heaven and earth, you are their Inanna.”102 The famous Poem of the Righteous Sufferer opens with the following proclamation: I will praise the lord of wisdom, the [deliberative?] god, [who lays] hold on the night, but lets free [the day] Marduk, the lord of wisdom.103 1989), 124, 127. 100  Did Moses have the knowledge necessary to identify the divine speaker and the target group? This question is, of course, disputable. However, the parallel narrative explains that Moses could not have known who was addressing him (6:2–3). Some cases of parallelism in the exodus narrative are discussed below, pp. 465–467. 101  W. G. E. Watson , Classical Hebrew Poetry: A Guide to its Techniques (JSOTSup 26; Sheffield: JSOT Press, 1984), 336–337. Sumerian and biblical “particularizing parallelism” are discussed by Berlin, “Shared Rhetorical Figures” (see n. 96), 35–38. 102  W. W. H allo and J. J. A. van D ijk , The Exaltation of Inanna (New Haven, CT: Yale University Press, 1968), 14–15; see also A. J. F errara , Nanna-Suenʼs Journey to Nippur (Studia Pohl, S.M., 2; Rome: Biblical Institute Press, 1973), 44–45, 82–84 (lines 1–8, 28–31); C. W ilcke , “Die Anfänge der akkadischen Epen,” ZA 67 (1977), 153–216, here 175–77, 181–82, 187–88, 191–95. 103  W. G. Lambert, Babylonian Wisdom Literature (Oxford: Clarendon, 1967), 343–344; see also F errara , Nanna-Suenʼs Journey (see n. 102), 44–45, 82–84 (lines 1–8, 28–31);

Storytelling and Redaction

463

In the Moses tale, the structure of the delay pattern is based on repetition, parallelism, and substitution in four steps, relating to the Israelites, the divine speaker, and the Egyptian oppressor. At first, the distress of the Israelites is stated in three terms (Exod 3:7): ‫ענִ י ַע ִמי ֲא ֶשר ְב ִמ ְצ ָריִ ם‬,ֳ which has been perceived visually; their outcry (‫)צ ֲע ָק ָתם‬, ַ which is heard; and their pains (‫)מ ְכא ָֹביו‬, ַ which are thus known.104 The second step announces the divine decision to save the suffering people, indicated only by a pronominal suffix (‫ל ַה ִצּילֹו‬,ְ v. 8). The third step (v. 9) repeats the divine perception of the suffering, paralleling v. 7, but substitutes “the Israelites” (v. 9a) for “my people” (v. 7a). In this description, two terms are repeated: “their outcry,” which has reached God (v. 9a//7bα), and the visual perception (v. 9b//7a), which now has a concrete object, “the oppression (‫)ה ַלּ ַחץ‬ ַ wherewith the Egyptians oppress (‫)ֹלח ִצים‬ ֲ them” (v. 9b, ASV). The fourth and final stage repeats the theme of deliverance (v. 10b) but now centers on Mosesʼs human role and substitutes the full identity of the people to be rescued: “to bring my people, the Israelites, out of Egypt.” The latter term parallels and condenses the circumscription of step two (v. 8, “to deliver them from the Egyptians, and to bring them up out of that land”).105 The innovations of the fourth stage include the identification of Pharaoh as the potentate to be confronted (v. 9a) and the specification of the general terms Egypt and Egyptians. Thus the opening stage highlights divine perception, whereas the final stage specifies what is to be done. The opening stage mentions “my people” and “its taskmasters” (v. 7), in contrast to the full identification of “my people the Israelites” and the specification of Pharaoh. The delayed-identification pattern appears likewise in the establishment of the identity of Mosesʼs interlocutor.106 The divine speech opens with the self-presentation as “the God of your father, the God of Abraham, the God of Isaac, and the God of Jacob” (Exod 3:6), not mentioning the divine name, unlike the divine self-presentation in Exod 6:2 and Gen 15:7; 17:1; 28:13; 35:11. Thus Moses has to ask for his interlocutorʼs name.107 It would, however, be highly inappropriate, and in interaction with a divine speaker positively dangerous,108 to ask straightforwardly, “Who are you?,” a question always put in the mouth of

Wilcke, “Die Anfänge” (see n. 102), 175–77, 181–82, 187–88, 191–95. 104  This threefold parallelism is discussed below, pp. 465–466. In Exod 2:24–25, the rhythmic and semantic shaping is less structured and thus hardly the source for 3:7. 105  As against the view of Gertz, Tradition und Redaktion (see n. 37), 277–279, the stylistically flat promise of Gen 46:4 could never form the example for Exod 3:8. The call pattern shared by Exod 3:6–12 and Gen 46:2–4 is formulaic; see Polak, “Oral Platform” (see n. 1), p. 421, n. 67. 106  A delay of a different kind is involved in the reference to YHWH in Gen 18:13–14. 107  As noted by Fischer, Jahwe unser Gott (see n. 99), 146, in the context of a polytheistic culture, the identity of the speaking deity is in need of clarification. 108  Mosesʼs fears are demonstrated by his hiding his face (v. 6).

464

Frank H. Polak

persons in authority.109 Hence Moses raises the question vicariously, by means of “a friend”:110 “When I come to the Israelites and say to them ‘The God of your fathers has sent me to you,’ and they ask me, ‘What is his name?’ what shall I say to them?” (v. 13). The information provided in the divine response is as mysterious as Mosesʼs question was evasive. Appropriately, the first person addressed is Moses himself: “Thus God said to Moses, ‘ʾEhyê ʾăšer ʾehyê’ ” (v. 14a), unlimited in its implications but without going beyond implication.111 The message to the Israelites appears in second place: “Thus shall you say to the Israelites, ‘ʾEhyê has sent me to you’ ” (v. 14b). The information Moses had asked for is only given in the next verse, repeating the notion of sending: “Thus shall you speak to the Israelites: ‘YHWH, the God of your fathers, the God of Abraham, the God of Isaac, and the God of Jacob, has sent me to you’ ” (v. 15a). This is a development of the previous statement by repetition and substitution: name and full title take the place of the enigmatic ʾehyê. This statement is rounded off by a particularly solemn proclamation in full parallelism: “This is my name (‫)שׁ ִמי‬ ְ forever, this is my appellation (‫ )זִ ְכ ִרי‬for all eternity” (v. 15b).112 The divine answer, then, consists of three stages, the last of which is marked by a solemn elaboration. Literary-stylistic analysis shows that this configuration leads from hāyâ to šālaḥ and thus presents the divine name as a promise of divine action,113 ʾehyê suggesting, “I shall step in.”114 Only after this gradual unfolding of the divine name and its implications does the narrative proceed to the actual program for action (v. 16), which thus forms the fourth stage of a step-by-step transition that had its starting point in the divine self-presentation (v. 6). A second feature that demands our attention is the use of parallelism.

 The father: Gen 27:18, 32; the king: 2 Sam 1:8; and similarly: 1 Sam 17:58; 30:13.  See A. F ern , “ ‘Iʼm Asking for A Friend’: 46 Questions You Would Never Ask For Yourself,” Elite Daily, Sept. 16, 2014, http://elitedaily.com/humor/questions-you-would-never-ask-for-yourself/756367/ (accessed 02/24/15). 111  A semiotic analysis of this saying is offered by J.-P. S onnet , “Ehyeh asher ehyeh (Exodus 3:14): Godʼs ‘Narrative Identity’ among Suspense, Curiosity, and Surprise,” Poetics Today 31 (2010), 331–351. 112  See pp. 465–467. 113  A. G. van Daalen, “The Place where YHWH Showed Himself to Moses: A Study of the Composition of Exodus 3,” in Voices from Amsterdam: A Modern Tradition of Reading Biblical Narrative (ed. M. Kessler; SBL Symposium Studies; Atlanta: Scholars Press, 1994), 133–144, here 140–141; and Polak, “Theophany and Mediator” (see n. 86), 122–124. 114  For this interpretation, and hāyâ as a verb of motion in general, see F. P olak , “Hebrew hāyāh: Etymology, Bleaching, and Discourse Structure,” in Tradition and Innovation in Biblical Interpretation: Studies Presented to Professor Eep Talstra on the Occasion of His Sixty-Fifth Birthday (ed. W. T. Van Peursen and J. W. Dyk; SSN 57; Leiden: Brill, 2011), 379–398, here 395. 109 110

Storytelling and Redaction

465

3.3 Parallelism in the Exodus Narrative Parallelism, originally viewed as the hallmark par excellence of poetry, has long been recognized as a valid option in narrative.115 The tale of the primeval garden contains a number of famous cases.116 Many scholars have recognized the use of parallelism in the divine invitation to the covenant (19:3–6).117 Thus we should not disregard the tricolon in the description of divine perception of Israelite oppression in Egypt: Exodus 3:7 ‫ת־מ ְכא ָֹביו‬ ַ ‫ ִכּי יָ ַד ְע ִתּי ֶא‬// ‫ת־צ ֲע ָק ָתם ָשׁ ַמ ְע ִתּי ִמ ְפּנֵ י נֹגְ ָשׂיו‬ ַ ‫ וְ ֶא‬// ‫ת־ענִ י ַע ִמּי ֲא ֶשׁר ְבּ ִמ ְצ ָריִ ם‬ ֳ ‫יתי ֶא‬ ִ ‫ָראֹה ָר ִא‬ I have marked, yes marked the plight of my people in Egypt // and have heeded their outcry because of their taskmasters // yes, I am mindful of their sufferings.118

One notes the threefold parallelism of perception terms: ‫יתי‬ ִ ‫ר ִא‬,ָ ‫שׁ ַמ ְע ִתּי‬, ָ and the conclusion, ‫כּי יָ ַד ְע ִתּי‬, ִ 119 as the third term of the tricolon.120 The largely poetic use  J. S. K selman , “The Recovery of Poetic Fragments from the Pentateuchal Priestly Source,” JBL 97 (1978), 161–173, here 162–167; J. L. K ugel , The Idea of Biblical Poetry (New Haven, CT: Yale University Press, 1981), 59–62; various essays in J. C. de Moor and W. G. E. Watson (eds.), Verse in Ancient Near Eastern Prose (AOAT 42; Kevelaer: Butzon & Bercker; Neukirchen-Vluyn: Neukirchener Verlag, 1993); and last, but not least, F. I. Andersen, “What Biblical Scholars Might Learn from Emily Dickinson,” in Words Remembered, Texts Renewed: Essays in Honour of J. F. A. Sawyer (ed. J. Davies et al.; JSOTSup, 156; Sheffield: Sheffield Academic Press, 1995), 52–74. For the Mesha Stela, see idem, “Moabite Syntax,” Orientalia 35 (1966), 81–120, here 85–88; J. C. de Moor, “Narrative Poetry in Canaan,” UF 20 (1988), 149–177. 116  See Gen 2:5–9 (and similarly, 7:11; 8:2); C. A. Briggs, “The Poem of the Fall of Man,” Reformed Quarterly Review 32 (1885), 311–333; and P olak, “Poetic Style and Parallelism in the Creation Account (Gen. 1.1–2.3),” in Creation in Jewish and Christian Tradition (ed. H. Reventlow and Y. Hoffman; JSOTSup 319; Sheffield: Sheffield Academic Press, 2002), 2–31, here 22, 28. 117  H olzinger , Exodus (see n. 38), 64, 66–67 (modeled on Deut 32:11); N oth , Exodus (see n. 52), 157. 118  In this verse, ‫ כי‬is to be taken as deictic marker, like Ugaritic k. For ‫ ידע‬as “paying attention,” see W. G esenius , Thesaurus philologicus criticus linguæ Hebrææ et Chaldææ Veteris Testamenti (3 vols.; Leipzig: Vogel, 1829–1858), 2:570. 119  The triad ‫ ידע‬/ ‫ שמע‬/ ‫ ראה‬appears in Isa 6:9; Lev 5:1; Deut 29:3; Neh 6:16. The collocation ‫ ידע‬/ ‫ ראה‬for divine perception appears in Ps 31:8; 138:6; Job 11:11, and in prose in Gen 18:21; Exod 2:25; in general, in poetry, in Deut 33:9; Isa 5:19; 6:9; 29:15; 41:20; 44:9, 18; 58:3; 61:9; 66:14; Jer 2:19, 23; 5:1; 11:18; 12:3; Ezek 10:20; 14:23; Pss 31:12; 74:9; Qoh 6:5; 8:16; in prose, in Exod 6:3; 16:15; Deut 4:9, 35; 11:2; Judg 13:21; 1 Sam 12:17; 14:38; 16:18; 17:28; 18:28; 23:22, 23; 24:12; 25:17; 26:12; 18:29; 24:13; 1 Kgs 20:7, 13, 22; 2 Kgs 5:7; Neh 4:5. The collocation ‫ ראה‬/ ‫ שמע‬appears in poetry in Num 24:16–17; 2 Kgs 19:16 (Isa 37:17); 20:5 (Isa 38:5); Isa 6:9, 10; 18:3; 21:3; 29:18; 30:30; 32:3; 33:15, 19; 42:18, 20; 52:15; 64:3; 66:8, 19; Jer 4:21; 5:21; 23:18; Ezek 1:28; 12:2; Mic 6:9; Ps 45:11; 48:9; 66:18; 106:44; Job 13:1, 29:11; 42:5; Prov 20:12; Cant 2:12, 14; Qoh 1:8; Lam 1:18; Dan 9:18; and in prose in Gen 24:30; Exod 8:11; Num 14:22; Deut 4:12, 28, 36; 5:24; 18:16; 26:7; Judg 13:23; 1 Kgs 3:28; 10:7 (2 Chr 9:6); 12:16 (2 Chr 115

466

Frank H. Polak

of ‫ ַ מ ְכאֹב‬121 is in keeping with the prosody of this logos. The use of parallelism suits the solemn tone of this proclamation and is matched by the structuration of the bicolon in v. 15b: Exodus 3:15b

‫ וְ זֶ ה זִ ְכ ִרי ְלד ֹר דּ ֹר‬// ‫ה־שּׁ ִמי ְלע ָֹלם‬ ְ ֶ‫ז‬ This is my name forever, and this is my appellation for all generations.

By the same token, one notes the tricola in the narrative description: Exodus 1:7

‫אָרץ א ָֹתם‬ ֶ ‫ וַ ִתּ ָמּ ֵלא ָה‬// ‫ וַ יִּ ְרבּוּ וַ יַּ ַע ְצמוּ ִבּ ְמאֹד ְמאֹד‬// ‫וּבנֵ י יִ ְשׂ ָר ֵאל ָפּרוּ וַ יִּ ְשׁ ְרצוּ‬ ְ But the Israelites were fertile and prolific, and multiplied and increased very greatly,122 and the land was filled with them. Exodus 1:12

‫ וַ יָּ ֻקצוּ ִמ ְפּנֵ י ְבּנֵ י יִ ְשׂ ָר ֵאל‬// ‫ וְ ֵכן יִ ְפר ֹץ‬// ‫ ֵכּן יִ ְר ֶבּה‬// ‫וְ ַכ ֲא ֶשׁר יְ ַענּוּ אֹתֹו‬ But the more they were oppressed, the more they increased, and the more they spread out, so that they came to dread the Israelites.

A significant instance of poetic patterning is revealed by the incantation uttered by Zipporah when warding off the divine attack against Moses.123 This charm would have the form of an expanded colon (staircase parallelism)124 were it not for the quotation formulae:

10:16); 2 Kgs 13:4; Ezek 40:4; 44:5; Neh 9:9. The dismissal of the third colon by W eimar , Berufung (see n. 98), 40–41, is criticized by G ertz , Tradition und Redaktion (see n. 37), 286–287. 120  On this “a-b-c pattern,” see M. W eiss , The Bible from Within: The Method of Total Interpretation (Jerusalem: Magnes, 1984), 251–255; T. C ollins , Line-Forms in Hebrew Poetry (Studia Pohl, S.M., 7; Rome: Biblical Institute Press, 1978), 223–225. 121  Note Ps 32:10; 38:18; 69:27; Isa 53:3–4; Jer 30:15; 45:3; 51:8; Job 33:19; Eccl 1:8; 2:23; Lam 1:12, 18; 2 Chr 6:29. 122  The phrase ‫( ַרב וְ ָעצוּם‬Exod 1:9) represents a poetic pair in Amos 5:12; Isa 8:7; 53:12; Joel 2:2, 11; Mic 4:3; Zech 8:22; Ps 35:13; 135:10; Prov 7:26; and in prose, Num 32:1; as junction, Deut 7:1; in the inverse order, 9:14; 26:5; see E. Blum, “Die literarische Verbindung von Erzvätern und Exodus,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 119–156, here 148. The pair ‫רבה‬-‫ פרה‬is frequent in the creation-flood sequence (Gen 1:22, 28; 8:17; 9:1, 7) and the El Shaddai blessings (17:20; 28:3; 35:11; 48:4; and similarly 47:27); further only in Lev 26:9; Ezek 36:11; and in Jer 3:16; 23:3. The sound play suggests a poetic origin. The use of ‫ שרץ‬for humans fits Gen 9:7 but not the other texts in which this root is used. 123  See P olak , “Theophany and Mediator” (see n. 86), 125–126; Avishur , Studies in Biblical Narrative (see n. 86), 153–154. 124  Another case of the expanded colon is Samsonʼs saying (Judg 15:16); for a similar analysis of Judg 4:18, see W. G. E. Watson, Traditional Techniques in Classical Hebrew Verse (JSOTSup 170; Sheffield: Sheffield Academic Press, 1994), 259.

Storytelling and Redaction

467

Exodus 4:25b, 26b

‫ן־ד ִמים ַא ָתה ִלי‬ ָ ‫אמר) ִכי ֲח ַת‬ ֶ ֹ ‫(וַ ת‬ ‫– ַלּמּוֹלת‬ ‫(אז ָא ְמ ָרה) ֲח ַתן ָד ִמים‬ ָ (And she said) “You are truly ‘protected125 by blood’ to me!” (Then she said) “ ‘Protected by blood’ – by the circumcision.”

3.4 The Connection between the Episodes On the level of the microtext, then, the stylistic patterns of the exodus narrative, as in biblical prose in general, are far richer and more complex than allowed for by the presuppositions of literary criticism. The case of macrotextual connectivity is no different. Many scholars have been troubled by the assertion “A long time after that, the king of Egypt died” (Exod 2:23aα), detached as it is from the sequence that speaks of the fate of the Israelites who “were groaning under the bondage and cried out” (v. 23aβ). Wellhausen solves this problem by means of the hypothesis that the note on the kingʼs death was originally connected to the command to Moses: “Go back to Egypt, for all the men who sought to kill you are dead” (4:19).126 Moreover, this command seems problematic, since it comes immediately after Mosesʼs declaration that he intends to return (v. 18). However, the supposed connection of 2:23 to 4:19 is hardly immediate and certainly not “lückenlos”: without previous mention of a divine message to Moses, the present announcement would come out of the blue.127 It is true that the divine call in the Abraham tale (Gen 12:1–3) likewise lacks all preparation, but this case is different. The pericope preceding Abrahamʼs call lacks any details apart from the family connections (Gen 11:27–30), whereas the Moses narrative details the birth of the hero, his exploits in Egypt, his flight to Midian, and his acceptance by the family of the Midianite priest. Thus the sudden appearance of the divine announcement in Exod 4:19 would hardly be in keeping with the method of the preceding episodes. Moreover, in the critical reconstruction, the divine address  Since the expression “bridegroom of blood” defies explanation, I adhere to the derivation from Akkadian ḫatānu, “to protect,” first proposed but in the end rejected by J. Hehn, “Der ‘Blutbräutigam’ Ex 4 24–26,” ZAW 50 (1932), 1–8, here 3; and established definitely by J. Blau, “Ḥatan Damīm,” Tarbiz 26 (1957), 1–3 (Hebrew, with summary in English). 126  J. W ellhausen , Die Composition des Hexateuchs und der historischen Bücher im Alten Testament (4th ed.; Berlin: de Gruyter, 1963), 71; B lum , “Literarische Verbindung” (see n. 122), 123, n. 20, with further references. The repetition of the note on Pharaohʼs death (LXX 4:19) probably represents secondary adaptation, like the harmonistic expansions in, e.g., 6:20; 8:1–2, 12–13, 18, 28; 9:8–9; 10:4, 6, 12; 11:3, 10; see B. Lemmelijn, A Plague of Texts? A Text-Critical Study of the So-Called ‘Plagues Narrative’ in Exodus 7:14–11:10 (OTS 56; Leiden: Brill, 2009), 150–196, 210–216. 127  This problem has been recognized, but not solved, by Wellhausen, Die Composition (see n. 126). For criticism of this analysis, see B erner , Die Exoduserzählung (see n. 44), 57–58. 125

468

Frank H. Polak

does not include any call to action, although action is the main point. Thus the scholarly construction is extremely misleading. In some form the call narrative is indispensible. The problematic appearance of the divine instruction to return could easily be viewed as encouragement, since Moses did not utter any formal assent to the divine call. The intrusive appearance of the announcement of Pharaohʼs death, which is not directly connected to any theme in its context, fits a well-known pattern in biblical narrative, which was discovered by the early medieval French exegetes R. Joseph Kara and R. Shmuel ben Meir (Rashbam):128 the opening of the narrative mentions data that will not play a role in the tale until after a series of intervening passages, by way of anticipatory preparation, ‫שלא תתמה‬, “in order that you may not wonder.” 3.5 Narrative Patterns Many discussions of the covenant episode in Exod 19–24 discern an embarrassing amalgam of variant traditions.129 However, a very different picture emerges when we take the texts from Mari into account. The descriptions of the covenant procedures in these texts130 show that the narrative of the proceedings in Exod 19–24 is based on the customary usage of Northwest Semitic political culture. The “invitation” to the covenant in 19:3–6 is paralleled by the preliminary steps embodied by the “little tablet” (ṭuppum ṣeḫrum), whereas the agreement itself is committed to writing in the “big tablet” (ṭuppum rabûm). These documents also evidence the doubling of oath ceremonies at the residences of both partners in “bilocal” ceremonies, like the doubling of the ceremonies near the Israelite camp at the foot of the mountain (24:4–8) and in the divine abode (vv. 9–11).131 128  J. J acobs , “Rabbi Joseph Kara as an Exegete of Biblical Narrative: Discovering the Phenomenon of Exposition,” JSQ 19 (2012), 73–89; G. Brin, Studies in the Biblical Exegesis of R. Joseph Qara (Tel Aviv: Rosenberg School of Jewish Studies, 1990), 86–90. 129  See E. Z enger , Die Sinaitheophanie: Untersuchungen zum jahwistischen und elohistischen Geschichtswerk (Würzburg: Echter-Verlag; Stuttgart: Katholisches Bibelwerk, 1971); W. O swald , Israel am Gottesberg: Eine Untersuchung zur Literaturgeschichte der vorderen Sinaiperikope Ex 19–24 und deren historischem Hintergrund (OBO 159; Universitätsverlag Freiburg; Göttingen: Vandenhoeck & Ruprecht, 1998). 130  D. Charpin, “Les Representants de Mari à Babylone (I),” in Archives Épistolaires de Mari I (ed. D. Charpin et al.; ARM 26; 2 vols.; Paris: Éditions Recherche sur les Civili­ sations, 1988), 2:139–205, here 144; idem , “Un traité entre Zimri-Lim de Mari et Ibālpī-Il II dʼEšnunna,” in Marchands, Diplomates et Empéreurs: Études sur la civilisation mésopotamienne offertes à Paul Garelli (ed. D. Charpin and F. Joannès; Paris: Éditions Recherche sur les Civilisations, 1991) 139–166; and Polak, “The Covenant at Mount Sinai in the Light of Texts from Mari,” in Sefer Moshe: The Moshe Weinfeld Jubilee Volume; Studies in the Bible and the Ancient Near East, Qumran and Post-Biblical Judaism (ed. C. Cohen et al.; Winona Lake, IN: Eisenbrauns, 2004), 119–134, here 123–125. 131  See Polak, “Covenant and Mari Texts” (see n. 130), 125.

Storytelling and Redaction

469

The conclusion proposed in these paragraphs is not that the diachronic-genetic enterprise should be abandoned. Far from that. Redaction-historical study is a necessity by virtue of the protracted and problematic transmission history. However, we must not disregard the particular stylistics and rhetorics of biblical narrative as an ancient Near Eastern text, the possibilities revealed by ethnopoetic studies, and the achievements of literary theory, from New Criticism and Werkinterpretation to narratology, cognitive poetics, and complexity theory.

4 Redactional Stratification in the Exodus-Sinai Narrative The findings concerning syntactic-stylistic differentiation indicate that only part of the sections in the tales of the exodus and the conclusion of the covenant are linguistically marked as redactional revisions or expansions. In the tale of Mosesʼs call, the IES only appears in a redactional expansion (Exod 3:17–22), like the expansions/revisions in the tale of the conclusion of the covenant at Mount Sinai (19:9b,132 20–25). The IES register in the tale of the first confrontation with Moses (5:1–6:1) points to a revision in light of the ensuing second revelation to Moses (6:2–13; 6:26–7:6), which serves as a response to his complaints.133 In some of these sections, the role of the redaction or later narrator is well-known, and with regard to 5:1–6:1 a Persian-era origin has recently been proposed.134 Thus we note a convergence between some aspects of redaction history and syntactic-stylistic analysis. Accordingly, syntactic-stylistic analysis is a helpful tool in diachronic-genetic study. This tool is all the more important since it is not dependent on exegetical considerations. The episodes of the plagues include a number of passages in the elaborate style: Exod 7:8–13 (P, the introductory sign of the serpents), 9:1–12 (the pestilence of livestock and boils), 9:13–26 (the first part of the plague of hail), and 11:1–10 (the announcement of the plague of the firstborn). The redactional input in many of these passages is well-known in diachronic-genetic study. In particular we note the role of the plague of hail, which, as the seventh plague, forms the first high point in the series of the plagues135 but is introduced by a passage that points to another seventh item, namely, pestilence (9:14–15). One could view this allusion as a mere redactional expansion, aimed at underlining  On the repetitive resumption of vv. 9a/10a, see n. 48, above.  See Greenberg, Understanding Exodus (see n. 98), 148–149. 134  See Gertz, Tradition und Redaktion (see n. 37), 343–344; Berner, Die Exoduserzählung (see n. 44), 97, 137–149; H. Utzschneider and W. Oswald, Exodus 1–15 (IEKAT; Stuttgart: Kohlhammer, 2013), 159–160. 135  S. E. Loewenstamm, “An Observation on the Source-Criticism of the Plague-Pericope,” VT 24 (1974), 374–378; see also idem, Exodus Tradition (see n. 87), 94–96, 103; S.B Noegel, “The Significance of the Seventh Plague,” Bib 76 (1995), 532–539. 132 133

470

Frank H. Polak

the ensuing theodicy (9:16; 10:1–2, all IES).136 However, the poetic tradition presents us with pestilence as the seventh and last plague (Ps 78:50),137 alongside and conflated with the death of the firstborn (v. 51). This sequence is often considered secondary. It is, however, to be borne in mind that two other plagues are related to the pestilence: boils, which involves symptoms that are comparable to pestilence though far less grievous, and the pestilence of livestock, which in Ps 78 precedes pestilence directly (v. 48, MT ‫ל ָבּ ָרד‬,ַ but the lethal epidemic is indicated by Symmachusʼs λοιμῷ/‫)ּד ֶבר‬. ֶ 138 It is highly significant, then, that in Exod 9 the plagues preceding hail are close to the pestilence theme, the pestilence of livestock hitting the animals and the boils as less grievous emulation of the pestilence.139 Notably, the passages describing these two plagues are both couched in the IES, which is also much in evidence in the first sections of the plague of hail. However, the closing section of this plague adheres to the lean, brisk style (type 1, with type-2 intrusions),140 indicating the oral-epic background of the hail plague itself and of the epiphanic fire involved (9:24). Thus the first high point of the plagues narrative embodies redactional revision, affecting the section 9:1–10:2 by interpolation (9:14–17; 10:1–2) and rewriting. I conclude, then, that syntactic-stylistic analysis confirms extensive redactional activity in the presentation of the plagues. In the continuation of the narrative, large-scale redactional revision or expansion is apparent in the announcement of the final plague (Exod 11:1–10), the description of the inception of the march (12:38–39, 42; 13:3–10, 17–19, 21–22), and the P (or Dtr) sections in the episodes of the Sea of Reeds and the manna (14:1–4, 8–10*, 15–18, 22–23, 27*–29; 16:1–3, 6–12, 15b–21a, 22–26, 32–35). In addition one notes the Jethro tale (Exod 18), part of the last sections of the book of the covenant (23:1–19), and parts of the second revelation to Moses, including the section often referred to as the Cultic Decalogue (34:1–4, 11–28).141

 So, for instance, Noth, Exodus (see n. 52), 80.  Loewenstamm, “Observation” (see n. 135), 376–377. 138  In Ps 78:48, the reading λοιμῷ (Syrohexapla: lmawtānā) is matched by ‫ל ְר ָשׁ ִפים‬,ָ “pestilence,” like the parallelism in Hab 3:5 (similarly Deut 32:34), as discussed by Loewenstamm, “Observation” (see n. 135), 377. The reading ‫ ַל ָבּ ָרד‬reflects an adaptation to the description of the hail hitting both men and animals (Exod 9:19, 22, 25; unlike Ps 105:33). 139  The pairing of the plague of “boils” (9:9, ‫)שׁ ִחין פּ ֵֹר ַח ֲא ַב ְע ֻבּעֹת‬ ְ and the pestilence of the livestock is noted by M. Greenberg, “The Redaction of the Plague Narrative in Exodus,” in Near Eastern Studies in Honor of W. F. Albright (ed. H. Goedicke; Baltimore: Johns Hopkins University Press, 1971), 243–252, here 246. 140  Type-2 intrusions are indicated by the strikingly large number of subordinate clauses, including many clauses in complex subordination. 141  See nn. 51, 81, above, and S. Bar-On, “The Festival Calendars in Exodus XXIII 14–19 and XXXIV 18–26,” VT 48 (1998), 161–195. 136 137

Storytelling and Redaction

471

Thus syntactic-stylistic analysis provides independent confirmation of a series of observations that were made in diachronic-genetic analysis. However, two passages are exceptional. First, the plague of lice, commonly ascribed to P because of its parallels with the section on the serpents,142 is couched in the VoLB type-1 style. This finding is surprising but lacks all significance since the unit at stake is too small for analysis (twenty clauses only). Second, earlier source criticism ascribes the tale of Jethro (Exod 18) to the Elohistic stratum, but in light of its syntactic-stylistic profile this tale in its present form is probably to be ascribed to a secondary stratum.143 Notably, this tale reflects the same kind of rationalization as the Deuteronomic recasting of the appointment of the elders (Deut 1:8–17) and contrasts strongly with the parallel narrative, with its emphasis on prophetic/magic inspiration (Num 11:16–17, 24–30; VoLB-1). One could speculate that the Jethro tale in Exodus represents a critical rejoinder to the tale of the elders that centers on Mosesʼs weakness vis-à-vis the populace (Num 11:11–14). However, many sections that in recent diachronic-genetic discussions have been attributed to secondary sections reveal a style that is quite different from the elaborate style of the redactional episodes discussed in the present section. The VoLB style characterizes such units as the tale of the forced labor (Exod 1:1–22),144 the first call of Moses (3:1–16), and the Sinai pericope (ch. 19–24), which thus are all anchored in the oral-epic platform.

5 Structures Underlying the Thematic Flow of the Exodus-Sinai Narrative 5.1 Narrative Sequence and Oral-Epic Platform The syntactic-stylistic profile of the ESN, dominated as it is by the VoLB style, indicates that this narrative is grounded in an oral-epic platform that includes all stages of the tales of the exodus, from the description of the bondage in Egypt and the revelation to Moses, through the episodes of the plagues, the rescue near the Sea of Reeds, and the wanderings through the desert, until the revelation at Mount Sinai. The heart of this platform is the pilgrimage pattern, posited by  So also Greenberg, “Plague Narrative” (see n. 139), 245–248.  The difficulties regarding Mosesʼs sons Gershom (2:22; 18:3) and Eliezer (18:4) and the appellations of Mosesʼs father-in-law, variously called Reuel (2:18), Jethro (3:1; 4:18), and Jether (4:18), are discussed by Berner, Die Exoduserzählung (see n. 44), 123–126. 144  The syntactic pattern of 1:21, with ‫ ויהי כי יראו‬as causative clause (LXX ἐπειδὴ ἐφοβοῦντο), differs from the temporal construction described by Joosten, “Wayhi” (see n. 11), 56, and thus remains problematic. On ‫ ָל ֶהם‬as dual form (Puah and Shiphra, v. 15; cf. Gen 31:9), see G. A. Rendsburg, “Dual Personal Pronouns and Dual Verbs in Hebrew,” JQR 73 (1982), pp. 38–58, here 42–43. 142 143

472

Frank H. Polak

Mark Smith as the work of the Priestly author.145 In this pattern, the “wilderness” represents a liminal space, in the terms of Victor Turner.146 This framework is not dependent on redaction processes within the Deuteronomic or the Priestly school but forms an independent platform in its own right. I do not claim authorial unity.147 I do claim, however, that the tales in this style all represent one and the same narrative platform, which came into being in oral performance, was used by various different narrators, and in the end provided the plot structure and the stylistic norms for the written compositions. The extent of the redactional layer, whether Deuteronomic or Priestly, is rather limited. These findings necessarily raise the question what it is that keeps the ESN together. In my view, the coherence of this narrative is dependent on a number of overarching themes, first and foremost the theme of the theophany in tandem with the growth of Mosesʼs insight into the ways of divine guidance.148 The herdsman who at first shrinks away from the divine appearance (Exod 3:6) in the end requests to be shown the divine kābôd (33:18).149 Particularly significant is the transition from forced labor and bondage in the service of Pharaoh to the worship of Israelʼs divine overlord.150 Both stages are indicated by the root ‫עבד‬. This term appears in the description of the forced labor as “imposed [. . .] labors” (Exod 1:13–14) but also in the representation of the service of God: “You shall worship (‫)תּ ַע ְבדוּן‬ ַ the deity at this mountain” (3:12, similarly 4:23).151 The transition is likewise expressed in the reviews of 145  M. S. S mith , The Pilgrimage Pattern in Exodus (JSOTSup 239; Sheffield: Sheffield Academic Press, 1997), 192–263. 146  V. W. T urner , The Ritual Process: Structure and Anti-Structure (Ithaca, NY: Cornell University Press, 1969); R. L. Cohn, The Shape of Sacred Space (AAR Studies in Religion 4; Chico, CA: Scholars Press, 1981), 7–23; Dozeman, Exodus (see n. 39), 347–348. 147  Notably, the parallel tales of Gen 12:10–20 and Gen 20 both are couched in the VoLB-2 style, and both have their place in the current plot sequence, although they ultimately represent different versions of the same theme. 148  For the importance of this theme, see J. P. Fokkelman, “Exodus,” in The Literary Guide to the Bible (ed. R. Alter and F. Kermode; Cambridge, MA: Belknap, 1987), 56–65, here 63; G. W. Savran, Encounter with the Divine: Theophany in Biblical Narrative (JSOTSup 420; London: T&T Clark, 2005), 27–28, 75–78, 108–116, 198–229, B. D. Sommer, “Revelation at Sinai in the Hebrew Bible and in Jewish Theology,” JR 79 (1999), 422–451; and see Polak, “Theophany and Mediator” (see n. 86), 116–144. 149  The relationship between the revelation in Exod 33–34 and the call narrative is discussed in Polak, “Theophany and Mediator” (see n. 86), 143–146. 150  J. D. Levenson, The Hebrew Bible, the Old Testament and Historical Criticism: Jews and Christians in Biblical Studies (Louisville: Westminster John Knox, 1993), 127–59, here 142–144. 151  So also 9:1, 13 (IES); 10:3, 8, 11, 26; 12:31 (all VoLB); as against the subjection to Egypt (14:5, 12; 20:2, all VoLB). The phrase ‫( ֲע ָב ַדי ֵהם‬Lev 25:42, 55, VoLB-2) has been commented upon by E. L. Greenstein, “Book of Exodus,” in The HarperCollins Study Bible (ed. W. A. Meeks; San Francisco: Harper-Collins, 1993), 86.

Storytelling and Redaction

473

the past that juxtapose the divine acts in Egypt and the relationship to the divine overlord (19:4; 20:2–3). A metaphorical expression of this contrast is embodied by the geographic symbolism of the Nile and Sea of Reeds, as water,152 and Sinai, as mountain. Water and mountain (or rock) appear together in a typical, often symbolic landscape (Jer 46:18; Ps 46:3–4; 98:8; 104:10) and thus form a thought pattern,153 which comes to the fore in many recastings of the exodus narrative (Ps 114:3–6).154 In particular, we note the transition pattern that leads from the Sea of Reeds, through the wells of Marah and Elim (Exod 15:22–27) and the scenes at Massah and Meribah (17:1–7) and Rephidim (17:8–13), to Mount Sinai. This pattern includes the triad155 water, wood or maṭṭê (15:25, 27; 17:5, 9),156 and rock, hill, or stone (17:6, 9–10, 12)157 and thus ties the point of origin, the Sea of Reeds, as water, to the goal, Mount Sinai, as rock. This configuration reflects an underlying thought pattern or symbolic landscape that holds the entire ESN together. In this symbolic landscape the wood fulfils a mediating role, as burning bush, divine staff (3:2; 4:2, 20), or divine ark (VoLB/ epic: Num 10:33–35; 14:44; IES/P recasting: Exod 25:10). 5.2 The Exodus-Sinai Narrative and the Abraham-Jacob Narrative In more than one sense, the ESN is a counterpoint to the Abraham-Jacob narrative. Though both frameworks are based on migration and journey, the tales of Jacob and Joseph ultimately reunite the patriarchal family in Egypt, whereas the ESN leads the Israelites out of Egypt to Mount Sinai in order to establish the covenant with the divine savior, which in a sense is the counterpart of the 152  P. Sabo, “Drawing out Moses: Water as a Personal Motif of the Biblical Character,” in Thinking of Water in the Early Second Temple Period (ed. E. Ben Zvi and C. Levin; BZAW 461; Berlin: de Gruyter, 2014), 409–436; D. Edelman , “The Nile in Biblical Memory,” in Thinking of Water in the Early Second Temple Period (ed. E. Ben Zvi and C. Levin; BZAW 461; Berlin: de Gruyter, 2014), 77–102, here 83–85. 153  The dyad water/mountain is found in, e.g., Isa 54:9–10; Deut 8:7; 33:19; Isa 7:19; 10:26; 11:9; 17:13; 18:7; 25:10; 30:25; 32:2; as echoes of the Massah-Meribah tale: Num 20:8, 10–11; Deut 8:15. For parallels in Ugaritic, Sumerian, and Akkadian poetry, see Polak, “Water, Rock and Wood: Structure and Thought Pattern in the Exodus Narrative,” JANES 25 (1997), 19–42, here 34–38. 154  The same contrast figures in Exod 15:5, 17; Ps 78:15–16, 20, 53–54; in the flood narrative: Gen 7:19, 20; 8:4–5; and in creation poetry: Ps 104:6; Prov 8:24–25; Isa 40:12; and as metaphor: Ps 36:7; Isa 42:15. 155  This triad stands out in 24:3, 6, 8, 10, 12; Ps 104:10–12; and, e.g., 2 Kgs 19:23–24 // Isa 37:24–25; Ezek 17:3–5; 47:1–2, 8–9, 12; see Polak, “Water, Rock and Wood” (see n. 153), 20–29, 34–38. 156  For the dyad water/wood tree, see, e.g., Ps 1:3; Gen 2:5–10; Isa 41:18–19; Jer 17:7–8; Isa 1:29–30. 157  The dyad wood/rock is also found in Judg 9:7–13 (note the liquids wine and oil); 9:48. In addition one notes the well-known pair ‫אבן‬-‫עץ‬, e.g., 2 Kgs 19:18 (// Isa 37:19); Jer 2:27; Hab 2:11, 19.

474

Frank H. Polak

promise to the patriarchs. Thus the ESN and the Abraham-Jacob narrative form two panels of a diptych, hinging on the narrative of Mosesʼs birth, flight, and call to prophetic leadership (Exod 2–4). Ronald Hendel has pointed to a large number of common features, including not only the birth theme, the flight, the scene at the well, and the marriage with the hostʼs daughter but also the role as shepherd in the hostʼs service, the meeting with the brother, and the dangerous encounter with the deity.158 These shared themes go far beyond the commonalities of the traditional biographical pattern. The Hoseanic diatribe confirms the connection between the two tales,159 which is of particular importance in view of the correspondence of the shepherd-marriage connection alluded to in the description of Jacobʼs acts (Hos 12:13) but also valid for Moses. The migration theme positions Jacobʼs flight in contrast to the prophetic leadership by which Israel was brought out of Egypt (v. 13).160 A highly notable point is the implied contrast between the revelation to Jacob at the Jabbok and the divine call to Moses (v. 5). The contrast is heightened by the reference to the divine name, “Yet YHWH, the God of Hosts, his name is YHWH” (v. 6).161 The use of the term ‫זִ ְכרֹו‬, “his name,” is extremely meaningful in light of the fact that the entire proclamation “his name is YHWH” is redundant after the mention of the deity as subject. In an intertextual vista, it is difficult to separate this declaration from the proclamation of the divine name in Mosesʼs call narrative (Exod 3:15).

 R. Hendel, The Epic of the Patriarch: The Jacob Cycle and the Narrative Traditions of Canaan and Israel (Atlanta: Scholars Press, 1987), 137–151. Hendel (163–165) places the parallels in the oral tradition but does not consider their function in the unfolding of a largescale narrative platform. Not all shared features are explained by the traditional theme of the fugitive hero (like Idrimi of Alalakh), for which see E. L. Greenstein, “Biblical Interpretation in the Light of Its Ancient Cultural Context,” Studies in Jewish Education 9 (2004), 61–73. The thesis that the Jacob stories have influenced the Moses tales is defended by A. de Pury, “Situer le cycle de Jacob: Quelques réflexions, vingt-cinq ans plus tard,” in Studies in the Book of Genesis: Literature, Redaction and History (ed. A. Wénin; BETL 155; Leuven: Leuven University Press, 2001), 213–241, here 235–237; see also D. M. Carr, “Genesis in Relation to the Moses Story: Diachronic and Synchronic Perspectives,” in Studies in the Book of Genesis: Literature, Redaction and History (ed. A. Wénin; BETL 155; Leuven: Leuven University Press, 2001), 273–295, here, 282, n. 33. 159  See E. B lum , “Hosea 12 und die Pentateuchüberlieferungen,” in Die Erzväter in der biblischen Tradition: Festschrift für Matthias Köckert (ed. A. C. Hagedorn and H. Pfeiffer; BZAW 400; Berlin: de Gruyter, 2009), 291–321, here 310–318, with further references. 160  The reference to the messenger and the encounter at Bethel in 12:5 suggest a subtext regarding the divine guidance promised at Beth-el (Gen 28:13–15). The radical depreciation of Jacob is also implied by the false balances (‫)מֹאזְ נֵ י ִמ ְר ָמה‬, alluding to the ‫מ ְר ָמה‬/‫עקב‬ ִ theme in the Jacob narrative; one also notes the game with the double meaning of ‫ און‬in vv. 4, 9. 161  The divine praise in 12:6 contrasts the divine overlord with the messenger of v. 5 (like the double contrast in vv. 7, 8–9, 10) and is not to be viewed as secondary. 158

Storytelling and Redaction

475

Thus I concur with Konrad Schmid and Jan Gertz in the recognition of the relative independence of the patriarchal narrative and the ESN.162 Both narratives are panels in themselves and are not to be viewed as two parts of a straightforward, continuous narrative. But this independence is only a relative one. Within the oral-epic platform, the ESN and the Abraham-Jacob narrative constitute a diptych, with the Moses tale as hinge. If this platform had a continuation in the narrative of the trek and the conquest, it has largely disappeared,163 for the present version of this narrative, in Numbers, Deuteronomy and Joshua, is dominated by the intricate style of the Judean corpus.

162  G ertz , Tradition und Redaktion (see n. 37), 389–393; K. S chmid , Genesis and the Moses Story: Israelʼs Dual Origins in the Hebrew Bible (trans. J. D. Nogalski; Winona Lake, IN: Eisenbrauns, 2010), 156–196, 224–279. 163  The VoLB is limited to Num 10:29–12:8; 21–24; Josh 2; with residues in Josh 10. One notes some remains of poetic sections, for instance in Deut 34:7b.

Part Four The Significance of Second Temple Literature and the Dead Sea Scrolls for the Formation of the Pentateuch

Introduction Bernard M. Levinson Any complete account of the Pentateuch must recognize that it has a rich and vivid double life. One dimension of its life is the story that we as scholars tell of its compositional history and the way it came together from its literary sources and traditions (whether we employ a Neo-Documentary model of four literary sources that existed separately until they were compiled together in a single editorial moment or, alternatively, a supplementary model, in which there was a process of literary growth whereby one or more sources interacted with and commented upon other sources and were integrated in a series of redactional stages). The discipline remains divided about these alternative approaches to tracing the process of formation of the Pentateuch, and the various alternatives are well represented in this volume. The compositional history of the Pentateuch, however, provides only one dimension of its larger story. Equally important is the afterlife of the Torah: the history of its reception, interpretation, reworking, and expansion during the Second Temple period, both biblically and in the Dead Sea Scrolls. There is no doubt that this reception history of the Pentateuch, as well as its various translations and rewritings, tell us a great deal about how this corpus was treated and viewed in the immediate centuries after which it was formulated and accepted in the various communities of Second Temple Judaism. But the question arises whether the interpretation, exegetical reworking, and rewriting of the Pentateuch in late biblical and postbiblical literature also shed light on the process of its composition, compilation, or redaction.1 Does the evidence provided by the textual versions, the Dead Sea Scrolls, the Apocrypha, and the Pseudepigrapha help us to understand the birth story of the Pentateuch? In many ways, this part raises a series of specific scholarly questions about methodology, literary analogy, and the nature of evidence. In so doing, it intersects with the kinds of questions also being examined in part 1, “Empirical Perspectives on the Composition of the Pentateuch.” In addition, however, I 1  See further B. M. Levinson, A More Perfect Torah: At the Intersection of Philology and Hermeneutics in Deuteronomy and the Temple Scroll (Critical Studies in the Hebrew Bible 1; Winona Lake, IN: Eisenbrauns, 2013), 92–93; and see M. Bartos and B. M. Levinson, “ ‘This Is the Manner of the Remission’: Implicit Legal Exegesis in 11QMelchizedek as a Response to the Formation of the Torah,” JBL 132 (2013), 351–371.

480

Bernard M. Levinson

believe that we are also implicitly addressing a gap in the intellectual history of our discipline, the nature of the subfields within it, the way we are trained as scholars, and the tools we provide to our graduate students. It is increasingly recognized that the final stages of the formation of the Pentateuch took place in postexilic Judah during the Persian and Hellenistic periods. Yet standard biblical studies takes insufficient notice of the important evidence available within academic Jewish Studies for the literature and history of the Second Temple period, the emergence of textual collections, the relation between sectarian conflict and scriptural formation, or exegesis as a vehicle for the construction of community boundaries. This part investigates such issues in two ways. The first three essays focus on the manuscript 4Q158, an example of the genre of Reworked Scripture from the Dead Sea Scrolls, dating from the middle of the first century BCE. Because of the methodological questions that this manuscript raises, the volume editors thought that our biblicist colleagues might welcome the opportunity to be given a guided tour through this text by three experts. The larger intellectual goal is to use this text, which is manageable in size, as a case study to investigate whether it can shed light on the scribal and editorial processes that gave rise to the Pentateuch and complicate or enrich the models that are currently employed within pentateuchal theory. Does the genre of Rewritten Scripture, in other words, provide meaningful information about the formation of Scripture? The essays here place 4Q158 in its proper historical and literary context while also reflecting on what this manuscript can tell us about the shape of the Pentateuch at the time of its composition. These three papers have a sequential logic. Sidnie White Crawfordʼs introductory essay, “What Constitutes a Scriptural Text? The History of Scholarship on Qumran Manuscript 4Q158,” discusses the problematic publication history of 4Q158 and its impact on how scholars came to read or misread this text. Molly Zahn then investigates how the scribal practices reflected in the Second Temple period complicate many of the methodological assumptions of pentateuchal theory. Her essay is entitled “Scribal Revision and the Composition of the Pentateuch: Methodological Questions Raised by 4Q158 Fragments 1–2.” Reinhard Kratz, in “Reworked Pentateuch and Pentateuchal Theory,” shows how the reception and reworking of the biblical text evident in 4Q158 can illuminate earlier stages in the development of the biblical text. The next two essays continue the same line of investigation but move intentionally “backward” from postbiblical to biblical material of the Second Temple period: from Wadi Qumran to Jerusalem and Yehud. The logic of this sequence is to challenge some of the standard dichotomies that exist between biblical and nonbiblical, or canonical and noncanonical, and to look at the Second Temple material in a more comprehensive and nuanced way. Both essays have for their focus the reception and reinterpretation of pentateuchal material in Ezra and Nehemiah. Richard J. Bautch, in “Holy Seed: Ezraʼs Rhetoric and the Formation of the Pentateuch,” argues that the Ezra narrative provides a

Introduction

481

window into the compositional history of the Pentateuch as it was still taking shape, before the P and non-P material were combined. He places his focus on the origin and significance of the “holy seed” language and the ways it was applied to Israel. Coming to different conclusions, Sara Japhet concludes this part with her valuable study, “What May Be Learned from Ezra-Nehmiah about the Composition of the Pentateuch?” She argues that, based on the evidence of Ezra–Nehemiah, the Pentateuch must already have been a complete document in the Persian period and cannot be dated to the Hellenistic period. This kind of dialogue about assumptions and methodology is precisely what our volume seeks to encourage. By studying the canonical material and the Reworked Pentateuch material from the Dead Sea Scrolls together, this part demonstrates that the reception and reinterpretation of the Pentateuch in the Second Temple period opens a valuable window into the compositional history of the Pentateuch. This comparative approach helps illumine both the editorial techniques and the hermeneutical stakes that played such a significant role in the formation of Torah. Greater dialogue between the two subfields of the discipline of academic biblical studies can only enhance the methodology and theoretical models employed by each.

What Constitutes a Scriptural Text? The History of Scholarship on Qumran Manuscript 4Q158 Sidnie White Crawford The purpose of this brief article is to present an overview of scholarship on the Qumran manuscript 4Q158 as an illustration of how radically the understanding of what constitutes a scriptural text has changed since 4Q158 was first published in 1968. At the end of the article I will offer a suggestion concerning the purpose of 4Q158 and how it exemplifies the use of Scripture by learned scribes in the late Second Temple period. 4Q158 was originally assigned to the lot of John Allegro, indicating that the cave 4 editorial team did not consider it a “biblical” manuscript (otherwise it would have been assigned to Frank Moore Cross or Patrick Skehan).1 The title given to it indicates as much: “Biblical Paraphrase: Genesis, Exodus.” Allegro published it, along with the rest of his manuscripts, in Discoveries in the Judaean Desert (DJD) V in 1968.2 Allegroʼs stated goal was to make his lot of manuscripts publicly available as quickly as possible,3 but the result was an edition that is almost useless for scholars. Some examples will suffice to illustrate the problems. The photograph of the manuscript in DJD V is of the original plate taken in the Palestine Archaeological Museum (PAM) series (Allegro evidently did not request further photographs, although others were made).4 The fragments are arranged as they were placed on the original glass plate; no attempt is made to place them in order or in relation to each other. The reader is left with no sense of the dimensions of the manuscript. In the edition itself, no information at all is given of the physical properties of the manuscripts.5 A cursory glance at the plate reveals that frags. 1, 4, and 7 have the remnants of previous columns on the right; Allegro does not mention this. Fragments 5, 13, and 14 have left margins, and frag. 14 has a remnant of a second column on the left; all of this

 W. F. F ields , The Dead Sea Scrolls: A Full History (2 vols.; Leiden: Brill, 2009), 1:231–239. 2  J. M. Allegro, Qumrân Cave 4.I (4Q158–4Q186) (DJD V; Oxford: Clarendon, 1968). 3  Allegro, Qumrân Cave 4.I (see n. 2), preface. 4  Allegro, Qumrân Cave 4.I (see n. 2), plate 1. 5  Allegro, Qumrân Cave 4.I (see n. 2), 1. 1

484

Sidnie White Crawford

goes unacknowledged. There is no organized study of the orthography of the manuscript or its paleography. Most glaringly, the textual study of 4Q158 is incomplete. The text of the manuscript is only compared to the Masoretic Text (MT), which leads to a complete misunderstanding of the nature of the text. No attempt is made to explain the textual sequence or what the nature or purpose of the manuscript was. The result of this faulty publication was that the importance of 4Q158 for the history of the text of the Pentateuch and its use in the late Second Temple period escaped scholarly notice. In 1970 John Strugnell published a review article of DJD V in Revue de Qumran, “Notes en marge du volume V des ‘Discoveries in the Judaean Desert of Jordan.’ ”6 This article became indispensable to scrolls scholars attempting to work with DJD V.7 Strugnell gives the manuscript a paleographic date of the mid-first century BCE. He gives the PAM photograph numbers. He does not reproduce the text of all the fragments, but he does give a reconstructed text of frag. 6, a most important fragment, as well as a new transcription of frag. 14 with additional material not found in DJD V. Crucially, he compares the text of 4Q158 not only with the MT but also with the Septuagint (LXX) and the Samaritan Pentateuch (SP) as well as the daughter versions of MT and LXX. He thus recognizes numerous small readings in which 4Q158 agrees with the LXX and, critically, that the text of frags. 6 and 7–8 reflect (but do not reproduce exactly) the Samaritan text of Exod 20:19–21b. He also notes the times in which 4Q158 introduces changes not reflected in any of the versions, changes that he calls “exegetical.”8 Strugnellʼs article points to the textual complexity of 4Q158, but it is by no means a complete study. Since his article appeared in a journal dedicated to Qumran studies and was written in French, the importance of 4Q158 for the textual history of the Pentateuch again went unnoticed. 4Q158 would continue to languish in obscurity for another twenty years. In the late 1980s Emanuel Tov and the present author began work on an edition of a group of manuscripts originally assigned to John Strugnell, 4Q364–367. These manuscripts were first entitled Pentateuchal Paraphrase and had originally been assigned to Frank Moore Cross as a copy of the books of the Pentateuch. Cross, however, rejected them as insufficiently “biblical” and turned them over to Strugnell.9 Thus, at the beginning of their scholarly history these four manuscripts, which were eventually retitled 4QReworked Pentateuch, were, like 4Q158, classified as “nonbiblical” and placed on the far side of the clear dividing line that was thought to separate “biblical” from “nonbiblical” manuscripts. 6  J. Strugnell, “Notes en marge du volume V des ‘Discoveries in the Judaean Desert of Jordan,’ ” RQ 7/2 (1970), 163–276. 7  All of the following statements refer to Strugnell, “Notes en marge” (see n. 6), 168–175. 8  Strugnell, “Notes en marge” (see n. 6), 168. 9  Fields, The Dead Sea Scrolls (see n. 1), 232.

What Constitutes a Scriptural Text?

485

In 1991 Tov, in a preliminary article on 4Q364–367, tentatively identified 4Q158 as a fifth “copy” of the “composition” reflected in these manuscripts.10 Note the quotation marks placed around “copy” and “composition”; our understanding of the nature of these manuscripts and their relationship to one another has changed drastically over the years, but this was our understanding twenty-five years ago. Tov gave four reasons for identifying 4Q158 with 4Q364–367: (1) 4Q158 contains running biblical text interlaced with a few exegetical comments. (2) The biblical texts are presented in a sequence that sometimes differs from the other witnesses. (3) Sometimes material is omitted from the known biblical text. (4) 4Q158, like 4Q364, reflects the text of the SP in its major harmonizing characteristics. Further, 4Q158 and 4Q364 agree in three textual details, one of which is unique to those two manuscripts (4Q364, frag. 5ii:13 // 4Q158 frag. 1:7), and 4Q158 and 4Q366 overlap once, agreeing in important details with LXX and SP (4Q366 frag. 1:10 // 4Q158 frags. 10–12:7).11 In the editio princeps of 4Q364–367 (DJD XIII), Tov states, “the first allegiance of 4Q366 is with 4Q158.”12 These reasons were, and continue to be, compelling enough to posit a relationship of some kind among these five manuscripts. It was what that relationship was that was initially misunderstood. In fact, placing the five manuscripts in a group under the rubric Reworked Pentateuch proved to be misleading, as it implied a similarity of function that was incorrect. It would take a further two decades to arrive at a proper understanding of the nature of 4Q158 and the other four manuscripts placed under the umbrella Reworked Pentateuch. Michael Segal reopened the discussion concerning the nature and function of 4Q158 and the other four manuscripts in the late 1990s in a series of articles concerned with all five manuscripts. His 1998 article, “Biblical Exegesis in 4Q158: Techniques and Genre,”13 offered for the first time a close analysis of the exegetical techniques used in 4Q158. One should note the thirty-year gap between 4Q158ʼs initial publication in 1968 and the appearance of a thorough analysis devoted exclusively to it. Segal categorized the exegetical techniques in 4Q158 as thematic association, exegetical citations, paraphrase, rearrangement, harmonistic additions, and literary expansions.14 As a result of his analysis he drew several conclusions. First, he recognized the relationship between the text 10  E. Tov, “The Textual Status of 4Q364–367 (4QPP),” in The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid, 18–21 March 1991 (ed. J. Trebolle Barrera and L. Vegas Montaner; 2 vols.; Leiden: Brill, 1992), 1:43–82, here 46–48. 11  Tov, “Textual Status” (see n. 10), 47–48. 12  E. Tov and S. A. White, “Reworked Pentateuch,” in Qumran Cave 4.VIII: Parabiblical Texts, Part 1 (ed. H. Attridge et al.; DJD XIII; Oxford: Clarendon, 1994), 187–352, here 190. 13  M. S egal , “Biblical Exegesis in 4Q158: Techniques and Genre,” Textus 19 (1998), 45–62. 14  Segal, “Biblical Exegesis” (see n. 13), 47–61.

486

Sidnie White Crawford

of some of the fragments and the Samaritan Pentateuch, identifying 4Q158 as an example of a manuscript with pre-Samaritan affinity. Second, Segal was the first to suggest that 4Q158 is in fact an Exodus manuscript with interpolations from Genesis and Deuteronomy. Third, he noted that exegesis is an integral part of the text, “perhaps the reason for its creation.”15 Fourth, he also observed that the scribe of 4Q158 intervened in his base text, adding material beyond his presumed pentateuchal source. Thus, Segal states, “this composition therefore represents the boundary between biblical and parabiblical compositions, between text and explicit exegesis.”16 In this article Segal concluded that Reworked Pentateuch is an accurate name for the manuscript.17 In a related publication, “4QReworked Pentateuch or 4QPentateuch?,” he tackled the question of the relationship of the five manuscripts to one another, arguing that this “group” of manuscripts should be deconstructed, with 4Q364 and 365 understood as Pentateuch manuscripts, 4Q367 as an excerpted Leviticus text, and 4Q366 and 4Q158 as different exegetical texts.18 In the same volume, Eugene Ulrich suggested for the first time that the Reworked Pentateuch manuscripts contained a variant literary edition of the Pentateuch and should simply be considered a “biblical” text.19 In the next few years, following and building on Segalʼs work, there were several treatments of the five Reworked Pentateuch manuscripts, including 4Q158. For example, in my own 2008 monograph, Rewriting Scripture in Second Temple Times, I devoted a portion of the chapter on Reworked Pentateuch to 4Q158. In discussing the Reworked Pentateuch manuscripts, I abandoned my 1994 position that they were “copies” of the same “composition.” Rather, I argued that the five manuscripts all exhibit scribal interventions for exegetical purposes in different degrees, with 4Q158 occupying a place along the spectrum of scribal intervention past either 4Q364 or 365.20 Emanuel Tov has also changed his view concerning Reworked Pentateuch, now considering them a group of Scripture texts like any other Scripture texts.21  Segal, “Biblical Exegesis” (see n. 13), 61.  Segal, “Biblical Exegesis” (see n. 13), 62. 17  Segal, “Biblical Exegesis” (see n. 13), 62. 18  M. Segal, “4QReworked Pentateuch or 4QPentateuch?,” in The Dead Sea Scrolls Fifty Years after Their Discovery (ed. L. Schiffman et al.; Jerusalem: Israel Exploration Society, 2000), 391–399. 19  E. Ulrich, “The Qumran Scrolls and the Biblical Text,” in The Dead Sea Scrolls Fifty Years after Their Discovery (ed. L. Schiffman et al.; Jerusalem: Israel Exploration Society, 2000), 51–59, here 57. 20  S. W. Crawford, Rewriting Scripture in Second Temple Times (SDSSRL; Grand Rapids, MI: Eerdmans), 52–56. 21  E. Tov, “From 4QReworked Pentateuch to 4QPentateuch (?),” in Authoritative Scriptures in Ancient Judaism (ed. M. Popović; JSJSup 141; Leiden: Brill, 2010), 73–92, here 81–82, 91–92. Tov refers to the “identity crisis” these manuscripts went through, from their first identification as “paraphrases” to their present status as Scripture texts. 15 16

What Constitutes a Scriptural Text?

487

The most recent full-length study of 4Q158 is that of Molly Zahn in her 2011 monograph Rethinking Rewritten Scripture.22 Zahn makes two preliminary observations before launching into her analysis: the first is that 4Q158ʼs Vorlage was a pre-Samaritan text (on which there has been agreement since Strugnellʼs article); the second is that 4Q158ʼs unique readings may be original to it, that is, the product of its scribe.23 Zahn divides the types of intervention into six broad categories, which are similar to Segalʼs: addition of new material, addition of material from elsewhere in the Pentateuch, omission, paraphrase, rearrangement, and minor alterations. She notes that 4Q158 appears to have a particular exegetical concern to build textual bridges between related pericopes and that it has a “distinctive profile” among the Reworked Pentateuch manuscripts.24 Finally, she affirms that the 4Q Reworked Pentateuch manuscripts, taken together, are a diverse lot and that their original placement under the same rubric (originally Pentateuchal Paraphrase) has obscured how different from one another they in fact are.25 The purpose of this long historical overview is to give an idea of how much our understanding of what a scriptural text is has changed over the last forty years as a result of the Qumran manuscripts. 4Q158 is a microcosm of that change. The original editor, Allegro, at first comparing it only with the text of MT, classified it as a paraphrase of Genesis and Exodus. Its second editor, Strugnell, took the step of comparing it with LXX and SP and recognizing its affinities with those texts. Thus 4Q158 became one of the manuscripts that confirmed the existence of Hebrew manuscripts of scriptural books, dated to the late Second Temple period and in the possession of a Jewish community, that were related to the texts of the LXX and SP rather than MT. This was a major breakthrough in the history of textual criticism. However, the text of 4Q158 differed enough from the texts of the major witnesses (especially in its addition of new material) that it was still not classified as a simple manuscript of Exodus (or Genesis–Exodus) but as a “reworked” manuscript. Tov puts the position this way: “in the 1990s I did not consider 4QRP a Bible text, the main reason being that in my textual Weltanschauung there was no room for Scripture texts that contained such major deviations from MT as those in 4QRP.”26 Thus it was not until the year 2000 (a mere sixteen years ago) that the manuscripts of 4Q Reworked Pentateuch, including 4Q158, began to be classified, first by

22  M. M. Z ahn , Rethinking Rewritten Scripture: Composition and Exegesis in the 4QReworked Pentateuch Manuscripts (STDJ 95; Leiden: Brill, 2011). Zahn is also responsible, with Moshe Bernstein, for the forthcoming new edition of 4Q158 in a reissue of DJD V. 23  Zahn, Rethinking Rewritten Scripture (see n. 22), 73. 24  Zahn, Rethinking Rewritten Scripture (see n. 22), 71. 25  Zahn, Rethinking Rewritten Scripture (see n. 22), 229. 26  Tov, “From 4QReworked Pentateuch to 4QPentateuch (?)” (see n. 21), 77.

488

Sidnie White Crawford

some and eventually by almost all scrolls scholars, as scriptural texts of one kind or another. However, that general classification may have done a disservice to 4Q158 (as well as 4Q366 and 367), since, as Segal (followed by Zahn) pointed out, the manuscripts of the 4Q Reworked Pentateuch group are not all alike.27 4Q364 and 365 present themselves, in their breadth of coverage, as manuscripts of the entire Pentateuch, and the classification of those two manuscripts as scriptural texts (that is, scrolls that preserved the running text of the five books of the Pentateuch) is justified. The other three, however, are small and poorly preserved, so it is difficult to say with any certainty what they contained when they were whole. Thus, we may ask the question, what was 4Q158? Was it an entire manuscript of Exodus, or something else? We are limited by the fact that we do not have the beginning or the end of the manuscript, so our conclusions must necessarily be tentative. But nonetheless a suggestion can be made. What does 4Q158 contain? It contains excerpts from Exodus, first of all, in a pre-Samaritan version. These excerpts are bound together by the theme of covenant and how the covenant given to the ancestors is fulfilled by the Sinai covenant. Thus the scribe (for I tend to agree with Zahn, while acknowledging the uncertainty, that the contents of 4Q158 are the work of its scribe) interpolated passages from Genesis pertinent to this theme as well as adding additional material of the same sort. This content brings to mind other manuscripts that collected scriptural passages around a theme. For example, 4QTestimonia and 4QFlorilegium both collect scriptural passages around a specific theme. In 4QTestimonia the theme is messianic expectations, and in 4QFlorilegium it is eschatology. Interestingly, 4QTestimonia quotes Exod 20:21 in the same pre-Samaritan version as 4Q158 (thus perhaps pointing to the same scribal milieu). Both Testimonia and Florilegium collect quotations from different books of Scripture, so their nature as anthologies was clear from the beginning. A more pertinent example is 4QDeutn, classified, as its name suggests, as a manuscript of Deuteronomy, for which I was the editor.28 On examination this manuscript proved not to be a manuscript of the entire book of Deuteronomy but a manuscript of excerpts of Deuteronomy created for study or liturgical purposes. It contains Deut 8:1–5, a passage used for grace after meals, and 5:1–6:1 (where it breaks off), a passage containing the Decalogue, in that order. Its text of the Decalogue is unique; it interpolates into the Sabbath commandment of Deuteronomy the Exodus reason for observance.29 This interpolation is most likely (again with the caveat of uncertainty) the work of the scribe of 4QDeutn. I would suggest that 4Q158 is a work of the same type. A learned scribe was collecting passages from Exodus  Segal, “4QReworked Pentateuch or 4QPentateuch?” (see n. 18), 393.  S. W. C rawford , “4QDeutn,” in Qumran Cave 4.IX: Deuteronomy, Joshua, Judges, Kings (ed. E. Ulrich et al.; DJD XIV; Oxford: Clarendon, 1995), 117–128. 29  For discussion, see Crawford, Rewriting Scripture (see n. 28), 30–32. 27

28

What Constitutes a Scriptural Text?

489

concerning covenant and emphasizing, by using techniques described by Segal and Zahn, their connection with the ancestral covenants. So here we actually see, in a manuscript, a scribe at work, using the texts of Scripture (by this period the classical literature of Israel) as a tool to make an exegetical statement. This manuscript may have been used for teaching or study, as we assume was the case for 4QTestimonia or 4QFlorilegium. Scrolls of entire biblical books, such as Genesis or Exodus, would have been long and unwieldy, and difficult to use when one wanted to “check a reference.” A reliance on memory in a teaching or study situation could have led to these small collections or anthologies, to gather together pertinent references on a particular topic of interest. 4Q158 may have had just such a function in the Qumran community. Thus 4Q158 gives us a glimpse of the work of learned scribes, both as transmitters of tradition and as teachers, in the Second Temple period.

Scribal Revision and the Composition of the Pentateuch Methodological Issues Molly M. Zahn This essay originated in a session of the 2014 Jerusalem conference that used one of the 4QReworked Pentateuch manuscripts, 4Q158, as a lens for exploring the possible relevance of these and related Second Temple texts to the question of the composition of the Pentateuch. I was asked to reflect on these issues in light of one particular fragment of 4Q158 that is particularly rich in thought-provoking features.1 In what follows, I will use these features to address two main questions. First, if a text like 4Q158 is deemed relevant to pentateuchal theory, what might this relevance look like? How would it change the ways we approach the traditional pentateuchal text?2 Second, what methodological challenges does this sample fragment raise for pentateuchal theory in general and in particular for the use of Second Temple texts as part of that endeavor? I do not want to dwell too much on the question whether 4Q158 and similar texts should be deemed relevant to the understanding of earlier, nondocumented stages in the composition of the Pentateuch. Broadly speaking, the textual remains discovered at Qumran constitute the corpus most historically and culturally proximate to the context or contexts in which the traditional text of the Pentateuch likely took shape.3 We should therefore expect there to be certain 1  I refer to a single fragment but in fact am discussing the joined fragments 1–2. Frag. 1 contains the bulk of the material in the joined fragment presented below, with frag. 2 preserving just a few letters at the end of lines 5–7. Besides the editio princeps of J. M. A llegro , Qumran Cave 4.I (4Q158–4Q186) (DJD 5; Oxford: Clarendon, 1968), see now the new color image available online as part of the Leon Levy Dead Sea Scrolls Digital Library, http://www. deadseascrolls.org.il/explore-the-archive/image/B-298174 (accessed 2/20/15). 2  By traditional text I mean essentially the Masoretic Text, as the complete text that generally forms the starting point for compositional analyses of the Pentateuch. For the potential difficulties with this reliance on MT, see below and M. M. Zahn, “ ‘Editing’ and the Composition of Scripture: The Significance of the Qumran Evidence,” HBAI 3 (2014), 298–316, here 315. 3  The ancient Near East, of course, constitutes another rich source of comparative material, some of which (i.e., texts dating to the Neo-Assyrian and Neo-Babylonian periods) is likely contemporaneous with some biblical materials. On scribal practices in ancient Near Eastern texts and their potential relevance to biblical studies, see the seminal volume of J. T igay (ed.), Empirical Models for Biblical Criticism (Philadelphia: University of Pennsylvania

492

Molly M. Zahn

continuities in the ways scribes approached their exemplars. Initial explorations support this: the work of Michael Fishbane, David Carr, Reinhard Kratz, and others has highlighted similarities of scribal approach and exegetical technique between Second Temple materials and the text as it has come down to us in the MT.4 Thus I begin from the partially evidence-based assumption that the tools scribes used to reshape existing texts and compose new ones would have been substantially the same in the late Second Temple period as they were in earlier times. If this is our starting point, 4Q158 can become a resource for illustrating the kinds of things scribes might do with an earlier text. Insofar as we believe that pentateuchal texts draw on earlier materials, we might expect them to rewrite and reconfigure their sources in ways comparable to those attested in 4Q158 and other Second Temple materials. Thus, a survey of a few of the notable cases of rewriting attested in 4Q158 1–2 is one way of considering its relevance to pentateuchal materials. Of course, this single fragment does not at all represent the full diversity of ways texts were rewritten in Second Temple contexts; this survey is meant only to be suggestive.

1 Rewriting in 4Q158 1–2 As this transcription indicates, lines 3–13 of 4Q158 1–2 present material from Gen 32:25–33, Jacobʼs wrestling match with the divine figure at the Wadi Jabbok.5 The extant text here corresponds fairly closely to what we find in known versions of Genesis until lines 7–10, which contain previously unknown material (underlined above). The new material comes after the notice ‫ויברך אותו שם‬, Press, 1985); also H. J. Tertel, Text and Transmission: An Empirical Model for the Literary Development of Old Testament Narratives (BZAW 221; Berlin: de Gruyter, 1994); and more recently D. M. Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (Oxford: Oxford University Press, 2005), and idem, The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011), 37–101; J. Pakkala , Godʼs Word Omitted: Omissions in the Transmission of the Hebrew Bible (FRLANT 251; Göttingen: Vandenhoeck & Ruprecht, 2013), 54–63. 4  Carr, Formation (see n. 3), 48–101; M. Fishbane, “Use, Authority and Interpretation of Mikra at Qumran,” in Mikra: Text, Translation, Reading and Interpretation of the Hebrew Bible in Ancient Judaism and Early Christianity (ed. M. J. Mulder; CRINT 2.1; Assen: Van Gorcum/Philadelphia: Fortress, 1988), 339–377; R. G. Kratz, Das Judentum im Zeitalter des Zweiten Tempels (FAT 42; Tübingen: Mohr Siebeck, 2004), 126–156; Pakkala, Godʼs Word Omitted (see n. 3), 93–102, 155–181; C. Berner, “The Redaction History of the Sinai Pericope (Exod 19–24) and Its Continuation in 4Q158,” DSD 20 (2013), 378–409; D. A. Teeter, “The Hebrew Bible and/as Second Temple Literature: Methodological Reflections,” DSD 20 (2013), 349–377. 5  For a fuller analysis of this fragment and its relationship to existing scripture, see Zahn, Rethinking Rewritten Scripture (see n. 5), 25–74.

Scribal Revision and the Composition of the Pentateuch

493

4Q158 1–2: Gen 32:25–33 + Exod 4:27–28 + new/unidentified material (underlined)6 ]    [‫ [      ]ץ למען‬1 [      ]ṣ so that[    ] ]   [‫ [  ]◦[    ]שרית ופ‬2 [  ]?[    ]šryt and p[   ] ‫ל]בדו שמה‬ ֯ ‫י]ע ֯ק[ב‬ ̇ ‫ ו]י֯ ו̇ [תר‬Gen 32:25[ 3 [Gen 32:25 And J]ac[ob ]re[mained a]lone ‫וירא כי לוא‬26 ‫ויאבק [איש עמו עד עלות השחר‬ there, and [a man] wrestled[ with him until ֯ the dawn came. 26And he saw that he could ‫יכול לו ויגע בכף ירכו ותקע כף‬ not prevail over him, so he struck him on the hollow of his hip socket, ‫אמ ֯ר‬ ֯ ֯‫ו̇ י‬27 ‫יעקו]ב בהאבקו עמו[ ו]י̇ אחזהו‬ ֯ ‫ ירך‬4 and Jaco]b[ʼs hip socket was wrenched out of joint] when he wrestled with him.[ And] he ‫אל[יו שלחני כי עלה השחר ויאמר לוא‬ seized him 27and said to [him, “Release me, ]‫אשלחכה כי אם ברכתני‬ for dawn has come!” And he said to him, “I will not release you unless you bless me] ‫ויאמר לו ̇מה שמכה[ ויאמר‬28 ̇‫ [ ???]◦ אלי‬5 [ ???] to me.” 28He said to him, “What is ‫ויאמר לוא יעקוב יאמר עוד‬29 ‫]לו [יעקוב‬ your name?”[ And he said ]to him, [“Jacob.” ֯ 29 ]‫שמכה כי אם ישראל כי שרית עם אלוהים‬ He said, “No longer shall you be called Jacob, but rather Israel, because you have striven with God] ‫וישאל י[ע]קוב‬30 ‫[ו]ע[ם ]אנשים ותוכל‬ 6 [and ]wi[th ]humans and you have prevailed.” ֯ 30 ‫[ו]י̇ אמ[ר הגי]ד נא לי מ[ה שמכה ויאמר למה‬ J[a]cob asked [and ]sai[d, “Tel]l me, wh[at ] ‫זה תשאל לשמי‬ is your name?” But he said, “Why do you ask my name?”] ‫ [ויבר]ך אותו שם ויאמר לו יפרכה‬7 [And he bless]ed him there, and he said to ]  [‫[וה וירב]כה‬ ̇ ‫יה‬ ̇ him, “May YH[WH] make you fruitful [and multiply ]you[  ] ]  [◦ ̇‫בינה ויצילכה מכול חמס ו‬ ̇ ̇‫[ד]עת ו‬ ̇ 8 [kno]wledge and understanding, and may he deliver you from all violence and ?[  ] ]   [‫עולם‬ ̇ ‫ עד היום הזה ועד דורות‬9 until this day and until eternal generations[   ]” ‫וי̇ [קרא יעקוב‬31 ‫ וילך לדרכו בברכו אותו שם‬10 And he went on his way when he had blessed ‫שם המקום פנואל כי ראיתי אלוהים פנים אל‬ him there. 31And [Jacob] ca[lled the name ]‫ויזרח‬32 ‫פנים ותנצל נפשי‬ of the place Penuʾel, because “I have seen God face to face, and yet my life has been saved.”] ‫ לו השמש כאשר עבר את פנוא[ל והואה צולע‬11 32And the sun [rose] upon him as he passed ]   ??? ‫על ירכו‬ by Penue[l, and he was limping because of his hip ???   ] ‫ ביום ההואה ויאמר אל תוא[כל(ו) את גיד‬12 on that day, and he said, “Do not ea[t the ‫על כן לוא‬33 ‫הנשה אשר על שתי כפות הירך‬ tendon of the sciatic nerve that is upon the two sockets of the hip.” 33Therefore the ]‫יואכלו בני ישראל את גיד הנשה אשר‬ Israelites do not eat the tendon of the sciatic nerve that is]  Transcription and translation follow M. M. Z ahn , Rethinking Rewritten Scripture: Composition and Exegesis in the 4QReworked Pentateuch Manuscripts (STDJ 95; Leiden: Brill, 2011), 246–247. 6

494

Molly M. Zahn

‫ על שתי כפות הירך עד ̇ה[יום‬13 upon the two sockets of the hip until ‫הזה כי נגע בכף ירך יעקוב בגיד‬ to[day, because he struck Jacob on the ]‫ויאמר יהוה‬Exod 4:27  ??? ‫הנשה‬ hip socket, on the tendon of the sciatic nerve ???  Exod 4:27YHWH spoke] ‫]מ[ושה‬ ֯ ‫לקרא[ת‬ ̇ ‫ אל אהרון לאמור לך‬14 to Aaron, saying “Go to mee[t ]M[oses in ‫המדברה וילך ויפגשהו בהר האלוהים וישק לו‬ the wilderness.” So he went, and he met him ]‫ויגד מושה לאהרון את כול‬28 on the mountain of God, and he kissed him. 28 And Moses related to Aaron all] ‫כול[ האותות אשר‬ ̇ ‫שלחו ואת‬ ֯ ‫ דברי יהוה אשר‬15 the words of YHWH with which he had sent ‫צוהו ??? ויאמר מושה אל אהרון (לאמור) זה‬ him and all[ the signs with which he had ]‫האות אשר צוה‬ commanded him. ??? And Moses said to Aaron, “This is the sign that] ‫את[ העם ממצרים‬ ̇ ‫ י̇ הוה לי לאמור בהוציאכה‬16 YHWH [commanded] me, saying ‘When you ] ??? ‫תעבדון את האלהים על ההר הזה‬ bring[ the people out from Egypt, you shall worship God upon this mountain ??? ] ] [◦‫והנה המה שלו̇ ש‬ ̇ ‫◦ב◦ים‬ ̇ ‫ללכת‬ ̇ 17 to go ?b?ym, and see, they are thirt[y(?) ] ] [‫אלוהים[ ]ה[ ]ל‬ ֯ ‫יהוה‬ ֯ 18 YHWH God[ ]h[ ]to[ ] ] [◦] [‫ נשל[ ]ל‬19 drew off(?) [ ]to[ ]?[ ]’ ”

“and he blessed him there,” which in the MT and other known versions closes 32:30. In the traditional version of the story, these words may simply indicate the stranger departing from Jacob, i.e., saying farewell. In any case, the contents of the blessing are not included; there is only the narrative report that the blessing occurred (“he blessed him there”). In contrast, the added material in 4Q158 ̇ ‫יפרכה‬, “May YH[WH] seems to provide the missing blessing. It starts with ‫יה[וה‬ make you fruitful,” and continues in line 8 with ‫ויצילכה מכול חמס‬, “and may he deliver you from all violence.” The last few added words specifically indicate the strangerʼs departure after the blessing, creating a nice repetitive resumption in the process: ‫וילך לדרכו בברכו אותו שם‬, “And he went on his way when he had blessed him there” (line 10). A very similar move occurs farther down in the fragment, in lines 15–16, where Moses is addressing Aaron. Here too, the MT version notes indirectly that Moses spoke to Aaron (‫ויגד משה לאהרן את כל דברי יהוה‬, “And Moses told Aaron all the words of YHWH,” Exod 4:28). 4Q158, on the other hand, seems to insert the contents of the speech in direct address, as implied by the words ]. . .[ ‫יהוה לי לאמור‬, “YHWH to me, saying [. . .],” at the beginning of line 16. Thus in two cases this fragment contains direct speech that, in other versions of the text, is only alluded to (e.g., “he blessed him”) but is not actually reported. Using 4Q158 as a model, we might ask whether there are cases of direct speech in the MT Pentateuch that might have the same kind of history, representing secondary expansions of texts where the speech was originally simply referred to rather than recounted. A third interesting move in 4Q158 also involves added direct speech, but here the issues appear to be different. The extant material in line 12 does not occur

Scribal Revision and the Composition of the Pentateuch

495

in known versions of Genesis 32. It comes between Jacobʼs hobbling departure from Penuel in 32:32 and the notice in the next verse that, because of Jacobʼs injury, the Israelites do not eat the tendon of the sciatic nerve. What is preserved of this addition reads . . .)‫ביום ההואה ויאמר אל תוא[כל(ו‬, “on that day. And he said: Do not ea[t . . . .” The rest of the contents of the addition are uncertain, but it would seem that the logical connection between Jacobʼs injury and the Israelitesʼ dietary practice is being strengthened by means of a command not to eat the sciatic nerve. The context does not allow us to determine conclusively who the speaker of this command is; perhaps Jacob is simply instructing his family. But it is tempting to argue that God somehow appears here as the speaker, addressing Jacob. This would give the force of divine commandment to an element of later rabbinic halakah that is depicted in the traditional version of Genesis as merely a customary practice.7 Either way, this addition witnesses a concern with the origins and authority of Israelʼs religious and cultural practices. There may not be many other exactly analogous cases in known versions of the Pentateuch, where we might argue that what was originally depicted as a traditional practice has been given the force of a command.8 But there is certainly much reflection in the Pentateuch on the origins and authority of law. If we were to use 4Q158 as a model, we might look for evidence of additions or alterations meant to clarify or demonstrate legal authority or for connections between law and the events of Israelʼs past.9 A final noteworthy aspect of this fragment is the odd juxtaposition of Gen 32:25–33 in lines 3–13 and Exod 4:27–28 in lines 13–15. As it stands, the connection between these two pericopes is not entirely clear. Students of this text have long suggested that the real connection is between the Jacob story and the

 In fact, the practice is referred to as a divine command in Song of Songs Rabbah 1.2.5.  There is certainly a case in Jub. 28:6, where what Laban describes as the local custom of marrying off an older daughter before a younger one can be married (see Gen 29:26) is described as a command “ordained and written on the heavenly tablets.” 9  One very basic example of the latter, recognizable because its literary source is preserved, can be found in Deuteronomyʼs law of manumission (Deut 15:12–18). Unlike the earlier version in the Covenant Code (Exod 21:2–6), Deuteronomyʼs law is explicitly grounded in the Israelitesʼ experience of divine manumission from slavery in Egypt: ‫וזכרת כי עבד היית‬ ‫בארץ מצרים ויפדך יהוה אלהיך על כן אנכי מצוך את הדבר הזה היום‬, “Remember that you were a slave in the land of Egypt and YHWH your God redeemed you; therefore I am giving you this command today” (Deut 15:15). Might a reverse case (insertion of law into narrative contexts) perhaps be preserved in Gen 9:4, where God qualifies his permission to Noah to eat animals as food with the command that their blood is not to be eaten (compare Lev 17:10–13)? Regarding redactional moves meant to demonstrate legal authority, see Morton Smithʼs observations on how originally anonymous laws in the P sections of the Pentateuch (e.g., Lev 13–15) were transformed into divine speech by means of an “editorial framework”; M. Smith, “Pseudepigraphy in the Israelite Literary Tradition,” in Pseudepigrapha I (ed. K. von Fritz; Entretiens sur lʼAntiquité classique 18; Geneva: Fondation Hardt, 1972), 191–215, here 201–202. 7 8

496

Molly M. Zahn

immediately preceding section of Exod 4, namely vv. 24–26, the so-called bridegroom-of-blood episode where God “seeks to kill” Moses.10 These two episodes are connected by the theme of apparent divine aggression toward the hero in the context of return to a homeland and encounter with a brother: Jacob encounters Esau on his way back to Canaan, and Moses encounters Aaron on his way back to Egypt. This proposal presumes that there must have been some mention of the bridegroom-of-blood episode in the lost portions of the manuscript surrounding frags. 1–2, though of course there is no textual evidence for this. If this explanation is correct, we appear to have a rearrangement or repetition such that materials that are thematically related are brought into proximity with one another, even though in this case the two episodes are separated by some four hundred years in the chronological framework of the Pentateuch. If we assume that 4Q158 generally adheres to this same temporal framework (and we have no evidence that it did not), then we must presume some kind of narrative device that addresses this chronological problem. Otherwise, how can Jacob and his distant descendant Moses appear at the same point in the story? A flashback is probably the most likely candidate here: somehow, in the context of the Exodus narrative, Jacobʼs story is introduced as a parallel. Of course, this suggestion, which is based only on what narrative logic would seem to demand, is entirely hypothetical. The main point in the context of my investigation here, though, is that the scribe responsible for this rearrangement has joined two different episodes based on thematic – rather than formal or stylistic – parallels.11 Again, it may or may not be possible to identify precisely analogous cases in the MT Pentateuch. It is difficult to think of many instances in which two episodes with thematic connections but unrelated formal content occur side by side.12 But perhaps an exact analogue is not the only thing we should be looking for. This is just one – and a rather idiosyncratic one, at that – of many  For bibliography, see Zahn, Rethinking Rewritten Scripture (see n. 5), 55.  For more on this technique in 4Q158, see M. M. Z ahn , “Building Textual Bridges: Towards an Understanding of 4Q158 (4QReworked Pentateuch A),” in The Mermaid and the Partridge: Essays from the Copenhagen Conference on Revising Texts from Cave Four (ed. G. J. Brooke and J. Høgenhaven; STDJ 96; Leiden: Brill, 2011), 13–32. The technique in a way constitutes the inverse of the rabbinic interpretive technique of smichut parshiyot, according to which the situation of two pericopes next to one another is read as exegetically significant, even if their contents appear unrelated. That is, there is assumed to be a thematic connection between two adjoining passages. (I thank Bernard Levinson for suggesting this analogy.) Here in 4Q158, instead of textual proximity generating a thematic link, the thematic link serves as the basis for a rearrangement that creates textual proximity. 12  Perhaps one example might be Num 16, where the narrative of the Korahite rebellion against Aaronʼs priestly authority has been conflated with a story in which Dathan and Abiram challenge Mosesʼs authority. Here, of course, the two thematically similar narratives have been woven together rather than simply juxtaposed; see B. A. Levine, Numbers 1–20 (AB 4; New York: Doubleday, 1993), 405–406. Note also the insightful analysis by Benjamin Sommer of Num 11 as a combination of two narratives, unrelated by plot, that address a single theme (the 10 11

Scribal Revision and the Composition of the Pentateuch

497

documented instances in which a scribe has increased the connections between parallel scriptural passages. These range from changing a single word to match the wording of a parallel passage to insertion of whole new sections meant to coordinate parallel accounts, as in the Samaritan Pentateuchʼs treatment of the wilderness narratives in Numbers.13 Carr, in particular, has already argued for the likely presence of additions serving this function within the MT itself.14 This somewhat quirky example from 4Q158 might encourage us to think a bit more creatively about what such coordinating additions or rearrangements might have looked like. They need not have been restricted to cases of formal parallels but might have involved looser or more thematic correspondences.

2 Methodological Implications With these examples in mind, I want to turn now to some more explicitly methodological considerations, beginning with a brief return to the issue of relevance or analogy. One cannot help but notice the highly exegetical nature of all of the major scribal interventions discussed above. Gaps are filled through the addition of missing speech; a customary practice is rooted in a command; two thematically related sections are brought together. These are all interpretive responses to perceived shortcomings in the source text. Thus, one might ask: Do not all of these changes belong to a late stage in the transmission of the Pentateuch, one in which scriptural texts have reached such a level of authority and sanctity that they are subjected to increasing levels of exegetical attention? And if so, are they not likely to be different from the kinds of activities engaged in by scribes working at earlier stages? There are several issues here. On the one hand, these highly exegetical moves in 4Q158 cannot represent the full range of ways that earlier scribes would have worked with their materials. But neither do they represent the full range of late Second Temple scribal activity. Much of this activity is considerably less exegetical in nature, if by exegetical we mean changes that appear to respond directly to some feature of an earlier form of the text. So, rather than leading pentateuchal theorists to reject the relevance of Second Temple texts for earlier stages of composition, limits in the applicability of these particular examples status of Moses) in opposite ways; B. D. Sommer, “Reflecting on Moses: The Redaction of Numbers 11,” JBL 118 (1999), 601–624. 13  See, for example, E. Tov, “The Nature and Background of Harmonizations in Biblical Manuscripts,” JSOT 31 (1985), 3–29; idem , “Textual Harmonization in the Stories of the Patriarchs,” in Rewriting and Interpreting the Hebrew Bible: The Biblical Patriarchs in the Light of the Dead Sea Scrolls (ed. D. Dimant and R. G. Kratz; BZAW 439; Berlin: de Gruyter, 2013), 19–50; Carr, Formation (see n. 3), 90–98. 14  Carr, Formation (see n. 3), 104, 134–137.

498

Molly M. Zahn

should instead lead to a broader study of Second Temple textuality and the consideration of other possible analogues. On the other hand, it is too hasty to write off high levels of exegetical engagement, such as those attested in 4Q158, as appropriate only to later stages of transmission. One of the few points of consensus in recent pentateuchal theory has been the recognition of the role played by innerbiblical exegesis in the Pentateuchʼs formation. That is, it has become widely accepted that much of the scribal activity that led to the Pentateuch as we know it was hermeneutical in nature.15 4Q158 illustrates that the (presumably) relatively high level of textual authority necessary for a text to be subjected to exegetical revision or rewriting did not limit the amount of intervention or change that a scribe could introduce – this is a very exegetical text, and one that drastically reworks its sources.16 Therefore, it makes good sense to examine the Pentateuch in its traditional form for signs of exegetical moves similar to those attested in 4Q158 and related texts. Other aspects of our fragment, unfortunately, suggest more sobering methodological reflections. One has to do with the question of assigning different materials to different stages of composition. Scholars often distribute material to various sources or redactional layers on the basis of similar language or ideological Tendenz. Here in 4Q158 1–2, such clues are in short supply. We can clearly distinguish the new material from a textual substrate that closely resembled known versions of Genesis and Exodus – but this is only because those known versions are available for comparison. The various major changes in 4Q158 1–2 do not share distinctive linguistic or ideological features. There are, it is true, two cases of supplying absent direct speech, but the other alterations move in different directions. We cannot, therefore, assume that all these changes represent the work of a single reviser (i.e., a single “redactional layer”). Nevertheless, none of the changes contradict any of the others or differ drastically in a linguistic or stylistic sense, so it is possible that they were, in fact, all introduced by a single scribe. We simply lack the evidence to decide. In a way, it is precisely the exegetical nature of these changes – their rootedness in some problem of the text and lack of explicit Tendenz – that makes them difficult to attribute. For the purposes of pentateuchal theory, the point is that if similar types of changes did take place in the course of the transmission of the Pentateuch, it may prove very difficult to sort these changes into multiple redactional “layers.” That is, even in those cases where secondary materials can be isolated with some confidence, we may lack  See, e.g., B. M. L evinson , “The Right Chorale”: Studies in Biblical Law and Interpretation (FAT 54; Tübingen: Mohr Siebeck, 2008), 7–39, here 23; Kratz, Das Judentum (see n. 4). 16  See Emanuel Tovʼs observation that, though we might expect Second Temple scribes to have approached the Torah with “special care and a lower level of intervention” because of its sacred status, in fact the opposite is the case: “Paradoxically, due to its popularity, the Torah was rewritten and changed more extensively than the other biblical books in the Second Temple period”; Tov, “Textual Harmonization” (see n. 13), 19–20. 15

Scribal Revision and the Composition of the Pentateuch

499

the textual evidence necessary to decide whether these materials were inserted all at once or derive from successive redactors working at different times. Another issue is highlighted by the text of 4Q158 1–2, lines 3–6, as presented in the above transcription. The reconstruction here is based on MT, which is very similar to other known witnesses at this point. Unfortunately, however, this reconstruction cannot be correct, since it results in a text with uneven line lengths: lines 3 and 5 are much longer than lines 4 and 6. The impossibility of this reconstruction indicates that the text must have differed from other versions here, even though the extant text of these lines corresponds quite closely to MT. In other words, we have clear evidence for some kind of otherwise unattested text form, but we cannot determine what that form actually was. The unknown text form attested in 4Q158 raises the question of how pentateuchal theory handles textual unknowns. This example is just one of myriad cases of divergent readings in Second Temple biblical manuscripts and rewritten compositions, which in turn point to a fundamental situation of textual fluidity in this period.17 The texts of the compositions that ended up in our Bibles were not fixed but were subject to constant revision. As I have argued, we have no reason to think that such scribal updating was only a product of the late Second Temple period and would not have occurred in earlier contexts. Yet pentateuchal theory for the most part bases itself, explicitly or implicitly, on the text of MT. This is of course completely understandable: MT represents a relatively noninterventionist text of the Pentateuch. And besides the occasional more original reading preserved in the Septuagint, a Qumran manuscript, or some other ancient version, there is no other concrete evidence to go on.18 But when we work backward from MT to reconstruct earlier sources or redactional layers, we tend to arrange these in linear fashion: we excavate successive layers or tease out independent sources. Herein lies the methodological problem: in doing this, we act on the presumption that we have all the evidence at our disposal – that the entire history of the text is contained, like an archaeological tel, in the text as we now have it preserved in MT. If scribal practice in earlier stages was anything like it was later on, however, this presumption is misguided. The history of the pentateuchal text leading up to MT is likely to have been complex and messy, with multiple versions of the same text existing side by side and influencing one another. Sections may have been totally rewritten, or replaced with alternate versions from another source, in ways that leave no trace in the textual record.19 In other 17  See especially E. C. Ulrich, “The Text of the Hebrew Scriptures at the Time of Hillel and Jesus,” in Congress Volume; Basel, 2001 (ed. A. Lemaire; VTSup 92; Leiden: Brill, 2002), 85–108. 18  On the centrality of MT in biblical scholarship, see E. Tov , Textual Criticism of the Hebrew Bible (3rd ed.; Minneapolis: Fortress, 2012), 365. 19  See C arr , Formation (see n. 3), 111–117; and, with a focus on omissions, Pakkala , Godʼs Word Omitted (see n. 3), 11–91.

500

Molly M. Zahn

words, considerable portions of a textʼs compositional history may simply be unrecoverable – much as the true contents of lines 3–6 of this fragment are unrecoverable to us. It will have become clear by now that I view 4Q158 and other rewritten texts from Qumran as full of promise but also full of challenges if they are taken seriously as relevant to the enterprise of pentateuchal theory. These texts constitute a treasure trove of examples of scribal activity and thus can draw our attention toward a range of features in the traditional text of the Pentateuch that might reflect similar activities. Further research is required, of course, both to compile and to analyze the extant Second Temple materials and to determine the precise nature of their relevance to earlier stages of composition. Even with this new body of comparative material, though, pentateuchal theory will remain at the level of hypothesis: barring spectacular new manuscript discoveries, our theories cannot be confirmed but only judged as relatively more or less compelling based on the textual evidence itself. This is where Second Temple models present a challenge, or rather, highlight methodological limitations that were there all along. It is the fragmentary Qumran texts that in fact illustrate how fragmentary the ostensibly complete text of the Pentateuch is as a record of its own composition history. They point to the lacunae in our understanding, gaps that likely can never be confidently reconstructed. I understand that this is a problematic, perhaps even heretical, conclusion for a field that has stood at the center of academic biblical studies since its inception. And I do not at all mean to imply that nothing of substance can be said about the composition of the Pentateuch. Based on my own work with Second Temple manuscript materials, however, I must reiterate Carrʼs call for methodological modesty – for a greater willingness to recognize the constraints of our evidence as well as the possibility, even the likelihood, that we are missing critical pieces of information that would change our conclusions.20 If scholarship moves in this direction, we may not be able to say as much as we would like about the history and development of the pentateuchal text. But what we do say will be more firmly rooted in the extant evidence for the production and transmission of texts in early Judaism.

20

 Carr, Formation (see n. 3), 144–149.

Reworked Pentateuch and Pentateuchal Theory Reinhard G. Kratz 1 Introduction After more than 250 years of critical research, the pentateuchal question is not solved yet. The current discussion is lively but, with the exception of some basic points of consensus, remains highly controversial and fractured.1 A fresh approach is the view from outside, which takes the external evidence into account. There are two roads to be taken: one is to look for analogies in the ancient Near East (Gilgamesh Epic, Assyrian inscriptions) and in Second Temple literature (such as Jubilees, Temple Scroll, as well as Serekh Ha-Yaḥad, Damascus Document); the other is to start with the different forms of the Pentateuch in the manuscripts, versions (Septuagint [LXX], Samaritan Pentateuch [SP]), and excerpts of Rewritten Scripture (Jubilees, Temple Scroll, Genesis Apocryphon, etc.). Both roads have already been taken in scholarship and have led, as is to be expected, to different results.2 Jeffrey Tigay used ancient Near Eastern analogies in order to confirm the Documentary Hypothesis; Stephen Kaufman rejected, or rather relativized, the possibility of analyzing the Pentateuch and reconstructing its literary history on the basis of external analogies, taking the Temple Scroll as an example.3 David Carr and others have looked more broadly and found the evidence of analogies extremely complex, so much so that it would be (too) difThis paper was finished in the 2014–2015 academic year, which I spent as an Overseas Visiting Scholar at St Johnʼs College, Cambridge. I would like to express my deep gratitude to the college as well as to my host, Dr. Nathan McDonald, who invited me and provided me with an extraordinary working environment, both hospitable and stimulating, which enabled me to complete this piece. English translation by Ruth Ludewig-Welch (Göttingen). 1  See R. G. Kratz, “The Pentateuch in Current Research: Consensus and Debate,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 31–61, and the discussion documented in this volume. 2  See R. G. K ratz , “Innerbiblische Exegese und Redaktionsgeschichte im Lichte empirischer Evidenz,” in Das Judentum im Zeitalter des Zweiten Tempels (ed. R. G. Kratz; 2nd ed.; FAT 42; Tübingen: Mohr Siebeck, 2013), 126–156. 3  J. H. T igay , “An Empirical Basis for the Documentary Hypothesis,” JBL 94 (1975), 329–342; idem, The Evolution of the Gilgamesh Epic (Philadelphia: University of Pennsylvania Press, 1982; repr., Wauconda, IL: Bolchazy-Carducci, 2002); idem (ed.), Empirical Models

502

Reinhard G. Kratz

ficult to apply it to the analysis of the Hebrew Bible.4 More recent contributions, therefore, focus more and more on the textual history of the Pentateuch and other writings of the Hebrew Bible as a starting point for the analysis of a composition and the reconstruction of its literary and redactional history.5 In this paper I would like to follow this latter path6 and take the so-called Reworked Pentateuch (RP) manuscripts from Qumran (4Q158 and 4Q364–367) – especially 4Q158 – as an example. These manuscripts move between a copy and a rewriting of the Pentateuch and thus have the advantage that they can function both as a different version and as an analogy. In other words: although I am taking the second road of textual history, the specific texts I am dealing with blur the distinction between the two roads of analogy and version. Furthermore, I want to confront the evidence in the RP manuscripts with the scholarly analysis of the parallel Masoretic Text in order to compare the compositional processes going on in both versions. Before we go into detail, a few general remarks are in order.

2 General Remarks The edition and classification of the relevant material is still in process. Fragments of five manuscripts (4Q158 and 4Q364–367) are preserved.7 It is still an for Biblical Criticism (Philadelphia: University of Pennsylvania Press, 1985); S. A. Kaufman, “The Temple Scroll and Higher Criticism,” HUCA 53 (1982), 29–43. 4  D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011); H. J. T ertel , Text and Transmission: An Empirical Model for the Literary Development of Old Testament Narratives (BZAW 221; Berlin: de Gruyter, 1994). 5  See already J. W ellhausen , Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871); also T igay , “Empirical Basis” (see n. 3); idem , Empirical Models (see n. 3); Carr, Formation (see n. 4); recently, J. Pakkala, Godʼs Word Omitted: Omissions in the Transmission of the Hebrew Bible (FRLANT 251; Göttingen: Vandenhoeck & Ruprecht, 2013); D. A. T eeter , Scribal Laws: Exegetical Variation in the Textual Transmission of Biblical Law in the Late Second Temple Period (FAT 92; Tübingen: Mohr Siebeck, 2014); R. M üller et al., Evidence of Editing: Growth and Change of Texts in the Hebrew Bible (RBS 75; Atlanta: Society of Biblical Literature, 2014); U. B ecker and H. B ezzel (eds.), Rereading the relecture? The Question of (Post)chronistic Influence in the Latest Redactions of the Books of Samuel (FAT 2/66; Tübingen: Mohr Siebeck, 2014). 6  For the composition of the Temple Scroll as an analogy for the pentateuchal question, see R. G. K ratz , “Sources, Fragments, and Additions: Biblical Criticism and the Dead Sea Scrolls” (forthcoming) and the relevant literature mentioned there. 7  J. M. Allegro, Biblical Paraphrase: Genesis, Exodus, Qumrân Cave 4 I (4Q158–186) (DJD V; Oxford: Oxford University Press, 1968), 1–6 + pl. 1; E. Tov and S. W. Crawford, “Reworked Pentateuch,” in Qumran Cave 4 VIII, Parabiblical Texts, Part 1 (ed. H. Attridge et al.; DJD XIII; Oxford: Clarendon, 1994), 187–352 + pl. 13–26; the images are available in the Leon Levy Dead Sea Scrolls Digital Library of the IAA, Jerusalem (http://www.

Reworked Pentateuch and Pentateuchal Theory

503

open question whether the five manuscripts are part of the same or different copies of the Pentateuch.8 The question whether the five manuscripts cover the entire Pentateuch or only parts of it is also a matter of debate; at least 4Q364 and 4Q365 clearly represent the whole, containing fragments from all parts, though 4Q364 does not contain any fragments from Leviticus. The text form moves between the (Proto-)Masoretic and the Samaritan versions of the Pentateuch. It has a close affinity to the Proto-Samaritan text attested elsewhere in some manuscripts from Qumran but also has its peculiarities, which differ from both the Masoretic and the Samaritan versions. Therefore, it is much disputed whether the five manuscripts represent an edition of the “biblical” writing or another exemplar of Rewritten Scripture.9 Significant for this problem are, most notably, the contributions of Emanuel Tov, who in the course of time deadseascrolls.org.il). For an introduction to the material, see E. Tov, “4QReworked Pentateuch: A Synopsis of its Contents,” RevQ 16 (1995), 647–653, reprinted in idem, Hebrew Bible, Greek Bible and Qumran: Collected Essays (TSAJ 121; Tübingen: Mohr Siebeck, 2008), 21–26; S. W. C rawford , Rewriting Scripture in Second Temple Times (Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans 2008), 19–30 and 39–59. 8  See G. J. Brooke, “4Q158: Reworked Pentateucha or Reworked Pentateuch A?,” DSD 8 (2001), 219–241. 9  See E. Ulrich, “The Dead Sea Scrolls and the Biblical Text,” in The Dead Sea Scrolls after Fifty Years: A Comprehensive Assessment (ed. P. W. Flint and J. C. VanderKam; 2 vols.; Leiden: Brill, 1998), 1.79–100, here 88; idem, The Dead Sea Scrolls and the Origins of the Bible (Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans, 1999), 32, n. 51; idem, “The Qumran Scrolls and the Biblical Text,” in The Dead Sea Scrolls Fifty Years after Their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman et al.; Jerusalem: Israel Exploration Society, 2000), 51–59, here 57; idem, “The Qumran Biblical Scrolls: The Scriptures of Late Second Temple Judaism,” in The Dead Sea Scrolls in Their Historical Context (ed. T. H. Lim et al.; Edinburgh: T&T Clark, 2000), 67–87, here 76; M. Segal, “4QReworked Pentateuch or 4QPentateuch?,” in The Dead Sea Scrolls Fifty Years after Their Discovery: Proceedings of the Jerusalem Congress, July 20–25, 1997 (ed. L. H. Schiffman et al.; Jerusalem: Israel Exploration Society 2000), 391–399; idem, “Between Bible and Rewritten Bible,” in Biblical Interpretation at Qumran (ed. M. Henze; Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans 2005), 10–28; A. Lange, “The Status of the Biblical Texts in the Qumran Corpus and the Canonical Process,” in The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries (ed. E. D. Herbert and E. Tov; London: British Library, 2002), 21–30; M. M. Zahn, “The Problem of Characterizing the 4QReworked Pentateuch Manuscripts: Bible, Rewritten Bible, or None of the Above,” DSD 15 (2008), 315–339; eadem , “Building Textual Bridges: Towards an Understanding of 4Q158 (4QReworked Pentateuch A),” in The Mermaid and the Partridge: Essays from the Copenhagen Conference on Revising Texts from Cave Four (ed. G. J. Brooke and J. Høgenhaven; STDJ 96; Leiden: Brill, 2011), 13–32. For further discussion, see M. J. B ernstein , “ ‘Rewritten Bible’: A Generic Category Which Has Outlived Its Usefulness?,” Textus 22 (2005), 169–196; idem, “What Has Happened to the Laws? The Treatment of Legal Material in 4QReworked Pentateuch,” DSD 15 (2008), 24–49; both articles are reprinted in idem , Reading and Re-Reading Scripture at Qumran (STDJ 107; 2 vols.; Leiden: Brill, 2013), 1:39–62, 476–497; Crawford, Rewriting Scripture (see n. 7), 56–57.

504

Reinhard G. Kratz

has changed his mind and now advocates for their designation as “biblical” manuscripts.10 The same (Proto-Samaritan) textual tradition seems to underlie the excerpts and the rewriting of the biblical text in the Temple Scroll.11 This classification does not exclude changes that go beyond the usual smaller variants but, on the contrary, includes rearrangements, omissions, and additions as we find them in 4Q364 and especially in 4Q365 and 4Q158.12 If this classification of the 4QRP manuscripts as editions of the “biblical” text – equal to the Masoretic or the Samaritan versions – is correct, the implications of these different versions of the Pentateuch for the pentateuchal question, i.e., the question of the composition of the Pentateuch, are obvious. Taking 4Q158 as an example, the differences between this version and the other versions (Masoretic Text, Samaritan Pentateuch) – which concern both the arrangement of the material and the text – seem, at least to me, to be significant with respect to redaction history. The evidence allows two basic conclusions. First, we may draw a conclusion with respect to the comparability and applicability of the evidence. The kind of difference that we find in the RP texts 10  See E. Tov , “The Textual Status of 4Q364–367 (4QRP),” in The Madrid Qumran Congress: Proceedings of the International Congress on the Dead Sea Scrolls, Madrid, 18–21 March 1991 (ed. J. Trebolle Barrera and L. Vegas Montaner; STDJ 11; Leiden: Brill, 1992), 42–82; idem, “Biblical Texts as Reworked in Some Qumran Manuscripts, with Special Attention to 4QRP and 4QParaGen–Exod,” in The Community of the Renewed Covenant: The Notre Dame Symposium on the Dead Sea Scrolls (ed. E. Ulrich and J. C. VanderKam; Notre Dame, IN: University of Notre Dame Press, 1994), 111–134; idem, “Rewritten Bible Compositions and Biblical Manuscripts, with Special Attention to the Samaritan Pentateuch,” DSD 5 (1998), 334–354, reprinted in idem, Hebrew Bible (see n. 7), 57–70; idem, “Three Strange Books of the LXX: 1 Kings, Esther, and Daniel Compared with Similar Rewritten Compositions from Qumran and Elsewhere,” in Die Septuaginta: Texte, Kontexte, Lebenswelten; Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal, 20.–23. Juli 2006 (ed. M. Karrer and W. Kraus; WUNT 219; Tübingen: Mohr Siebeck, 2008), 369–393, reprinted in idem , Hebrew Bible, 283–308; idem , “The Many Forms of Hebrew Scripture: Reflections in Light of the LXX and 4QReworked Pentateuch,” in From Qumran to Aleppo: A Discussion with Emanuel Tov about the Textual History of Jewish Scriptures in Honor of his 65th Birthday (ed. A. Lange et al.; FRLANT 230; Göttingen: Vandenhoeck & Ruprecht, 2009), 11–28; idem, “From 4QReworked Pentateuch to 4QPentateuch (?),” in Authoritative Scriptures in Ancient Judaism (ed. M. Popović; JSJSup 141; Leiden: Brill, 2010), 73–91; idem, Textual Criticism of the Hebrew Bible (3rd ed.; Minneapolis: Fortress, 2012), 323. 11  See M. M. Zahn , “4QReworked Pentateuch C and the Literary Sources of the Temple Scroll: A New (Old) Proposal,” DSD 19 (2012), 133–158, contra S. W. C rawford , “Three Fragments from Cave Four and Their Relationship to the Temple Scroll,” JQR 85 (1994), 259–273. For the classification of 4Q365a as 4QTemplea, see Tov and Crawford, “Reworked Pentateuch” (see n. 7). 12  See M. S egal , “Biblical Exegesis in 4Q158: Techniques and Genre,” Textus 19 (1998), 45–62; M. M. Z ahn , Rethinking Rewritten Scripture: Composition and Exegesis in the 4QReworked Pentateuch Manuscripts (STDJ 95; Leiden: Brill, 2011); and the relevant contributions of Sidnie White Crawford and Molly M. Zahn in this volume.

Reworked Pentateuch and Pentateuchal Theory

505

is not a new phenomenon. It occurs not only in other versions such as the Proto-Samaritan Pentateuch but also in other cases. It is encountered, for example, in the relationship of Chronicles to the books of Samuel and Kings, for which there is an additional textual witness in the manuscript of 4QSama. Sometimes the objection is raised that it is precisely the external evidence that proves that our usual methods of internal analysis are inadequate. If we only had Chronicles, it is said, then, using the classical – some say “traditional” – methods of source and redaction criticism (Literar- and Redaktionskritik), we would never be able to reconstruct the older version of Samuel and Kings.13 Correspondingly, if we had only 4Q158 as a basis for the internal analysis, we would never obtain the biblical version of Genesis and Exodus. The objection is undoubtedly justified. However, it does not follow that internal literary-historical analysis is obsolete. Various textual witnesses show that on the way from the Vorlage to later versions, we must assume an infinite number of intermediate stages of which we are not aware. The versions that have been passed down to us are only sporadic stations in an extremely complex literary and textual history. And yet, they give us insight into the multiplicity of possibilities to be expected, some of which (with luck) can be proven or made plausible on the basis of the available textual evidence and others that will probably remain hidden forever. This means that 4Q158 and the Reworked Pentateuch texts, just like the Proto-Samaritan version, are witnesses to the way in which the version preserved in the Masoretic Text could have come into being – no more, no less. If this is the case, then the external evidence allows a further basic conclusion to be drawn. This conclusion concerns the methodological options for how to explain the evidence. At no point in 4Q158, the other RP texts, or even Chronicles does the Documentary Hypothesis immediately suggest itself as an explanation of the similarities and differences between the different versions of basically the same text. Nowhere are multiple versions – i.e., “sources” or “documents” containing the same story, which emerged independently and which later would have been conflated mechanically with each other – presupposed. In addition, the classical Fragmentary Hypothesis, which says that diverse, free-floating pieces or “sources” were fitted together by a redactor, cannot, as far as I can see, be observed in the relationship of the RP texts to their “biblical” parallels.14 13  See C arr , Formation (see n. 4), 57–65, 73–80, and especially 146; Tertel , Text and Transmission (see n. 4), 56–67, 156–171. 14  The Fragmentary or “Source” Hypothesis is often used to explain the new material (Sondergut) in the Rewritten Bible (or Rewritten Scripture) texts. For the Temple Scroll, see A. M. W ilson and L. W ills , “Literary Sources of the Temple Scroll,” HTR 75 (1982), 275–288; M. O. Wise, A Critical Study of the Temple Scroll from Qumran Cave 11 (SAOC 49; Chicago: The Oriental Institute of the University of Chicago, 1990); for Jubilees, M. Segal, The Book of Jubilees: Rewritten Bible, Redaction, Ideology and Theology (JSJSup 117; Leiden: Brill, 2007). There are, however, different approaches; for Jubilees, see C. Berner,

506

Reinhard G. Kratz

What can be observed, however, is the merging or even harmonization of two different versions of a text (e.g., Exod 20 and Deut 5 in 4Q158). What can further be observed is the supplementation of a presupposed text with material that came from elsewhere, be it the oral tradition or other written sources, some of which are known to us from biblical tradition and some of which are unknown (frag. 14 in 4Q158, etc.; cf. the distinctive material [Sondergut] in Chronicles). In both cases, the differences in the arrangement and the version of the text rest on a more or less direct relationship of literary dependence between the “biblical” text (the Vorlage) and the other versions. This means that, in order to explain the relationship between the different versions of the Pentateuch (MT, Proto-Samaritan, RP), the classical Supplementary Hypothesis or revisional approach seems to be the most appropriate. This hypothesis does include elements of the Documentary (or Source) and Fragmentary Hypotheses but cannot be replaced by them. If we assume this to be true for the external evidence, then the same must also be valid for the internal analysis and thus has important implications for pentateuchal theory. Now let us turn to our sample text, 4Q158. I chose three examples, which I want to consider from the two vantage points of the RP and the Masoretic Text: the patriarchal references, the mountain of God, and the Decalogue. In addition, I will briefly discuss the minor alterations of the text.

3 Patriarchal References Let us begin with the first example, the references to the patriarchs. 4Q158 is probably – or rather, certainly – a copy of the Exodus scroll, whether just this individual scroll or part of a larger composition that also included Genesis, or the entire Pentateuch.15 In any case, what we find in several places in the context of this Exodus scroll are additional references to the patriarchs, which we do not find in the Masoretic Text:16 Jahre, Jahrwochen und Jubiläen: Heptadische Geschichtskonzeptionen im Antiken Judentum (BZAW 363; Berlin: de Gruyter, 2006); J. L. Kugel, A Walk through Jubilees: Studies of the Book of Jubilees and the World of Its Creation (JSJSup 156; Leiden: Brill, 2012); for the Temple Scroll, Kratz, “Sources, Fragments, and Additions” (see n. 6). 15  For the identification of the book of Exodus, see J. S trugnell , “Notes en marge du volume V de ‘Discoveries in the Judaean Desert of Jordan,’ ” RevQ 7 (1970), 163–276, here 169; Segal, “Exegesis” (see n. 12), 48; Zahn, Rethinking (see n. 12), 55–56. 16  For the following I refer to the transcription and translation of Z ahn , Rethinking (see n. 12), 245–258; eadem, “Building” (see n. 9). For the text, see also A. B. Perrin, “Towards a New Edition of 4QReworkedPentateucha (4Q158): Text, Translation, Variants and Notes,” in Celebrating the Dead Sea Scrolls: A Canadian Collection (ed. J. Duhaime et al.; Atlanta: Society of Biblical Literature, 2011), 59–76; idem, “The Variants of 4Q(Reworked) Pentateuch: A Comprehensive List of the Textual Variants in 4Q158, 4Q364–367 in Biblical Sequence,”

Reworked Pentateuch and Pentateuchal Theory

507

A. In frags. 1–2 we find an additional reference to Jacob, with a repetition of Gen 32:25–32 before Exod 4:27–28.17 B. In frag 4:6–7 we find a reference to Abraham in the sacrifice scene of Exod 24:3–8, alluding to the showing of the land (Gen 12:1; 13:14–15; cf. Exod 6:3) and the covenant (Gen 17:7; Exod 6:4).18 C. Another added reference to the patriarchs is the promise of the land in frag. 3, with an allusion to Gen 49:1 or 47:29–30.19 D. In my opinion, the surplus in the blessing of Jacob in frags. 1–2, lines 7–9 belongs to the same category. Here, within the rendering of Gen 32, the combination of the roots ‫ ברך‬and ‫ פרה‬recalls not only Gen 28:3 but also Gen 35:9–12; thus, both passages wherein Jacob is renamed Israel (Gen 32 and 35) are in view.20 The compositional technique used here is intertextuality, which creates textual links within a larger composition. What implications do these intertextual connections have for pentateuchal theory? Let us begin with the repetition of Gen 32:25–33 in Exod 4 (frags. 1–2). In 4Q158 this repetition seems to be an addition that has been inserted between the “bridegroom of blood” pericope in Exod 4:24–26 and the meeting of Aaron and Moses at the mountain of God in Exod 4:27–28, which follows in frags. 1–2, lines 14–15. This assumption is supported by the word ‫ שרית‬in line 2 of frags. 1–2, which most commentators leave unexplained.21 It is most likely the verbal form śārîtā “you have wrestled” from Gen 32:29. What has irritated scholars is the fact that, in some sense, the word is in the wrong place, namely before the scene of Gen 32:25–33, in which it would have been expected in frags. 1–2, lines 5–6 (where one has to restore, ‫)כי שרית עם אלהים ועם אנשים ותוכל‬. This JJS 63 (2012), 127–157; for the analysis of the text, see Segal, “Exegesis” (see n. 12); Zahn, Rethinking, 25–74. 17  S egal , “Exegesis” (see n. 12), 47–48; Z ahn , Rethinking (see n. 12), 27–29, 41–43, 45–48, 54–56. 18  Segal, “Exegesis” (see n. 12), 49–50; Zahn, Rethinking (see n. 12), 48–54. 19  Segal, “Exegesis” (see n. 12), 53–54; Zahn, Rethinking (see n. 12), 62–63. 20  Segal, “Exegesis” (see n. 12), 59–60; Zahn, Rethinking (see n. 12), 27–29. There is a possible overlap with 4Q364 frag. 5, col. ii. 21  Zahn, Rethinking (see n. 12), 54–56, 246–247, 259; Segal, “Exegesis” (see n. 12), 48, rightly sees the relation to the Jacob story in Gen 32 and Hos 12:4, 5 but does not offer an explanation for the position of the word before the retelling of Gen 32. Hos 12, too, makes a connection between Jacob and Moses. The following interpretation is based on the assumption that the “bridegroom of blood” pericope preceded the retelling of Gen 32:25–33. A different interpretation is necessary if the reading ‫שרית ופליטה‬, “remnant and survivor,” is correct (for the expression, see 1 Kgs 19:31 // Isa 37:32; Ezra 9:14, as well as 1QS 4:14 // 4Q257 5:13; CD 2:6 // 4Q266 2 ii:6; 4Q374 2ii:4; further 1QM 1:6), which might also be an allusion to the Jacob story (Gen 32:9); for the reading, see E. Qimron, The Dead Sea Scrolls: The Hebrew Writings (3 vols.; Jerusalem: Yad Ben-Zvi, 2010–2014), 3:16.

508

Reinhard G. Kratz

suggests that the second-person singular “you” of ‫ שרית‬in line 2 is most likely referring not to Jacob but to Moses (or Zipporah) and is a secondary reference to the preceding pericope of the “bridegroom of blood” in Exod 4:24–26. As a ַ and the “bridegroom result of the similarity between the fight at Yabok (‫)ּכף ַהּיָ ֵרְך‬ of blood” (‫ן־ּד ִמם‬ ָ ‫)ח ַת‬, ֲ 22 the pericope in Gen 32 was inserted at this point in order to make the relationship explicit before the text continues with Exod 4:27–28 in 4Q158 frags. 1–2, lines 14–15. A retrospective, which compared the “wrestling” of Moses (or Zipporah) with the “wrestling” of Jacob, might have been used as the transition for interpolating the pericope at this point. However, it must be admitted that we are not certain which text preceded the rendering of Gen 32 in our Exodus scroll and that the interpretation offered here only works if it was the “bridegroom of blood” pericope. With the reproduction of Gen 32 within the context of Exod 4, further changes are made in the text of Gen 32 itself. One change – clearly recognizable as an addition by means of Wiederaufnahme23 – is the surplus in the blessing of Gen 32:30–31 in frags. 1–2, lines 7–9. Here, expressions are used that recall Gen 28:3 and Gen 35:9–12 in particular but that also contain new material. Another change within the rendering of Gen 32 in frag 1–2 of our manuscript is the interpretation of the dietary law in Gen 32:33 concerning the sciatic nerve (‫)ּגִ יד ַהּנָ ֶשׁה‬ as halakah, which is clearly ascribed to divine law (frags. 1–2, lines 11–13).24 Unlike other versions, such as Jub 48:2–3, events depicted in the “bridegroom of blood” pericope are not explained as being the result of an attack by Mastema or other evil powers but are conceived in the sense of Gen 32 as a narrative about blessing and halakah. This is probably the reason why this version of the text reproduces Gen 32 at this point in Exod 4. For the sake of a cross-check, let us for a moment forget the Masoretic Text. If we did not have the Masoretic Text but only 4Q158 (without Genesis), we would take the passage from Gen 32 in frags. 1–2 as an integral part of the Exodus scroll. Some scholars, especially those who believe in the literary integrity of the Scriptures and work with the canonical approach, would certainly take it as an original passage of this scroll. Still, literary critics with a trained eye for compositional techniques would not hesitate to regard the passage as an addition, which interrupts the older connection between Exod 4:20–26 and 4:27–28, regardless of the origin of the material. Since we know the Masoretic Text, the piece proves itself to be an additional (secondary) intertextual connection, which  For a contextual reading and the relation between the “bridegroom of blood” and Exod 4:21–23 as well as Gen 32, see B. P. Robinson, “Zipporah to the Rescue: A Contextual Study of Exodus IV 24–6,” VT 36 (1986), 447–461, 450–452; for earlier witnesses of this relation, see Zahn, Rethinking (see n. 12), 55, n. 52 with reference to Rashbam and Moshe Greenberg. 23  The phrase ‫ ויברך אותו שם‬from Gen 32:30 (= frags. 1–2, line 7) was the motivation for the addition, which ends with the words ‫ בברכו אותו שם‬in line 10, before the text of Gen 32:31 continues. 24  Zahn, Rethinking (see n. 12), 41–43. 22

Reworked Pentateuch and Pentateuchal Theory

509

multiplies the references to the patriarchs, of which there are already a number in Exod 3–4 and 6. The same can be said of the reference to Abraham in frag. 4. But the references that are to be found in the Masoretic Text are not only multiplied but also used to make certain implications explicit and to implement them in the basic narrative. Thus, the explicit prohibition of eating the sciatic nerve (‫)ּגִ יד ַהּנָ ֶשׁה‬, in frags.1–2, lines 11–13, and other examples stress the correspondence of divine commands and their exact execution,25 which is to be found in a number of other cases in the base text as well. In terms of pentateuchal theory, this means that we are dealing with extensions or supplements to an older base text. The Documentary or Fragmentary Hypothesis is excluded here. Let us now turn our attention to the Masoretic Text itself. What we can discern in 4Q158 can, to some extent, also be observed in the Masoretic version. I will not be able to offer a complete analysis here, but I will mention a few of the issues. There is consensus among scholars that the references to the patriarchs in Exod 3–4 and in Exod 6 are (relatively) late texts. Exodus 6:2–12 belongs to P (the Priestly writing). Here, it is a matter of debate whether P is a source or a revision.26 In light of 4Q158, the evidence for the revisional theory increases, regardless of whether the revision took place in the same literary context or in a separate writing. Be that as it may, with its concept of blessing, P seems to presuppose the blessing/curse theme in Gen 1–11 and 12:1–3 in the non-Priestly text.27 Exodus 3–4 is regarded by many scholars as a very late, post-Priestly text, serving as a secondary connection between the patriarchs and the exodus.28 Others divide the text into the sources J and E and assign the patriarchal references (with the exception of Exod 4:5) to E.29 But however we evaluate or divide the text, it is precisely the explicit references to the patriarchs that scholars

 Zahn, Rethinking (see n. 12), 37–48.  See R. G. K ratz , Die Komposition der erzählenden Bücher des Alten Testaments (Göttingen: Vandenhoeck & Ruprecht, 2000), 113–116, 247–248; English translation, idem, The Composition of the Narrative Books of the Old Testament (London: T&T Clark, 2005), 109–113, 244–246; idem, “Pentateuch” (see n. 1), 36–38. 27  See Kratz, Komposition (see n. 26), 233, 248 (Composition [see n. 26], 232, 245); idem, “Pentateuch” (see n. 1), 51–52, n. 65. 28  E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 20–28; K. Schmid, Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 186–208; English translation, idem, Genesis and the Moses Story: Israelʼs Dual Origins in the Hebrew Bible (trans. J. D. Nogalski; Siphrut: Literature and Theology of the Hebrew Scriptures 3; Winona Lake, IN: Eisenbrauns, 2010). 29  Thus M. N oth , Überlieferungsgeschichte des Pentateuch (Stuttgart: Kohlhammer 1948), 39; on Exod 4:4 see 31 (“J” in brackets). 25

26

510

Reinhard G. Kratz

have excluded as later additions to this text.30 The relevant passages containing references to the patriarchs that are held to be later additions to the text follow, with additions indented. Exodus 3:5–7 ‫׃‬ ‫עֹומד ָע ָליו ַא ְד ַמת־ק ֶֹדׁש הּוא‬ ֵ ‫ל־ּת ְק ַרב ֲהֹלם ַׁשל־נְ ָע ֶליָך ֵמ ַעל ַרגְ ֶליָך ִּכי ַה ָּמקֹום ֲא ֶׁשר ַא ָּתה‬ ִ ‫אמר ַא‬ ֶ ֹ ‫וַ ּי‬ ‫אֹלהי יַ ֲעקֹב‬ ֵ ֵ‫ֹלהי יִ ְצ ָחק ו‬ ֵ ‫ֹלהי ַא ְב ָר ָהם ֱא‬ ֵ ‫ֹלהי ָא ִביָך ֱא‬ ֵ ‫אמר ָאנ ִֹכי ֱא‬ ֶ ֹ ‫וַ ּי‬ ‫׃‬ ‫ֹלהים‬ ִ ‫ל־ה ֱא‬ ָ ‫וַ ּיַ ְס ֵּתר מ ֶֹׁשה ָּפנָ יו ִּכי יָ ֵרא ֵמ ַה ִּביט ֶא‬ ‫ת־צ ֲע ָק ָתם ָׁש ַמ ְע ִּתי ִמ ְּפנֵ י נֹגְ ָׂשיו ִּכי יָ ַד ְע ִּתי‬ ַ ‫ת־ענִ י ַע ִּמי ֲא ֶׁשר ְּב ִמ ְצ ָריִ ם וְ ֶא‬ ֳ ‫יתי ֶא‬ ִ ‫ֹאמר יְ הוָ ה ָראֹה ָר ִא‬ ֶ ‫וַ ּי‬ ‫׃‬ ‫ת־מ ְכא ָֹביו‬ ַ ‫ֶא‬ And He said, “Do not come closer. Remove your sandals from your feet, for the place on which you stand is holy ground. I am,” He said, “the God of your father, the God of Abraham, the God of Isaac, and the God of Jacob.” And Moses hid his face, for he was afraid to look at God. And YHWH continued, “I have marked well the plight of My people in Egypt and have heeded their outcry because of their taskmasters; yes, I am mindful of their sufferings.” Exodus 3:13–16 ֶ ‫בֹות‬ ֵ ‫ֹלהי ֲא‬ ֵ ‫ל־ּבנֵ י יִ ְׂש ָר ֵאל וְ ָא ַמ ְר ִּתי ָל ֶהם ֱא‬ ְ ‫ֹלהים ִהּנֵ ה ָאנ ִֹכי ָבא ֶא‬ ִ ‫ל־ה ֱא‬ ָ ‫אמר מ ֶֹׁשה ֶא‬ ֶ ֹ ‫וַ ּי‬ ‫יכם‬ ֶ ‫יכם ְׁש ָל ַחנִ י ֲא ֵל‬ ‫׃‬ ‫ה־ּׁשמֹו ָמה א ַֹמר ֲא ֵל ֶהם‬ ְ ‫רּו־לי ַמ‬ ִ ‫וְ ָא ְמ‬ ֶ ‫אמר ִל ְבנֵ י יִ ְׂש ָר ֵאל ֶא ְהיֶ ה ְׁש ָל ַחנִ י ֲא ֵל‬ ַ ֹ ‫אמר ּכֹה ת‬ ֶ ֹ ‫ֹלהים ֶאל־מ ֶֹׁשה ֶא ְהיֶ ה ֲא ֶׁשר ֶא ְהיֶ ה וַ ּי‬ ִ ‫אמר ֱא‬ ֶ ֹ ‫וַ ּי‬ ‫׃‬ ‫יכם‬ ‫ֹלהי ַא ְב ָר ָהם‬ ֵ ‫יכם ֱא‬ ֶ ‫ֹלהי ֲאב ֵֹת‬ ֵ ‫ל־ּבנֵ י יִ ְׂש ָר ֵאל יְ הוָ ה ֱא‬ ְ ‫ֹאמר ֶא‬ ַ ‫ֹלהים ֶאל־מ ֶֹׁשה ּכֹה־ת‬ ִ ‫ֹאמר עֹוד ֱא‬ ֶ ‫וַ ּי‬ ְ ֶ‫יכם ז‬ ֶ ‫אֹלהי יַ ֲעקֹב ְׁש ָל ַחנִ י ֲא ֵל‬ ֵ ֵ‫ֹלהי יִ ְצ ָחק ו‬ ֵ ‫ֱא‬ ‫׃‬ ‫ה־ּׁש ִמי ְלע ָֹלם וְ זֶ ה זִ ְכ ִרי ְלד ֹר ּד ֹר‬ ‫יכם נִ ְר ָאה ֵא ַלי‬ ֶ ‫ֹלהי ֲאב ֵֹת‬ ֵ ‫ֵלְך וְ ָא ַס ְפ ָּת ֶאת־זִ ְקנֵ י יִ ְׂש ָר ֵאל וְ ָא ַמ ְר ָּת ֲא ֵל ֶהם יְ הוָ ה ֱא‬ ‫ֹלהי ַא ְב ָר ָהם יִ ְצ ָחק וְ יַ ֲעקֹב‬ ֵ ‫ֱא‬ ‫׃‬ ‫ת־ה ָעׂשּוי ָל ֶכם ְּב ִמ ְצ ָריִ ם‬ ֶ ‫ֵלאמֹר ָּפקֹד ָּפ ַק ְד ִּתי ֶא ְת ֶכם וְ ֶא‬ Moses said to God, “When I come to the Israelites and say to them ‘The God of your fathers has sent me to you,’ and they ask me, ‘What is His name?’ what shall I say to them?” And God said to Moses, “ʾEhyê ʾăšer ʾehyê.” He continued, “Thus shall you say to the Israelites, ‘ʾEhyê sent me to you.’ ” And God said further to Moses, “Thus shall you speak to the Israelites: YHWH, the God of your fathers, the God of Abraham, the God of Isaac, and the God of Jacob, has sent me to you: This shall be My name forever, This My appellation for all eternity.” “Go and assemble the elders of Israel and say to them: YHWH, the God of your fathers has appeared to me, the God of Abraham, Isaac, and Jacob and said: ‘I have taken note of you and of what is being done to you in Egypt.’ ”

 C. L evin , Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993), 332; J. C. G ertz , Tradition und Redaktion in der Exoduserzählung: Untersuchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000), 270–271, 294–295, differently on Exod 4:5, 312, and cf. 363; C. Berner, Die Exoduserzählung: Das literarische Werden einer Ursprungslegende Israels (FAT 73; Tübingen: Mohr Siebeck, 2010), 83–85, 88, 90–91, 104, 111, 134. 30

Reworked Pentateuch and Pentateuchal Theory

511

Exodus 4:4–6

‫׃‬ ‫אמר יְ הוָ ה ֶאל־מ ֶֹׁשה ְׁש ַלח יָ ְדָך וֶ ֱאחֹז ִּבזְ נָ בֹו וַ ּיִ ְׁש ַלח יָ דֹו וַ ּיַ ֲחזֶ ק ּבֹו וַ יְ ִהי ְל ַמ ֶּטה ְּב ַכּפֹו‬ ֶ ֹ ‫וַ ּי‬ ‫׃‬ ‫אֹלהי יַ ֲעקֹב‬ ֵ ֵ‫ֹלהי יִ ְצ ָחק ו‬ ֵ ‫ֹלהי ַא ְב ָר ָהם ֱא‬ ֵ ‫ֹלהי ֲאב ָֹתם ֱא‬ ֵ ‫ְל ַמ ַען יַ ֲא ִמינּו ִּכי־נִ ְר ָאה ֵא ֶליָך יְ הוָ ה ֱא‬ ‫אמר יְ הוָ ה לֹו עֹוד ָ ֽה ֵבא־נָ א ָי ְ�ֽדָך ְּב ֵח ֶיקָך וַ ֵּיָבא יָ דֹו ְּב ֵחיקֹו‬ ֶ ֹ ‫וַ ּי‬ Then YHWH said to Moses, “Put out your hand and grasp it by the tail” – he put out his hand and seized it, and it became a rod in his hand – “that they may believe that the LORD, the God of their fathers, the God of Abraham, the God of Isaac, and the God of Jacob, did appear to you.” YHWH said to him further, “Put your hand into your bosom.” He put his hand into his bosom.

If the assumption is correct that the references to the patriarchs (noted by indentation) are later additions, the evidence in Exod 3–4 is in line with what we can observe in 4Q158: the additions create an explicit connection between the exodus narrative and the patriarchs in Exod 3–4. This connection is further explained by the Priestly layer in Exod 6, which provides a step-by-step system of revelation of Godʼs name! This process of intertextual linking between the patriarchal narrative and the Moses narrative is continued in 4Q158 with the repetition of Gen 32 in Exod 4 and the further references to the patriarchs.

4 The Mountain of God My second example is the interrelation of the mountain of God in Exod 3–4 and 19–24 which, again, is to be found in 4Q158 frags. 1–2 and 4.31 In frags. 1–2, lines 16–19, at Exod 4:27–28, the command for the exodus from Exod 3:12 is repeated, and in frag. 4 the execution of a sacrifice is described in the words of Exod 24:3–8. However, frag. 4 clearly alludes to Exod 3:12 and 24:3–8 in the context of Exod 4 before the Sinai pericope is reproduced in frags. 5–12 and 13. Based on the remnants of Exod 19–22 in frags. 5–12 and of Exod 30 in frag. 13, we do not know whether Exod 24 was reproduced again in 4Q158. I suspect that the chapter was reproduced at its original position in Exod 24. Similar to the case of ‫ שרית‬from Gen 32 in frags. 1–2, in frag. 4 we are dealing with an anticipation and secondary combination––in this case, the juxtaposition of the mountain of God in Exod 3–4 with Mount Sinai in Exod 19–24. The expression in frag. 4, line 1, ‫צוה לכה‬, “he commanded you,” refers to Exod 3:12 (and 3:18). Here, the exodus is connected with the “worship” at the mountain of God from ִ ‫ת־ה ֶא‬ ָ ‫)ת ַע ְבדּון ֶא‬. ַ It is this command that is executed in Exod 3 (‫ֹלהים ַעל ָה ָהר ַהּזֶ ה‬ frag. 4 in the context of Exod 4 according to the example of the later scene in Exod 24:4–8. Thus, the two mountains, the mountain of God in Exod 3–4 and Mount Sinai in Exod 19–24, are correlated, which is indeed suggested by the 31  For the following, see Strugnell, “Notes” (see n. 15), 169–170; Segal, “Exegesis” (see n. 12), 49–53; Zahn, Rethinking (see n. 12), 43–54.

512

Reinhard G. Kratz

name of the mountain, “the mountain of God” (‫ֹלהים‬ ִ ‫)הר ָה ֱא‬, ַ in Exod 3:1 and the description of the thorn bush, ‫ה ְּסנֶ ה‬, ַ in Exod 3:2. Again, intertextual references or implications, which are already to be found in the (Proto-Masoretic) base text, are made explicit and implemented in the basic narrative in 4Q158. Here, the sacrifice and the covenant from Exod 24:4–8 gets a new interpretation in the context of Exod 3–4. So in frag. 4, lines 6–8 the covenant of Exod 24 is rephrased in the light of the patriarchal blessings and the covenant formula from Gen 12–13, Gen 17, and Exod 6.32 Here, too, the secondary alignments would also be recognizable as such, even if we only knew the version in 4Q158. This means that the Documentary or Fragmentary Hypothesis is quite out of the question as a model for explanation; only the Supplementary or Revisional Hypothesis can be considered a suitable explanation for the redactional activity that is observed in 4Q158. How about the Masoretic Text itself? Again, the redactional process going on in 4Q158 is already in the offing.33 Here, the decisive element is Exod 4:27, where 4Q158 adds the command from Exod 3:12 in the conversation between Moses and Aaron. In the Masoretic version of Exod 4:27, the localization at the mountain of God is an anachronism, as Moses has long since returned to Egypt (v. 20, ‫)וַ ּיָ ָׁשב ַא ְר ָצה ִמ ְצ ָריִ ם‬. Originally this was followed either by the exodus (Exod 12:35–38 after 3:21–22) or the meeting with the pharaoh and the stories of the plagues (Exod 3:16–20/4:27–31). Therefore, Exod 4:27 is usually considered to be secondary.34 Exodus 4:20, 27–31

]. . .[ ‫ל־ה ֲחמֹר וַ ּיָ ָׁשב ַא ְר ָצה ִמ ְצ ָריִ ם‬ ַ ‫ת־ּבנָ יו וַ ּיַ ְר ִּכ ֵבם ַע‬ ָ ‫ת־א ְׁשּתֹו וְ ֶא‬ ִ ‫וַ ּיִ ַּקח מ ֶֹׁשה ֶא‬ ‫׃‬ ‫ֹלהים וַ ּיִ ַּׁשק־לֹו‬ ִ ‫ל־א ֲהר ֹן ֵלְך ִל ְק ַראת מ ֶֹׁשה ַה ִּמ ְד ָּב ָרה וַ ּיֵ ֶלְך וַ ּיִ ְפּגְ ֵׁשהּו ְּב ַהר ָה ֱא‬ ַ ‫אמר יְ הוָ ה ֶא‬ ֶ ֹ ‫וַ ּי‬ ‫׃‬ ‫אתֹת ֲא ֶׁשר ִצּוָ הּו‬ ֹ ‫ל־ה‬ ָ ‫ל־ּד ְב ֵרי יְ הוָ ה ֲא ֶׁשר ְׁש ָלחֹו וְ ֵאת ָּכ‬ ִ ‫וַ ּיַ ּגֵ ד מ ֶֹׁשה ְל ַא ֲהר ֹן ֵאת ָּכ‬ ‫׃‬ ‫ת־ּכל־זִ ְקנֵ י ְּבנֵ י יִ ְׂש ָר ֵאל‬ ָ ‫וַ ּיֵ ֶלְך מ ֶֹׁשה וְ ַא ֲהר ֹן וַ ּיַ ַא ְספּו ֶא‬ ‫׃‬ ‫אתֹת ְל ֵעינֵ י ָה ָעם‬ ֹ ‫ר־ּד ֶּבר יְ הוָ ה ֶאל־מ ֶֹׁשה וַ ּיַ ַעׂש ָה‬ ִ ‫ל־ה ְּד ָב ִרים ֲא ֶׁש‬ ַ ‫וַ יְ ַד ֵּבר ַא ֲהר ֹן ֵאת ָּכ‬ ‫׃‬ ‫ת־ענְ יָ ם וַ ּיִ ְּקדּו וַ ּיִ ְׁש ַּת ֲחּוּו‬ ָ ‫ת־ּבנֵ י יִ ְׂש ָר ֵאל וְ ִכי ָר ָאה ֶא‬ ְ ‫י־פ ַקד יְ הוָ ה ֶא‬ ָ ‫וַ ּיַ ֲא ֵמן ָה ָעם וַ ּיִ ְׁש ְמעּו ִּכ‬ So Moses took his wife and sons, mounted them on an ass, and went back to the land of Egypt; [. . .] YHWH said to Aaron, “Go to meet Moses in the wilderness.” He went and met him at the mountain of God, and he kissed him. Moses told Aaron about all the things that YHWH had committed to him and all the signs about which He had instructed him. Then Moses and Aaron went and assembled all the elders of the Israelites. Aaron repeated all the words that YHWH had spoken to Moses, and he performed the signs in the sight of the people,

 Cf. Segal, “Exegesis” (see n. 12), 49–51; Zahn, Rethinking (see n. 12), 52–53.  For the following, see C. Berner, “The Redaction History of the Sinai Pericope (Exod 19–24) and Its Continuation in 4Q158,” DSD 20 (2013), 376–407, here 399–406. 34  L evin , Jahwist (see n. 30), 333; G ertz , Tradition (see n. 30), 334; B erner , Exoduserzählung (see n. 30), 115. 32 33

Reworked Pentateuch and Pentateuchal Theory

513

and the people were convinced. When they heard that YHWH had taken note of the Israelites and that He had seen their plight, they bowed low in homage.

However, there is some controversy here: is only the mention of Aaron secondary35 or are all of the Aaron passages in Exod 3–4 (including 5:1–2 and the plagues) interpolations?36 It has also been suggested that only Exod 4:27 is a secondary addition.37 Be that as it may, it is obvious that this verse creates a problem. It interprets the return in v. 20 in terms of a departure and shifts the scene that follows in v. 28(–31) to the mountain of God. Here, Moses and Aaron, according to Exod 4:28–31, pass the instructions from Exod 3:16–18 on to the people. Exodus 3:16–18, in turn, mentions a “sacrifice in the wilderness” that serves as a pretext for the departure (see Exod 10). Furthermore, Exod 4:27 alludes to Mosesʼs stop at the mountain of God in Exod 3:1, 12. This allusion itself is also often considered to be secondary. Of particular interest is Exod 3:12b, which connects the exodus with worship at the mountain of God and is likely to be an addition within Exod 3:9–14.38 Exodus 3:11–12 ‫׃‬ ‫ת־ּבנֵ י יִ ְׂש ָר ֵאל ִמ ִּמ ְצ ָריִ ם‬ ְ ‫אֹוציא ֶא‬ ִ ‫ל־ּפ ְרעֹה וְ ִכי‬ ַ ‫ֹלהים ִמי ָאנ ִֹכי ִּכי ֵא ֵלְך ֶא‬ ִ ‫ל־ה ֱא‬ ָ ‫אמר מ ֶֹׁשה ֶא‬ ֶ ֹ ‫וַ ּי‬ ֶ ‫אמר ִּכ‬ ֶ ֹ ‫וַ ּי‬ ‫י־א ְהיֶ ה ִע ָּמְך‬ ‫׃‬ ‫ֹלהים ַעל ָה ָהר ַהּזֶ ה‬ ִ ‫ת־ה ֱא‬ ָ ‫ת־ה ָעם ִמ ִּמ ְצ ַריִ ם ַּת ַע ְבדּון ֶא‬ ָ ‫יאָך ֶא‬ ֲ ‫הֹוצ‬ ִ ‫ה־ּלָך ָהאֹות ִּכי ָאנ ִֹכי ְׁש ַל ְח ִּתיָך ְּב‬ ְ ֶ‫וְ ז‬ But Moses said to God, “Who am I that I should go to Pharaoh and free the Israelites from Egypt?” And He said, “I will be with you. And that shall be your sign that it was I who sent you. And when you have freed the people from Egypt, you shall worship God at this mountain.”

The motive of “worship at this mountain” in v. 12 is not quite in agreement with the “sacrifice in the wilderness” in v. 18. Still, it can be assumed that this motive, too, hints at the sacrifice in Exod 24 and the cult in the tabernacle of Sinai, along with similar worship references in the plague episodes.39 After all, it is already apparent in the Masoretic version that Mount Sinai in Exod 19–24 has been connected secondarily with the mountain of God and the “thorn bush” (‫)ה ְּסנֶ ה‬ ַ in Exod 3–4. This connection is realized firstly by Exod 3:12, which anticipates the sacrificial scene of Exod 24, and secondly by Exod 4:27, which sends Aaron to the mountain of God and refers back to Exod 3:1, 35  N oth , Überlieferungsgeschichte (see n. 29), 32, n. 104; L evin , Jahwist (see n. 30), 330, 333; also Berner, Exoduserzählung (see n. 30), 93 for 4:29 and 5:1; Blum, Studien (see n. 28), 27–28; Gertz, Tradition (see n. 30), 336 for 5:1, etc. 36  Kratz, Komposition (see n. 26), 296–300 (Composition [see n. 26], 287–291). 37  Berner, Exoduserzählung (see n. 30), 115. 38  L evin , Jahwist (see n. 30), 331; G ertz , Tradition (see n. 30), 292–294; B erner , Exoduserzählung (see n. 30), 80–83, 103. 39  Exod 7:16, 26; 8:16, “worship”; also 5:1, “celebration”; animals for sacrifice in the wilderness, 10:7–11, 24–26; 12:31–32.

514

Reinhard G. Kratz

12b (mountain of God) and 3:16–18 (sacrifice in the wilderness, reference to the patriarchs). 4Q158 puts two and two together and lets a sacrificial act follow the meeting of Moses and Aaron at the mountain of God in Exod 4:27–28. This sacrificial act, however, not only corresponds to the command in Exod 3:12, as well as the sacrifice on Mount Sinai in Exod 24:4–8, but it also connects the covenant of Exod 24 to the patriarchs.40 In this way, the mountain of God and Mount Sinai are identified as one and the same in 4Q158, and the covenant at Sinai is traced back to the covenant with the patriarchs, thus following the example of the Priestly writing in Gen 17 and Exod 29 and 40.41 So we can conclude that the literary processes to be observed in the biblical text, especially Exod 3–4 and 24, prepare for the further interpretation of 4Q158. In both cases, intertextual links are used to connect and interpret texts that are usually assigned to the independent sources J, E, and P. Thus, the evidence both in the Masoretic Text and in 4Q158 is to be explained most simply with the Supplementary or Revisional Hypothesis, rather than by either the Documentary or the Fragmentary Hypothesis.

5 The Decalogue Having considered two examples with intertextual links, our third example will lead us to another phenomenon of redactional activity: the issue of different versions of a text, and the tendency to harmonize parallel passages.42 This can be seen in 4Q158 frags. 6–9, which reproduce the Sinai pericope, particularly the material of Exod 20.43 The text of 4Q158 shows a close proximity to the Samaritan (SP) and the Proto-Samaritan (4QpaleoExodm) versions of the Pentateuch and at the same time contains a number of significant differences from both the Masoretic and the Samaritan traditions. Also, the arrangement of the text is similar to, but not at all identical with, that of SP. Following Exod 19 in frag. 5, we find the following sequence:44  See above, n. 32.  Kratz, Komposition (see n. 26), 246 (Composition [see n. 26], 243). 42  For the tendency of harmonization in the Proto-Samaritan Pentateuch and the 4QRP texts, see Crawford, Rewriting (see n. 7), 22–34, 40–46; for specification, see Zahn, Rethinking (see n. 12), 233–236. 43  See Segal, “Exegesis” (see n. 12), 55–59; Zahn, Rethinking (see n. 12), 27–34, 37–40, 63–67. 44  The chart follows, with a few specifications, S egal , “Exegesis” (see n. 12), 56. The placement of frag. 6 before frags. 7–9 is not questionable; it is based on the similarity to the Proto-Samaritan (Segal, “Exegesis,” n. 31) and the fact that the replacement of the Decalogue is connected to further extensions of the Proto-Samaritan version. See also Zahn, Rethinking 40 41

Reworked Pentateuch and Pentateuchal Theory

MT

(Proto-)Samaritan

4Q158

Decalogue

Decalogue

not preserved

Exod 20:18–19aα

Exod 20:18–19aα

not preserved

Exod 20:19aβ

Deut 5:24–27

Deut 5:(24–)2745

Exod 20:19b–21

Exod 20:19b–21

Exod 20:19b–21

Exod 20:22aα

Exod 20:22aα

Exod 20:22aα46

Deut 5:28b–29

Deut 5:28b–29

515

Frag. 6

Plus Deut 18:18–22

Deut 18:18–22 Frags. 7–9 Exod 20:12–17 (Decalogue) Exod 20:22aα47

Deut 5:30–31

Deut 5:30–31 Plus

Exod 20:22aβb, 23, etc.

Exod 20:22–23, etc.

Exod 20:2248–21:25

From this chart it is obvious that 4Q158 differs from the Proto-Samaritan in two respects: first, the position of the Decalogue (at least the second part of it) is changed; second, there are three extensions of the text, which seem to be connected to the replacement of the Decalogue (frag. 6, lines 5–6; frags. 7–9, lines 3, 4–5). The technique of rearranging the material is similar to the Proto-Samaritan version and works with the introductory speech formula of Exod 20:22a. As in the case of Exod 20:21a and 21b, Exod 20:22 is also broken up and filled with new material. Thus, in SP the combination of quotations from Deut 5:28–31 and 18:18–22 is inserted between two speech formulae of Exod 20:22a by means of Wiederaufnahme (repetitive resumption). In 4Q158 this

(see n. 12), 64, who argues that the continuation of the text in frag. 7 presupposes the request for a mediator (frag. 6) beforehand. 45  See Strugnell, “Notes” (see n. 15), 171; Segal, “Exegesis” (see n. 12), 56, n. 31. 46  MT reads ‫ ;ויאמר יהוה אל משה‬4Q158 frag. 6, line 4 reads ‫ וידבר] יהוה אל מושה ל[אמור‬in accordance with SP and ‫ את כל אשר ידבר יהוה אלהינו אליך‬in Deut 5:27; see Berner, “Redaction History” (see n. 33), 390, n. 31. The Samaritan Pentateuch is quoted according to A. Tal and M. F lorentin (eds.), The Pentateuch: The Samaritan Version and the Masoretic Version (Tel-Aviv: The Haim Rubin Tel Aviv University Press, 2010). 47  Here the reading agrees with MT: ‫( ויאמר יהוה אל משה‬frags. 7–9, line 3). 48  The beginning of Exod 20:22 is not preserved; according to the text of the Proto-Samaritan it can be restored as ‫ ידבר יהוה אל מושה לאמור‬as in frag. 6, line 4; see Zahn, Rethinking (see n. 12), 252.

516

Reinhard G. Kratz

extension is supplemented by further additions, again using the speech formula of Exod 20:22a as a means of interpolation. The differences are also significant with regard to the content. The main point is what the people saw and heard at Sinai and to what role Moses was assigned therein. The Proto-Samaritan version integrates Mosesʼs review of the events at Horeb (Sinai) from Deut 5 into the scene of Exod 20:18–21 itself. This has the consequence that the people not only “see” (as in Exod 20:18) but also “hear.” Moreover, Moses is introduced as a mediator in much more detail than is the case in Exod 20:19. This also includes the announcement of a prophet according to Deut 18:18–22, which is connected with Deut 5, and the divine commission to Moses to let the people go, the execution of which, however, is not reported in the SP version. 4Q158 presupposes not the Masoretic composition but this version from the Proto-Samaritan and supplements it further.49 The purpose is primarily to smooth transitions: A. In frag. 6, line 6, at the transition from Deut 5:29 to 18:18, the announcement of a prophet by means of a commission to Moses is introduced expressly and contextualized as a response to the peopleʼs request for a mediator (the source for the wording is Deut 18:16–17). B. In frags. 7–9, lines 3–5, at the transition from Deut 5:30–31 to Exod 20:22, it is reported that the people follow Mosesʼs instructions and return to their tents before God continues to speak with Moses. C. Finally, the scene from Exod 20:18–21 (followed by Deut 5:24–31 and 18:18–22) is divided between the situation before (frag. 6) and after (frags. 7–9) the proclamation of the Decalogue. The transition from the Decalogue back to Deut 5:30–31 is made by means of Wiederaufnahme of the speech formula from Exod 20:22a, which here (in frags. 7–9, line 3) uses the wording attested in MT (‫ ;)ויאמר יהוה אל מושה‬in frag. 6, line 4 and frags. 7–9, line 5, however, the wording of SP is used (‫וידבר יהוה אל מושה‬ ‫)לאמור‬. The relocation of the Decalogue right after the interpolation of Deut 5 and 18 between Exod 20:21 and 22, is a crux interpretum. On the one hand, the quotation from Deut 5:24–27 in frag. 6 already says that the people not only “saw” something (Exod 20:18) but also “heard” God speak. On the other hand, the declaration of the Decalogue follows only afterward, in frags. 7–9. Here, scholars usually follow the explanation of Michael Segal, who, on the basis of a later rabbinic tradition, suggests that before frag. 6 only the first two 49  On the relative chronology and direction of dependence, see S egal , “Exegesis” (see n. 12), 56–57. The direction from Proto-Samaritan (attested in 4QpaleoExodm and, apart from the Samaritan glosses, in the Samaritanus) to 4Q158 is also suggested by the clear tendency of extension in 4Q158.

Reworked Pentateuch and Pentateuchal Theory

517

commandments were proclaimed to the people and then afterward commandments 3–10 were proclaimed by Moses acting as mediator.50 However, along with Christoph Berner, we have to bear in mind that this explanation cannot be proven on the basis of the text as it is preserved. Berner points out that the introduction of a speech addressed to Moses in line 3 of frags. 7–9 and the subsequent dismissal of the people can also suggest that the entire Decalogue was publicly proclaimed between frags. 6 and 7–9, before God turns to Moses, commissioning him to let the people go.51 There is, however, a third possibility, which was suggested by John Strugnell.52 He believed that, between frags. 6 and 7–9, both the proclamation of the prophet and the Decalogue were communicated first to Moses and then by Moses to the people, before God in frags. 7–9, line 3 gives the order to let the people go. The quotation from Deut 5:24–27 in frag. 6 would, in terms of Exod 20:18, only refer to the theophany and Godʼs speaking to Moses in Exod 19:9–19. Following this last thought, i.e., an understanding of Deut 5:24–27 that implies that the people heard Godʼs voice but did not necessarily understand his words, I would like to propose yet another solution. We have to take into account that the whole interpolation between Exod 20:19b–21 and 20:20 is located and arranged by means of a Wiederaufnahme of the introductory speech formula in Exod 20:22a. Thus, the interpolation is most likely to be seen as a speech of God to Moses – in front of the people, who hear the voice but not necessarily the words. This fits very well with the narrative scene of Exod 19 and continues the back and forth of Moses, who is moving between God and the people. Here, too, the people see and hear that something is going on at the top of the mountain, but Moses must deliver the words of God to the people at the foot of the mountain. However, there is one passage where the situation is confused, or at least unclear, namely, at the transition from Exod 19:25 to the Decalogue in 20:1. Here, Moses is still with the people at the foot of the mountain and speaks to them. The continuation in 20:1 can be read in two ways: (1) either God begins speaking to Moses and the people or (2) Moses is reproducing the words of God to the people. 4Q158 solved this problem by relocating the Decalogue to the speech of God to Moses starting in Exod 20:22. Thus, the conversation between Moses and the people introduced in Exod 19:25 continues in Exod 20:18–19b + Deut 5:24–27 + Exod 20:19b–21: this is before Moses, again, meets with God, and the conversation between God and Moses begins in Exod 20:22a – first with the people seeing and hearing but not necessarily understanding (Deut 5:28b–29 + Deut 18:18–22 + Decalogue + Deut 5:30–31), then without the people, who were sent home according to their wish and the commandment  Segal, “Exegesis” (see n. 12), 57–58, followed by Zahn, Rethinking (see n. 12), 66–67.  Berner, “Redaction History” (see n. 33), 396–397. 52  Strugnell, “Notes” (see n. 15), 172–173; Zahn, Rethinking (see n. 12), 67, n. 83. 50 51

518

Reinhard G. Kratz

of God to Moses. Although the people were still present and could hear (!), Moses had to deliver the message of God to the people (frags. 7–9, lines 3–5 + Exod 20:22–24). In short, I think the sequence of the material is completely consistent if we apply the difference between “seeing and hearing” and “understanding” of Exod 19 to Deut 5:24–27, recognize the problem of the transition from Exod 19:25 to 20:1, and pay attention to the role of the speech introduction in Exod 20:22a for the whole rearrangement. Be that as it may, the arrangement of 4Q158 is in any case a (further) attempt to meaningfully integrate the different versions of the peopleʼs reactions to the theophany and Godʼs speaking (either to them or to Moses) in Exod 20:18–21 and Deut 5:22–31, which the Proto-Samaritan version had already conflated. These are the findings in 4Q158. What are their implications for the Masoretic Text and pentateuchal theory? In the case of the Sinai pericope, we are dealing with not just two but three versions (MT, SP, 4Q158) which are different but which did not evolve independently from each other. Whereas the Proto-Samaritan version appears to be a first revision of the older Proto-Masoretic Text of Exod 20, 4Q158 can be seen as a second revision. The main tendency of the two revisions (SP and 4Q158) is, without any doubt, the harmonization of Exod 20 with Deut 5. However, the different ways of harmonizing reflect different conceptions of the role of Moses. If we did not have the intermediate version of the Proto-Samaritan text, we would ascribe all changes to 4Q158. If we only had the Proto-Samaritan version or only 4Q158 without the Masoretic Text, we would certainly not recognize every change but would still recognize one or the other, particularly in cases where there is the technique of Wiederaufnahme, which is used in both versions in order to integrate Deut 5 and 18 and the Decalogue.53 This shows that we have to reckon with several intermediate textual versions in our analysis, some of which we will certainly miss but others that can be reconstructed by means of internal analysis. Although we are dealing here with the conflation of several sources (Exod 20 and Deut 5), the findings, seen individually, cannot be explained using the Source or Documentary Hypothesis but are best explained with the Fragmentary Hypothesis. However, since we know the Masoretic Text and an intermediate link (the Proto-Samaritan version), it is clear that we are dealing with a revision, which bears Exod 20 and Deut 5 in mind and is making an effort to harmonize both versions. Let us now turn to the Masoretic Text itself.54 The tendency to harmonize Exod 20 and Deut 5 can be observed here as well. It is immediately apparent that Deut 5 (vv. 22–31) offers not only a recapitulation but also an interpretation of the opening scene of the Covenant Code in Exod 20:18–21. If, in Exod 20, the 53 54

 See the chart above with nn. 46–48.  For the following, see Berner, “Redaction History” (see n. 33), 380–398.

Reworked Pentateuch and Pentateuchal Theory

519

people “saw” the theophany and the proclamation of the Decalogue, in Deut 5 they “saw” the theophany and “heard” the words of God spoken (v. 22–24). This, in itself, can be regarded as a first step to harmonize both passages. The Deuteronomic conception of “seeing” and “hearing” also occurs in Deut 5:4. It is taken up in the Proto-Samaritan text and 4Q158, and in both versions is inserted into the scene of Exod 20 itself. By an addition to Deut 5:5, however, this concept is corrected to the effect that the Decalogue is only made known to the people through Moses as a mediator. A similar tendency can be observed in Exod 19:20–25 and possibly in the rearrangement of the scene before the proclamation of the Decalogue in 4Q158. Taking a closer look, we can assume that the harmonization of Exod 20 and Deut 5 has taken shape gradually in the Masoretic version as well. Let us first have a look at Exod 20:18–21. Exodus 20:18–21 ‫ת־ה ָהר ָע ֵׁשן וַ ּיַ ְרא ָה ָעם וַ ּיָ נֻ עּו וַ ּיַ ַע ְמדּו‬ ָ ‫ת־ה ַּל ִּפ ִידם וְ ֵאת קֹול ַהּׁש ָֹפר וְ ֶא‬ ַ ‫ת־הּקֹוֹלת וְ ֶא‬ ַ ‫ל־ה ָעם ר ִֹאים ֶא‬ ָ ‫וְ ָכ‬ ‫׃‬ ‫ֵמ ָרחֹק‬ ‫׃‬ ‫ֹלהים ֶּפן־נָ מּות‬ ִ ‫ר־א ָּתה ִע ָּמנּו וְ נִ ְׁש ָמ ָעה וְ ַאל־יְ ַד ֵּבר ִע ָּמנּו ֱא‬ ַ ‫אמרּו ֶאל־מ ֶֹׁשה ַּד ֵּב‬ ְ ֹ ‫וַ ּי‬ ִ ‫ל־ה ָעם ַא‬ ָ ‫אמר מ ֶֹׁשה ֶא‬ ֶ ֹ ‫וַ ּי‬ ‫ֹלהים‬ ִ ‫ל־ּת ָיראּו ִּכי ְל ַב ֲעבּור נַ ּסֹות ֶא ְת ֶכם ָּבא ָה ֱא‬ ‫׃‬ ‫יכם ְל ִב ְל ִּתי ֶת ֱח ָטאּו‬ ֶ ֵ‫ל־ּפנ‬ ְ ‫ּוב ֲעבּור ִּת ְהיֶ ה יִ ְר ָאתֹו ַע‬ ַ ‫וַ ּיַ ֲעמֹד ָה ָעם ֵמ ָרחֹק‬ ‫׃‬ ‫ֹלהים‬ ִ ‫ר־ׁשם ָה ֱא‬ ָ ‫ל־ה ֲע ָר ֶפל ֲא ֶׁש‬ ָ ‫ּומ ֶֹׁשה נִ ּגַ ׁש ֶא‬ All the people witnessed the thunder and lightning, the blare of the horn and the mountain smoking; and when the people saw it, they fell back and stood at a distance. “You speak to us,” they said to Moses, “and we will obey; but let not God speak to us, lest we die.” Moses answered the people, “Be not afraid; for God has come only in order to test you, and in order that the fear of Him may be ever with you, so that you do not go astray.” So the people stood at a distance, and Moses approached the thick cloud where God was.

Exodus 20:19–21a are most probably an addition; this can be seen in the Wiederaufnahme of v. 18 (‫ )וַ ּיַ ַע ְמדּו ֵמ ָרחֺק‬in v. 21a (‫)וַ ּיַ ֲעמֹד ָה ָעם ֵמ ָרחֹק‬.55 In the older layer, the people have “seen” the theophany and the Decalogue and move away (v. 18), whereas Moses again approaches God (v. 21b). Only in the later layer do the people ask Moses to serve as mediator (v. 19), and Moses says to the people “Do not be afraid” (v. 20). It can be assumed that the additions in vv. 19–20 are dependent on Deut 5:22–31, where the people “see” and “hear” and where the subject of Moses as mediator is dealt with in detail. In addition, changes seem to have been made within Deut 5 itself in order to connect it with Exod 20. Deuteronomy 5:24–27 reads: 55  See W. O swald , Israel am Gottesberg: Eine Untersuchung zur Literargeschichte der vorderen Sinaiperikope Ex 19–24 und deren historischem Hintergrund (OB 159; Freiburg, Switzerland: Universitätsverlag; Göttingen: Vandenhoeck&Ruprecht, 1998), 98, 101, 258– 259; Berner, “Redaction History” (see n. 33), 381–383. The additions in Exod 20:19–21a may not be from a single hand: Oswald finds a later gloss in v. 20 and Berner in v. 19.

520

Reinhard G. Kratz

ְ ‫ֹלהינּו ֶא‬ ֵ ‫אמרּו ֵהן ֶה ְר ָאנּו יְ הוָ ה ֱא‬ ְ ֹ ‫וַ ּת‬ ‫ת־ּכבֹדֹו וְ ֶאת־ּגָ ְדלֹו וְ ֶאת־קֹלֹו ָׁש ַמ ְענּו ִמּתֹוְך ָה ֵאׁש ַהּיֹום ַהּזֶ ה‬ ‫׃‬ ‫ת־ה ָא ָדם וָ ָחי‬ ָ ‫ֹלהים ֶא‬ ִ ‫ָר ִאינּו ִּכי־יְ ַד ֵּבר ֱא‬ ‫אכ ֵלנּו ָה ֵאׁש ַהּגְ ד ָֹלה ַהּזֹאת‬ ְ ֹ ‫וְ ַע ָּתה ָל ָּמה נָ מּות ִּכי ת‬ ‫׃‬ ‫ֹלהינּו עֹוד וָ ָ ֽמ ְתנּו‬ ֵ ‫ִאם־י ְֹס ִפים ֲא ַ֗נ ְחנּו ִל ְׁשמ ַֹע ֶאת־קֹול יְ הוָ ה ֱא‬ ‫׃‬ ‫ּתֹוְך־ה ֵאׁש ָּכמֹנּו וַ ּיֶ ִחי‬ ָ ‫ֹלהים ַחּיִ ים ְמ ַד ֵּבר ִמ‬ ִ ‫ל־ּב ָׂשר ֲא ֶׁשר ָׁש ַמע קֹול ֱא‬ ָ ‫ִּכי ִמי ָכ‬ ‫ֹלהינּו‬ ֵ ‫ל־א ֶׁשר יְ ַד ֵּבר יְ הוָ ה ֱא‬ ֲ ‫ֹלהינּו וְ ַא ְּת ְּת ַד ֵּבר ֵא ֵלינּו ֵאת ָּכ‬ ֵ ‫אמר יְ הוָ ה ֱא‬ ַ ֹ ‫ל־א ֶׁשר י‬ ֲ ‫ּוׁש ָמע ֵאת ָּכ‬ ֲ ‫ְק ַרב ַא ָּתה‬ ‫׃‬ ‫ֵא ֶליָך וְ ָׁש ַמ ְענּו וְ ָע ִׂשינּו‬ and you said, “YHWH our God has just shown us His majestic Presence, and we have heard His voice out of the fire; we have seen this day that man may live though God has spoken to him. Let us not die, then, for this fearsome fire will consume us; if we hear the voice of YHWH our God any longer, we shall die. For what mortal ever heard the voice of the living God speak out of the fire, as we did, and lived? You go closer and hear all that YHWH our God says, and then you tell us everything that YHWH our God tells you, and we will willingly do it.”

Deuteronomy 5:24b–26 reflects the question under what circumstances humans can remain alive when hearing God. Thus, the peopleʼs fear is ascribed both to the hearing of Godʼs words and to the danger of the fire of theophany. The reflection seems to be secondary, at least in some parts.56 Perhaps its point of departure was v. 25b and the concern that listening to the voice of God would lead to death, which offers a reason for the distance in the original scene of Exod 19:18, 21b and justifies the request for Moses as mediator. In a second step, v. 25a (‫ )וְ ַע ָּתה ָל ָּמה נָ מּות‬introduced a further concern that if Godʼs speaking with a human being would not lead to death (v. 24b) then the fire might consume the people. Verse 26 is a Wiederaufnahme of v. 24b and combines both concerns. In any case, with the additions in vv. 24b–26, which are reflected in the additions in Exod 20:19–21a (‫ֹלהים ֶּפן־נָ מּות‬ ִ ‫)וְ ַאל־יְ ַד ֵּבר ִע ָּמנּו ֱא‬, a (further) bridge was constructed between Exod 20 and Deut 5. Deuteronomy 5:29 (]. . .[ ‫)מי־יִ ֵּתן וְ ָהיׇ ה ְל ָב ָבם זֶ ה ָל ֶהם ְליִ ְר ָאה א ִֹתי‬ ִ is also an addition in the context of Deut 5:28–30. Deuteronomy 5:28–30 ‫אמר יְ הוָ ה ֵא ַלי ָׁש ַמ ְע ִּתי ֶאת־קֹול ִּד ְב ֵרי ָה ָעם ַהּזֶ ה ֲא ֶׁשר‬ ֶ ֹ ‫יכם ְּב ַד ֶּב ְר ֶכם ֵא ָלי וַ ּי‬ ֶ ‫וַ ּיִ ְׁש ַמע יְ הוָ ה ֶאת־קֹול ִּד ְב ֵר‬ ‫׃‬ ‫ל־א ֶׁשר ִּד ֵּברּו‬ ֲ ‫יטיבּו ָּכ‬ ִ ‫ִּד ְּברּו ֵא ֶליָך ֵה‬ ‫׃‬ ‫נֵיהם ְלע ָֹלם‬ ֶ ‫יִיטב ָל ֶהם וְ ִל ְב‬ ַ ‫ּיָמים ְל ַמ ַען‬ ִ ‫ל־ה‬ ַ ‫ֹותי ָּכ‬ ַ ‫ל־מ ְצ‬ ִ ‫ת־ּכ‬ ָ ‫י־יִּתן וְ ָהיָה ְל ָב ָבם זֶ ה ָל ֶהם ְל ְיִר ָאה א ִֹתי וְ ִל ְׁשמֹר ֶא‬ ֵ ‫ִמ‬ ‫׃‬ ‫יכם‬ ֶ ‫ֵלְך ֱאמֹר ָל ֶהם ׁשּובּו ָל ֶכם ְל ָא ֳה ֵל‬ The LORD heard the plea that you made to me, and the LORD said to me, “I have heard the plea that this people made to you; they did well to speak thus. May they always be of such mind, to revere Me and follow all My commandments, that it may go well with them and with their children forever! Go, say to them, ‘Return to your tents.’ ”  See T. Veijola, Das 5. Buch Mose: Deuteronomium Kapitel 1,1–16,17 (ATD 8/1; Göt­tin­ gen: Vandenhoeck & Ruprecht, 2004), 127, 136–137, 146–147; Berner, “Redaction History” (see n. 33), 383–384, 386–387. 56

Reworked Pentateuch and Pentateuchal Theory

521

Verse 29 interrupts the connection between the self-commitment of the people and Godʼs approval (vv. 27–28: ‫ל־א ֶשׁר ִּד ֵּברּו‬ ֲ ‫יטיבּו ׇּכ‬ ִ ‫)ה‬ ֵ and the dismissal of the ֶ ‫)לְך ֱאמֹר ָל ֶהם ׁשּובּו ָל ֶכם ְל ָא ֳה ֵל‬. ֵ 57 The addition of v. 29 forms people (vv. 30–31: ‫יכם‬ a link to Exod 20:20: Godʼs intention to test the people and to bring them to fear God is fulfilled in Deut 5:29. Furthermore, the issue of fearing God is related to the admonition to keep the commandments in the future. From these examples we can draw the conclusion that the beginnings of the development that we observed in the Proto-Samaritan version and in 4Q158 are to be found in the literary history of Exod 20 and Deut 5 themselves. The links between Exod 20 and Deut 5 are gradually added. They find their continuation in the integration of Deut 5 (including also Deut 18) into Exod 20 both in the Proto-Samaritan version and, finally, in 4Q158. This innerbiblical interpretation and harmonization cannot be explained either by the classical Documentary Hypothesis or by the Fragmentary Hypothesis but only by the supplementary or revisional approach. At first glance, it appears as if two independent sources (Exod 20 and Deut 5) were joined, especially if we consider the evidence in the Samaritan version and 4Q158.58 Yet these two texts are already the result of innerbiblical interpretation, and they stand in a literary and compositional relationship with each other. The further revision in the Proto-Samaritan version and 4Q158 is just a continuation of this innerbiblical exegesis.

6 Textual Alterations In addition to the three examples that we have discussed so far, in this section I would like to complete the analysis, focusing on the individual elements of the manuscript 4Q158, according to the three categories proposed by Molly Zahn in her work on 4QRP: minor additions, omissions, and paraphrases.59 We have already discussed some of these issues above: A. The surplus in the blessing of Gen 32 in frags. 1–2, lines 7–10 partly recalls Gen 28:3 and 35:9–12 and partly features its own formulations – recognizable as an addition by Wiederaufnahme. B. The addition of a divine commission to communicate the announcement of the prophet like Moses (Deut 18:18) to the people in frag. 6, lines 5–6 is recognizable as an addition only if compared with the Masoretic Text. C. The introduction for the quotation from Deut 5:30–31 in frags. 7–9, line 3, owing to the rearrangement of Exod 20:18–21, became necessary as a transition from the proclamation of the Decalogue to the dismissal of the  Veijola, Deuteronomium (see n. 56), 127, 138, 144–145.  Cf. Tigay, “Empirical Basis” (see n. 3). 59  See Zahn, Rethinking (see n. 12), 35–37, 57–63; for the categories see 17–19. 57 58

522

Reinhard G. Kratz

people. This is recognizable as an addition because of the anticipation of the speechʼs opening from Exod 20:22a, which is likely to have been taken up again in line 5 (according to the Samaritan version). D. Certain additions produce a correspondence of commission and execution: in frags. 1–2, lines 12–13 (prohibition to eat the tendon of the sciatic nerve); frags. 1–2, lines 15–19 and frag. 4 (reproduction, partial paraphrase, new version of Exod 3:12, and the performance of worship according to the model of Exod 24); and frags. 7–9, lines 4–5 (dismissal and return of the people, recognizable as addition by Wiederaufnahme). Two techniques stand out here: the method of Wiederaufnahme and the correspondence of commission and execution. Both techniques are found in 4Q158 as well as in the Masoretic Text. For the method of Wiederaufnahme we refer to Exod 20:18–21 and for the correspondence of commission and execution to the relationship of Exod 29:19–20 and Deut 5:22–31. Both techniques are widely used in the Hebrew Bible and in many cases point to later reworking. This is almost the consensus for the method of Wiederaufnahme; for the correspondence of commission and execution, we can refer to the Deuteronomistic notes on the fulfilment of prophetic predictions (for instance, in 2 Kgs 9, with regard to 1 Kgs 19 and 21) or to the Priestly writing and its account of the creation in Gen 1, the flood in Gen 6–9, or the building of the sanctuary in Exod 25–40, where this principle is applied on a grand scale. In all cases where these techniques are applied, reworking with or without comparison to the Masoretic Text can be seen. However, the latter is not the case with the small changes and additions of one or two words that Zahn has listed for 4Q158, such as Gen 32:25a (‫וַ ּיִ ּוָ ֵתר‬ ‫ )יַ ֲעקֹב ְל ַבּדֹו‬versus 4Q158 frags. 1–2, line 3 ‫ל]בדו שמה‬ ̊ ‫י]ע ̊ק[ב‬ ֗ ‫ו]י̊ ו֗ [תר‬, “And Jacob remained alone there.”60 Even such small features are often significant and can smooth transitions or answer exegetical questions. Admittedly, in many cases these are purely stylistic or lexical variants, for which we cannot really determine the priority of one or the other variant. Unless they are strikingly different from the rest of the composition in terms of style, vocabulary, or grammar, such changes are almost impossible to identify without knowledge of a second textual witness. In these cases we find ourselves in an area where the borders between literary and textual history are particularly fluid. Almost no omissions can be detected in 4Q158. Due to the length of the lines, Zahn suspects a possible case of haplography at two places (frag. 5, line 3; frags. 10–12, line 8).61 However, the available text does not indicate a deliberate omission anywhere. That we nevertheless have to reckon with this case is shown  Zahn, Rethinking (see n. 12), 35.  Z ahn , Rethinking (see n. 12), 57–58. Another possible shortening or omission is suggested by Segal, “Exegesis” (see n. 12), 49–50, n. 16 for the text of Exod 24:4–6 in frag. 4, lines 4–5. 60

61

Reworked Pentateuch and Pentateuchal Theory

523

by other examples where more text is available, such as the excerpts of Samuel– Kings in Chronicles. Without a second textual witness, however, this means of composition and reconstruction cannot be proven and should, therefore, be used only sparingly when formulating a theory. It is equally difficult to reconstruct a Vorlage in cases of free paraphrasing. This is the case with 4Q158 in frag. 14 and also perhaps in frag. 3. The vocabulary in frag. 14 shows numerous points of contact with Exod 6:3–8 and Exod 15. The piece is possibly written in the context of Exod 6, as a proclamation of the victory over Egypt and the conquest of the promised land, which is described in Exod 15.62 In order to assess the extent to which we are dealing with a reworking and how to explain it, we need to have more text. The same is true for frag. 3, which may allude to Gen 49:1 or Gen 47:29–30.63 Nevertheless, it is indisputable that we have to take free paraphrasing into account in the reworking process. In itself, such a paraphrase appears – depending on the circumstances – as original material, part of a source, as a fragment, or as a free, independent formulation. However, if the reference texts (here Exod 6 and 15) or the Masoretic version are known, then the paraphrase proves itself to be a reformulation and part of a revision.

7 Conclusion Our subject in this paper was the possible implications of the 4QReworked Pentateuch manuscripts for the pentateuchal question. Looking in both directions, to 4Q158 as well as the Masoretic Text, I have to say that I cannot see any difference in their compositional strategies; on the contrary, I see a great similarity in terms of compositional techniques as well as content. As far as the Reworked Pentateuch manuscripts are concerned, we are dealing with “biblical” manuscripts, and so these texts are in every respect comparable to what we have in the Masoretic or the Sarmaritan Pentateuch. Even if we were to classify the Reworked Pentateuch manuscripts and the Samaritan Pentateuch as a kind of Rewritten Bible or Rewritten Scripture (the same sort of literature as, for example, Chronicles, Jubilees, the Temple Scroll, etc.), the comparability is given, not least since the Pentateuch itself seems to be a product of permanent rewriting of older traditions. This result has, of course, enormous implications for the pentateuchal question and the different hypotheses to answer it. I see two primary implications: First, on the methodological level, the text is an example of various compositional techniques and models for how a text (at least a narrative text,  Segal, “Exegesis” (see n. 12), 54–55; Zahn, Rethinking (see n. 12), 59–62.  S trugnell , “Notes” (see n. 15), 170; S egal , “Exegesis” (see n. 12), 53–54; Z ahn , Rethinking (see n. 12), 62–63. 62

63

524

Reinhard G. Kratz

but I think also texts of other genres) has grown within the biblical tradition. Second, on the historical level, the Reworked Pentateuch manuscripts, like all other versions (MT, SP, LXX, Rewritten Bible), are witnesses of the same text and part of the same literary and textual history. They progress further along the path, which begins in the literary and redactional history of a biblical text and indicates the direction in which we must go to trace the history of extension, supplementation, and revision – in short, Fortschreibung – and exegesis of the Pentateuch. Here, as well as there, the Supplementary or Revisional Hypothesis (Fortschreibungshypothese) has proven itself an appropriate model that is superior in explanatory power to the Source or Documentary Hypothesis.

Holy Seed Ezra 9–10 and the Formation of the Pentateuch Richard J. Bautch This study of Ezra–Nehemiah and the development of the Pentateuch focuses on the fourth century BCE. Building on the work of David Carr,1 I begin with certain presuppositions about the book of Ezra and the formation of the Pentateuch in this timeframe. Regarding the Pentateuch, I take it that P and non-P material were first combined into a Hexateuch, which preceded the formation of the Pentateuch. Specifically, biblical writers integrated P into a narrative of origins, featuring the Deuteronomic law, or D. Carr refers to this critical stage in the collectionʼs growth as “the formation of a combined P/non-P Hexateuch.”2 Subsequently, the writers sought support for this combination of the D-based account of Israelʼs origins and the previously separate and distinct Priestly narrative. The formation of this comprehensive document, the combined P/ non-P Hexateuch, quite likely dates to the first half of the fourth century. The fourth century was also a critical juncture in the textual history of traditions associated with the figures of Ezra and Nehemiah; it was during this time that Ezra–Nehemiah began to take shape as a unified account of the golah community in Yehud. Carr identifies an Ezra narrative that begins with the return from exile and follows with the rebuilding of the temple and the reading of the Torah, much like 1 Esdras (discounting 1 Esd 3:1–4:63).3 In MT Ezra–Nehemiah, these events occur in Ezra 1–6, 7–8, and 9–10, with Neh 8–9 understood to have been originally placed between Ezra 8 and 9. The Ezra narrative may be dated to the latter half of the Persian period or the early Hellenistic period, that is, the fourth century. 1  D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011), 204–225. At the conference “The Pentateuch within Biblical Literature: Formation and Interaction” (Jerusalem, May 2014), David Carr responded to this essay, and in fact we were in dialogue about the issues addressed here before, during, and after the conference. I am grateful to David for his insights and constructive suggestions, and to Bernard Levinson, who, as an editor of this volume, has made queries and recommendations that greatly enhanced this study. 2  Carr, Formation (see n. 1), 219. 3  Carr, Formation (see n. 1), 207. In his analysis, Carr refers to this text as the Rebuilding-Ezra Narrative; here it is referred to as the Ezra narrative.

526

Richard J. Bautch

Given these presuppositions, in this study I first draw attention to the fact that the combined P/non-P Hexateuch and the Ezra narrative are contemporaneous and that both constellations of traditions had their formative period late in the Persian era, in the early to mid-fourth century. Dating the Ezra narrative to the fourth century places it alongside the combined P/non-P Hexateuch in its seminal stage, and vice versa. Put differently, accounts of Ezra the scribe interacting with the Torah crystallized as the forerunner to the Pentateuch, the Hexateuch featuring D and P, was being structured.4 This coincidence explains, among other things, how words and expressions drawn from the book of Deuteronomy and the Priestly law are routinely conjoined in a single discourse, the Ezra narrative (Ezra 6:12, 18; 9:1, 4; Neh 8:14).5 An example of this phenomenon is the Sukkot legislation about living in booths during the seventh month, which is presented as a law that God commanded to Moses (Neh 8:14, drawing on Lev 23:34, 42, as well as Deut 16:13). In the second part of this study, I ask: What is the literary relationship between the combined P/non-P Hexateuch and the Ezra narrative? To state the obvious, there are textual correlations between a developing form of what would become the MT Pentateuch and the books of Ezra and Nehemiah. In Neh 8:1, Ezra reads “the book of the law of Moses” (‫ )את ספר תורת משה‬to the Judeans gathered in Jerusalem. Later, in 8:8, 18, Ezraʼs text is referred to as “the book of the law of God” (8:8, 18, ‫)ספר תורת האלהים‬. Another reference occurs in Ezra 7:26, “the law of Ezraʼs God” (‫)דתא די אלהך‬. In addition, there are exegetical passages that adopt and adapt commands described “as written in the Torah” (Ezra 6:18; 10:3; Neh 10:35). What do these references indicate about the development of the Torah? Do the books of Ezra and Nehemiah document the completion of the Pentateuch? I argue, rather, that expressions such as “the book of the law of Moses” refer to the preeminent traditions to be included in the Pentateuch rather than the completed document. The Torah read in Neh 8 is in the process of becoming a comprehensive document comprising the major legal traditions of the Pentateuch. To gain a better understanding of these textual processes, the third and fourth sections of this study take as their point of departure the fact that the Ezra materials describe the Israelite ethnos metaphorically as the “holy seed” but, interestingly, the combined P/non-P Hexateuch does not.6 Consequently, I investigate the holy-seed concept attested in Ezra 9:2 to show how the Ezra  On interaction between the figure of Ezra and Hexateuchal processes in the fourth century, see E. Otto, Mose: Geschichte und Legende (Munich: Beck, 2006), 60. 5  D. V. Edelman et al., Opening the Books of Moses (Sheffield: Equinox, 2012), 99. See also R. J. Bautch, Developments in Genre Between Post-Exilic Penitential Prayers and the Psalms of Communal Lament (SBL AcBib 7; Leiden: Brill, 2003), 125–135. 6  Of course, to assume that the Priestly writer should employ the designation holy seed for Israel is to argue circularly. I mitigate this point, however, by demonstrating below that (1) that among biblical writers, seed language was highly pervasive, with seed as common as terms 4

Holy Seed

527

narrative addresses the golah community in rhetorical terms and exhorts it to endogamy. There are two critical findings. First, although the endogamy campaign in Ezra 9–10 adopts the extreme measure of forcible divorce, in other respects it is not exceptional, and its signature reference to the Israelite ethnos, holy seed, joins together two common ideas in a relatively straightforward manner. Second, Leviticus combines legislation about distinguishing between clean and unclean animals with a divisive polemic against the nations whom God has driven from the land because of their abhorrent practices (Lev 20:22–26). When the passage urges readers to eschew the nations and practice endogamy, it echoes Ezra 9–10, and it shares with the Ezra text a range of literary features. Leviticus 20:22–26 does not, however, share the language of holy seed, even though other portions of Leviticus (e.g., 11:44–45) draw upon on the concept of Godʼs holy people as it is found in Deuteronomy. I conclude this study by offering a possible explanation for the absence of the term and concept holy seed in Leviticus.

1 Developments in the Fourth Century There are two major legal traditions in the Pentateuch. One comprises Deuteronomy and the Deuteronomic school of thought as it developed in Yehud, or postexilic Judah, during the fifth and fourth centuries. At this time, D is a distinct stratum of legal material rooted in the book of Deuteronomy and attested as well within the larger narrative of origins that encompasses the patriarchs, Moses and the exodus, and the entering of the land.7 On the question of a more theological Deuteronomistic (Dtr) dimension in the Pentateuch, increasingly scholars are identifying Deuteronomistic redactional texts in the books of Genesis through Numbers, but this endeavor has yet to yield a consensus and has no application for the present study of pentateuchal materials.8 The other tradition is the P such as Israel, and (2) that the writer responsible for Lev 20:22–26 adopted expressions and phrases from outside his ken and incorporated them into Priestly legislation. 7  Thus, the language and legal principles of D exert influence on pentateuchal materials outside of D (or any other identifiable strata) and have a presence across the larger narrative. See R. G. K ratz , “The Pentateuch in Current Research: Consensus and Debate,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Mohr Siebeck: Tübingen, 2012), 31–61, here 37, 47. 8  See R. F. P erson Jr. and K. S chmid , introduction to Deuteronomy in the Pentateuch, Hexateuch, and the Deuteronomistic History (ed. R. F. Person Jr. and K. Schmid; FAT 56; Tübingen: Mohr Siebeck, 2012), 1–7, here 3; E. O tto, Das Deuteronomium im Pentateuch und im Hexateuch (FAT 30; Tübingen: Mohr Siebeck, 2001); O. H. S teck , Israel und das gewaltsame Geschick der Propheten: Untersuchungen zur Überlieferung des deuteronomistischen Geschichtsbildes im Alten Testament, Spätjudentum und Urchristentum (WMANT 23; Neukirchen-Vluyn: Neukirchener Verlag, 1967). In contrast to the Pentateuch, Dtr theology does influence the other primary text in this study, Ezra 9–10, which contains a

528

Richard J. Bautch

or Priestly tradition as it existed in the second half of the Persian period.9 P included a distinct strand of Holiness teaching, or H material, that had been joined to it at a point relatively early in the Persian period, before the P tradition was integrated into a larger narrative of Israelʼs origins.10 The integration of P into the narrative of origins, featuring D, is especially revealing. This is the development referred to above as “the formation of a combined P/non-P Hexateuch.” Prior to this point the narrative of origins encompassed D in the form of Deuteronomy and may well have concluded with the book of Joshuaʼs account of settling the land. Joshuaʼs covenant at Shechem (Josh 24:25) is the pinnacle of this narrative sequence, which was designed to rival the founding legends of other nations. The combination of this (non-P) Hexateuch of Israelʼs origins and the then-separate and distinct Priestly narrative dates to the first half of the fourth century, and the indications of this are multiple. Carr points out that in postexilic texts such as the Nehemiah memoir, Haggai–Zechariah, and Trito-Isaiah, pentateuchal traditions are attested piecemeal and fragmentarily or not at all; the first half of the Persian period provides no evidence of a narrative of origins that consolidates the P and D traditions.11 By the end of the Persian period and the beginning of the vision of Israel and its history shaped in large part by the precepts of Dtr. The impact of Dtr theology on Ezra 9–10 is elaborated elsewhere in this essay (see n. 40). 9  Like D, P influenced other writings that are now found in the Pentateuch but are not part of a distinguishable tradition (so-called non-D/non-P texts). See Kratz, “Consensus and Debate” (see n. 7), 37; J. Stackert, Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007), 14–16. 10  Carr defines H as “a later expansion of that [P] material, an expansion characterized in part by the coordination of P material with Deuteronomic and Covenant Code materials now found in Exod 20–23 and Deuteronomy.” He adds that this coordination “preceded the coordination of P with non-P materials,” i.e., the integration of P into the Pentateuch. See D. M. C arr , “Scribal Processes of Coordination/Harmonization and the Formation of the First Hexateuch(s),” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen; Mohr Siebeck, 2012), 63–83, here 72. 11  C arr , Formation (see n. 1), 212. For a discussion of the Nehemiah memoir not citing Torah, see D. P olaski , “Nehemiah: Subject of the Empire, Subject of Writing,” in New Perspectives on Ezra–Nehemiah: History and Historiography, Text, Literature and Interpretation (ed. I. Kalimi; Winona Lake, IN: Eisenbrauns, 2012), 37–59, here 41. Regarding the postexilic prophets, see, for example, Hag 2:10–14, esp. v. 11, where Haggai requests an instruction (‫ )תורה‬that speaks to a certain sacrificial rite, and Zech 7:12, where Zechariah refers to a distinct form of prophetic revelation with the word ‫תורה‬. These and other contemporary references to “torah” suggest that at this time there was no whole greater than the sum of the parts. In the case of Deutero- and Trito-Isaiah, the prophet cites cultic and patriarchal traditions attested in the pentateuchal narrative but does so in a disconnected way that gives no indication of a collection of traditions with a narrative sweep. For descriptions of pentateuchal themes and motifs in the last twenty-six chapters of Isaiah, see U. Berges , “Kingship and Servanthood in the Book of Isaiah,” in The Book of Isaiah: Enduring Questions Answered Anew (ed. R. J. Bautch and J. T. Hibbard; Grand Rapids, MI: Eerdmans, 2014), 167–190.

Holy Seed

529

Hellenistic period, however, there are works such as 1–2 Chronicles that are based in part upon a combined P/non-P Hexateuch.12 Chronicles, it is generally agreed, is a fourth-century text.13 The genealogies in 1 Chr 1–9 are derived from the genealogies in Gen 1–9, and despite material differences between the sets of genealogies, they give the unavoidable impression that the Chronicler knows the Genesis genealogies and is utilizing them as a source.14 Thus, the late fourth century provides the terminus ad quem for the textual formation of a combined P/non-P Hexateuch, whereas the terminus a quo of this narrative is the beginning of the fourth century. In dating the Ezra narrative to the fourth century, Carr notes, “the positive depiction of the Persian role in the return, rebuilding and institution of Torah in the Ezra narrative may suggest a dating still during the Persian period, since later, post-Persian-period scribes probably had little reason to assert so strenuously that the Persians had supported each stage of the rebuilding process.”15 Carr is not alone in dating the composition of the Ezra narrative to the fourth century.16 The question depends, to a degree, upon oneʼs evaluation of the sources embedded in the Ezra narrative, such as the Aramaic letters in Ezra 5–6 and the memoir in Ezra 7–10. Bob Beckingʼs analysis of the sources and their authenticity leads him to conclude that “the organization of the past in the present book of Ezra has to be regarded as the authorʼs interpretation of the past.”17 That is, if the sources are putative and not historical, they themselves do not require a dating 12  Because Chronicles is not an exegetical work per se, it is important to clarify its relationship to pentateuchal materials. The issue is beyond the present study, but see B. M. Levinson, Legal Revision and Religious Renewal in Ancient Israel (Cambridge: Cambridge University Press, 2008), 176–181. 13  See S. L. McKenzie, 1–2 Chronicles (AOTC; Nashville: Abingdon, 2004), 31–32, who concludes that “citations and allusions to other biblical materials push the date of Chronicles well past 400 BCE.” 14  Japhet observes Chroniclesʼs “strict adherence to its source in Genesis.” See S. Japhet, I & II Chronicles (OTL; Louisville: Westminster John Knox, 1993), 56. Moreover, there is the case of 2 Chr 35:13, which describes the roasting of the Passover lamb with fire as well as a boiling of the offerings in pots, caldrons, and pans. It is noteworthy that the lamb is both roasted over fire “according to the ordinance” and boiled before serving. The P source stipulates roasting (Exod 12:8–9), whereas the D source calls for boiling the lamb (Deut 16:7). In Chronicles, the redundant process of preparation indicates that the Chronicler knew both the D and P legal traditions and sought to harmonize the two, because those two sources had by this point both been included in the latest version of a unitary narrative of origins (see Japhet, I & II Chronicles, 1052–1053). 15  Carr, Formation (see n. 1), 208. 16  See B. Becking, Ezra, Nehemiah, and the Construction of Early Jewish Identity (FAT 80; Tübingen; Mohr Siebeck, 2011), 28–29; L. Grabbe, “Reconstructing History from the Book of Ezra,” in Second Temple Studies, Vol. 1: Persian Period (JSOTSup 117; Sheffield: Sheffield Academic Press, 1991), 98–106. 17  Becking, Ezra, Nehemiah (see n. 16), 29.

530

Richard J. Bautch

of the text earlier than the fourth century. The only certainty is that the textual world of Ezra postdates Ezraʼs mission by a considerable span of time.18 The Ezra narrative may thus be considered historiography that was composed, in all likelihood, in the fourth century. Dating the Ezra narrative to the fourth century locates it squarely alongside the combined P/non-P Hexateuch in its formative period. The synergy between these two textual processes would extend forward from the fourth century and result in what Alexander Fantalkin and Oren Tal refer to as the “canonization of the Torah”: “The canonization of the Torah – or better, the proto-canonization under the authority of Ezra and his circle – should not be considered as the final redaction of the Torah that miraculously survived to these days, but rather as an initial attempt at canonization, with certain modifications made after Ezraʼs time.”19 Although I would neither factor the historical Ezra into a synoptic study of the P/non-P Hexateuch and the Ezra narrative nor use the term “canonization” to describe this process, I agree with Fantalkin and Tal that the gathering of the different law codes and narratives into a single collection that would have five parts commenced in the fourth century, precisely when the Ezra narrative began to assume the form in which we now have it.

2 Ezraʼs Relationship to the Pentateuch The fact that the combined P/non-P Hexateuch and the Ezra narrative were contemporaneous invites a new perspective. As Fantalkin and Tal imply, there was formerly an assumption that the Pentateuch was compiled in exile and was brought to Yehud by Ezra.20 In this reconstruction, the Pentateuch resembles a great figure in Israelite history and Ezra its firstborn son, who inherited a double  On dating the mission, the figure of Ezra is said to leave Babylon in the seventh regnal year of the Persian king Artaxerxes (Ezra 7:7). The reference could be to either Artaxerxes I (seventh year = 458 BCE) or Artaxerxes II, whose seventh year on the throne was 398 BCE. Scholars have defended both dates, and their explanations often draw on the plausibility of Ezra preceding or following Nehemiah, whose arrival in Yehud from Babylon is securely dated to 445 BCE. The temporal question about Ezraʼs mission remains open. 19  A. Fantalkin and O. Tal , “The Canonization of the Pentateuch: When and Why (Part 1),” ZAW 124 (2012), 1–18, here 3. 20  This view was expressed both in scholarship and especially in histories written for nonspecialists, such as B. W. A nderson , Understanding the Old Testament (Englewood Cliffs, NJ: Prentice Hall, 1957), 456. According to Anderson, “The ‘book of the law’ that Ezra read was the Pentateuch, as edited by priests during the Babylonian exile. [. . .] Ezraʼs greatest contribution was to establish the Pentateuch as the authoritative canon for Jewish faith and practice.” Other studies arriving at this conclusion included G. Widengren, “The Persian Period,” in Israelite and Judean History (ed. J. H. Hayes and J. M. Miller; Philadelphia: Westminster, 1977), 536; J. A. Sanders, Torah and Canon (Philadelphia: Fortress, 1972), 51. 18

Holy Seed

531

portion of the family wealth, here in the form of revealed laws and traditions. Given, however, the simultaneity of the Ezra narrative and the combined P/ non-P Hexateuch, their relationship calls for a rather different familial metaphor. In light of current scholarship, the metaphor should change from parent and child to close siblings, such as Leah and Rachel. As close siblings, Leah and Rachel had much in common, and so it is with the combined P/non-P Hexateuch and the Ezra narrative. Materials taken primarily from two major sources, D and P, were fused together to form, on the one hand, the P/non-P Hexateuch, and on the other hand the Ezra narrative. Although different scribes worked on the respective text blocks, their parallel content (see the example below of Lev 20:22–26 // Ezra 9:1–15) is more comparable than has previously been appreciated because it arose at more or less the same time and in more or less the same place. Strictly from the perspective of textual formation, the P/non-P Hexateuch and the Ezra narrative are siblings of a sort. The implications of a synchronous relationship between the combined P/ non-P Hexateuch and the Ezra narrative are well worth considering. Among other things, this provides the grounds to reconsider the design and contours of the highly rhetorical campaign to ban intermarriage described in Ezra 9–10.21 In this episode, Ezra and his supporters tell other Judeans that if they have intermarried with the “peoples of the land,” they must divorce their spouses, who will then be banished from the community. The language of “peoples of the land” is rhetorical; it is an attempt at othering individuals who were themselves Judeans and fellow worshippers of Yahweh.22 The issue with these spouses was that they were not part of the community that came back from exile, or they were part of an earlier wave of returnees. In either case, their religious practices were not identical to those of the returnees with Ezra. Because they were different, this other Yahwistic group was deemed a threat and rejected. It is not clear that Ezraʼs group of returnees followed through with the campaign of forced divorce and banishment; the text itself is inconclusive, especially Ezra 10:44.23  The literature on intermarriage in Ezra 9–10 is voluminous. For a cross section of current research, see the essays in C. F revel (ed.), Mixed Marriages: Intermarriage and Group Identity in the Second Temple Period (LHBOTS 547; New York: T&T Clark, 2011). For present purposes, especially pertinent are the contributions of J. Pakkala (“Intermarriage and Group Identity in the Ezra Tradition [Ezra 7–10 and Neh 8],” pp. 78–88) and G. N. Knoppers (“ ‘Married into Moab’: The Exogamy Practiced by Judah and his Descendants in the Judahite Lineages,” pp. 170–91). 22  On the ethnicity of the women referred to in Ezra 9–10, see B. Becking, Ezra, Nehemiah (see n. 16), 71–73. 23  After the long list of those from “Israel” who had married foreign wives, Ezra 10:44 ends with a statement. A literal translation of the MT reads: “And there were from them wives, and they put sons.” Some commentators hold that the Hebrew text is corrupt, and they follow 1 Esd 9:36: “they sent them away with their children.” All other textual witnesses, however, including the ancient Greek, leave some ambiguity as to how these events transpired. The ambiguity is 21

532

Richard J. Bautch

The campaign of forced divorce may have been a trial balloon that burst but is nonetheless in the historical record and part of the canon. From a pentateuchal perspective, it is worth noting that the chapters about intermarriage in Ezra (Ezra 9–10) engage both D and P materials to make clear the authorʼs awareness of the two preeminent legal traditions found in the Pentateuch. In Ezra 9–10, the legislation against Judeans marrying others borrows from the D legal source, or Deuteronomy, and more is said about this below in section 4.2. At this point I will focus rather on how the Ezra narrative utilizes a range of Priestly precepts and norms associated with Leviticus in order to champion endogamy. Clustered in the Holiness section of Leviticus, specifically chapters 18–22, are a number of expressions also found in Ezra. In terms of language, there are multiple correlations between Ezra 9–10 and Leviticus 18–22, and the more salient of these are catalogued in the appendix to this essay.

3 Textual Correlations between Ezra 9–10 and Leviticus 20:22–26 The textual correlations between Ezra 9–10 and Leviticus 18–22 can be divided into two types. In some cases the H material was combined with P well before the writing of Ezra. Here the Ezra narrative draws on the H text as a source and so is dependent on it, literarily. In other cases the Leviticus text containing H material comes from the hand of a fourth-century editor and thus is temporally closer to the Ezra narrative. In these cases, H material predominates the given section of Leviticus, with D phrasing sometimes attested as well.24 As a result, reflected at other points in the narrative as well; for example, Ezra 10:19 states that the priestsʼ descendants agreed to dismiss their wives, although there is no indication of such a pledge by all the others on the list. These and related difficulties in Ezra 10 preclude any determination as to whether the Judeans actually sent away their foreign wives and children. Sara Japhet, in fact, thinks it highly unlikely: “The consideration of the matter in historical terms, against the actual social and historical background, raises severe doubts whether the implementation of such a resolution was at all possible. [. . .] We are dealing with heads of households, people committed by marriage contracts and family ties; how could they ignore them and breach them? Where would the expelled women and children go? What arrangements would be made for their support? [. . .] All these considerations lead to one conclusion: that the resolution remained ‘on paper’ and was never implemented.” See S. Japhet, “The Expulsion of the Foreign Women (Ezra 9–10): The Legal Basis, Precedents, and Consequences for the Definition of Jewish Identity,” in “Sieben Augen auf Einem Stein” (Sach 3,9): Studien zur Literatur des Zweiten Temples; Festschrift für Ina Willi-Plein zum 65. Geburtstag (ed. F. Hartenstein and M. Pietsch; Neukirchen-Vluyn: Neukirchener Verlag, 2007), 141–161, here 155. 24  Christophe Nihan distinguishes further these two types of H-based texts found in Leviticus; one type exhibits no contact with D and may even offer viewpoints contrary to D, whereas the other type provides a combination and at times even a synthesis of H and D elements. See C. Nihan, “The Laws about Clean and Unclean Animals in Leviticus and Deuteronomy and

Holy Seed

533

there is no literary dependence between Ezra and Leviticus but rather the parallel, independent adaptation of shared material. The compositional stage in Leviticus lines up with Ezraʼs compositional stage, and this is the phenomenon referred to metaphorically as “close siblings.” A striking example of this temporal alignment is Lev 20:22–26,25 an exhortation set within a series of prohibitions. Leviticus 20:22–26 lines up with Ezra 9:1–2 in multiple ways, and there are three Priestly expressions in particular that the texts share. Both texts feature the verb ‫בדל‬, “to set apart”: in Lev 20:24, God, referenced in the first person, has set the addressees apart (‫ )הבדלתי‬from the peoples (‫)מן העמים‬, and in Ezra 9:1, the people of Israel have not separated themselves (‫ )לא נבדלו‬from the peoples of the land (‫)מעמי הארצות‬. Leviticus 20:25 supplies the motive clause for separating from the peoples: lest you draw abomination upon yourselves (‫ולא תשקצו את‬ ‫)נפשתיכם‬. Ezra 9:1 also refers to the abominations of the peoples of land but with a different Hebrew expression (‫)כתועבתיהם‬. Leviticus 20:26 concludes, “you shall be holy (‫ )קדשים‬to me,” whereas Ezra 9:2 inverts this idea: “the holy seed (‫ )זרע הקדש‬they have mixed with the people of the land.” Since these two passages have no verbatim phrases in common, literary dependence is unlikely. Nonetheless, the alignment of so many ideas suggests a parallel, independent adaptation of shared material. Lev 20:22–26 and Ezra 9:1–2 were written in the same place at about the same time, the fourth century, for essentially the same reason, to maintain boundaries between social groups. The next portion of this essay will focus on the relationship between Lev 20:22–26 and Ezra 9–10. The key to this inquiry is the metaphor of the Israelite ethnos as a “holy seed” (‫ )זרע הקדש‬in Ezra 9:2. The term holy seed provides a case study that illustrates the relationship between the book of Ezra and the Pentateuch, specifically Leviticus. As shown above, the term holy seed in Ezra 9:2 aligns with the command to “be holy” in Lev 20:26, and it resonates across the Holiness portion of the combined P/non-P Hexateuch. What more can be said about this concept of holiness?

4 Case Study: “Holy Seed” in Ezra 9–10 and Leviticus 20:22–26 The root ‫ קדש‬appears ten times in the Holiness Code, chapters 17–26 of Leviticus, and it seems that the writers/redactors of Leviticus and those responsible Their Place in the Formation of the Pentateuch,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 401–432, here 429–430. 25  That is, Lev 20:22–26 is among the editorial insertions that postdate the incorporation of H into P (see n. 10). On the basis of Lev 20:24 and its expression “a land flowing with milk and honey,” discussed below (see nn. 42, 43), Lev 20:22–26 may be dated to the period of the combined P/non-P Hexateuch, or the early fourth century.

534

Richard J. Bautch

for the Ezra narrative shared a common model of holiness. For the latter writers, Israel is said to be a thing holy in the cultic sense, a sacred seed whose admixture with foreigners is ‫( מעל‬Ezra 9:2). Clearly, Ezraʼs writers use cultic language of the type found in the Holiness Code to talk about Israel as a holy seed. The application of holiness to the noun seed is more challenging to explain. Katherine Southwood posits a connection between the holy seed mentioned in Ezra 9:2 and the agricultural laws against mixing different types of seeds that are stipulated in the Holiness Code (Lev 19:19).26 The concepts of seed and holiness, however, need not be derived from Leviticus uniquely. The concept of seed, like that of holiness, was broadly cast throughout the literature of the late Persian period. It is more likely that the comparison of the Israelite ethnos to a holy seed in Ezra 9:2 occurred by joining two terms already in common parlance. Likening the golah community to a holy seed was not especially innovative,27 and one might ask why a passage such as Lev 20:22–26, where God commands separation from the peoples of the land, has so many terms and concepts in common with Ezra (holiness, separation and mixing, uncleanness and impurity, iniquity and guilt, prohibition of intermarriage) but not the term holy seed. The question becomes acute in light of the contemporaneity of the combined P/non-P Hexateuch and the Ezra narrative. Focusing more closely on the concept of holy seed in Ezra 9:2 allows one to probe the termʼs usage28 and to further compare Ezra 9–10 with Lev 20:22–26. 26  Southwood argues that Ezra 9:2 is able to “define the boundaries of who constitutes the ethnos through reapplying agricultural injunctions.” K. E. S outhwood , Ethnicity and the Mixed Marriage Crisis in Ezra 9–10: An Anthropological Approach (Oxford: Oxford University Press, 2012), 129. 27  This finding goes against the grain of the burgeoning interest in Ezra 9–10 from a social-scientific viewpoint. Because the contents of these two chapters serve to illustrate a variety of sociological and anthropological concepts, there has been extensive analysis of the narrative and a privileging of the concept of holy seed as that which galvanized various parties in their opposition to exogamy. Christine Hayes holds that “the holy seed rationale of Ezra supports a universal and permanent prohibition of intermarriage.” See C. Hayes, “Intermarriage and Impurity in Ancient Jewish Sources,” HTR (1999), 3–36, here 13. As attributed to Ezra, the term holy seed is not insignificant, but its value is primarily as a marker that distinguishes and differentiates Ezra 9–10 from contemporary legal material prohibiting intermarriage, such as Lev 20:22–26 or Exod 34:6–16. 28  Hung-Chuan Lai, in his dissertation on the concept of seed in Ezra and Trito-Isaiah, has argued that this metaphor is a key to the rhetoric of genealogy in Ezra 9–10. That is, the vision of a golah community with highly circumscribed boundaries erected by Ezra pivots rhetorically on the concept of a holy seed. Among the many identity markers embedded in Ezra 9–10, “perhaps the most obvious is golah lineage, on which is conferred a holy quality (‫)זרע הקדש‬. The trauma of displacement can fashion a new group consciousness founded on the lineage of diasporic Judeans rather than on the general Judean descent.” H. C. L ai , “Reconceiving Seed, Redefining Foreignness: Markers of Judean Identity in Ezra–Nehemiah and Trito-Isaiah” (PhD diss., Union Theological Seminary, 2014), 239.

Holy Seed

535

How can the datum of the holy seed, attested in Ezra 9:2 but not in Lev 20:22–26, serve to differentiate these two texts in light of their temporal proximity? As a means to address this, I pose two focal questions related to the concept of the holy seed. The first question asks where the writer of the Ezra narrative might have encountered this term. Put differently, how commonly did seed denote social entities in the fourth century? The second question asks how the D material that informs the concept of the holy seed is incorporated into this portion of the Ezra narrative. 4.1 How Commonly Did “Seed” Denote Social Entities? Ezra 9:2 (NRSV) reads: “For they have taken some of their daughters as wives for themselves and for their sons. Thus the holy seed (‫ )זרע הקדש‬has mixed itself with the peoples of the lands, and in this faithlessness the officials and leaders have led the way.” Whence the notion of “seed” (‫ ?)זרע‬No one should assume that Ezraʼs term seed originated in Hʼs legislation on agriculture. Seed (‫ )זרע‬is attested 222 times in the Hebrew Bible, with almost half of these references in the Pentateuch (Gen 57; Exod 6; Lev 22; Num 8; Deut 14). Across the Hebrew Bible, ‫ זרע‬has four connotations: an agricultural commodity (Gen 47:19, 23; Num 24:7; Deut 28:38), virile semen (Num 5:28; Lev 22:4; Jer 31:27), offspring or descendants (Gen 3:15; 12:7; 16:10; Josh 24:3), and an entity distinguished by its moral excellence or turpitude (Isa 61:9; 65:9, 23).29 The final category includes Ezra 9:2, where the “seed” has mixed with the peoples of the land, and a number of other exilic or postexilic texts (Isa 1:4; 6:13; 57:3; 61:9; 65:23; Prov 11:21; Mal 2:15). The data suggest that when Ezra 9:2 was written into the Ezra narrative, the term ‫ זרע‬could be used metaphorically, and further analysis bears this out. Comparable to the “holy seed” in Ezra 9:2, the “blessed seed” is a motif in Trito-Isaiah (61:9; 65:23).30 The “blessed seed” is described as a covenant partner with Yahweh (61:8–9) and as the children of the blessed people (65:22–23). It is surely this “blessed seed” that forms the community standing with the Lord in the new heavens and new earth at the close of the book of Isaiah (66:22–24). In the eschatological framework of Isa 66, the seed (66:22) is a hybrid of Judeans and converted gentiles (66:18–21). Thus, Trito-Isaiah supports ethnic and religious interaction between Judeans and foreigners, with non-Israelites even enjoying access to the cult in Zion (56:7; 60:7). Although Trito-Isaiahʼs 29  This fourfold typology is standard in the lexica (see BDB 282–283) and allows distinctions to be made between attestations of seed that are materially different. For example, although Lev 20:2 speaks of seed metaphorically, the reference clearly belongs to the third category, offspring or descendants, and thus it has no bearing on the present discussion of seed as a moral entity (the fourth category). 30  This writer also speaks generically of “seed” (59:21, 65:9, 66:22) as well as the “evil seed” (1:4) and the “seed of lies” (57:4).

536

Richard J. Bautch

depiction of non-Israelites contains ideological tensions,31 the notion of seed therein is by and large inclusive, which is quite different from Ezra. Elsewhere in Isaiah, Isa 6:13 applies the figure of a tree stump to certain returnees from exile and concludes by calling them a “holy seed.” This is the only time, other than Ezra 9:2, that the exact expression holy seed, or ‫זרע קדש‬, occurs in the Hebrew Bible. Commentators such as Joseph Blenkinsopp read Isa 6:13 as a late addition by a specific group of returnees; these returnees are “anxious to remain apart from the native population of the province.”32 The figure of the tree stump may thus refer to a group not unlike that of Ezra, inasmuch as the Isaian tradent says of this remnant, “Its stump is the holy seed.” At the same time, Isaiahʼs “holy seed” is not identical to that of Ezra, in that it likely postdates the Ezra narrative and so is contemporary with more polemical uses of holy seed, such as those found in Jubilees.33 Jeremiah 33 is a text contemporaneous with the Ezra narrative and similarly conspicuous in its use of ‫ זרע‬to construct identity for certain groups, possibly kinship units, prominent in Yehud. In Jer 33:14–26, God, speaking through the prophet, promises to raise up a Davidic heir outstanding in righteousness (33:15–16). The promise of a Davidic ruler on the throne is matched by an assurance that there will be Levites offering sacrifice continually (33:17–18). Subsequently, an allusion to the stars and sand of the Abrahamic covenant (33:22a) is followed by a promise to multiply the “seeds” (‫ )זרעים‬of David and of the Levites who minister to God (33:22b). The Davidic group is again referred to as “seed,” but with a wrinkle, in 33:26a (NRSV): “Would I reject the offspring (‫ )זרע‬of Jacob and of my servant David and not choose any of his descendants (‫ )זרע‬as rulers over the offspring (‫ )זרע‬of Abraham, Isaac, and Jacob?” In this final verse of the section, “seed” refers to a Davidic group, a Levitical group, a group associated with Jacob and his legacy, and finally to a group linked to Abraham, Isaac, and Jacob as its namesake. The assemblage of phrases in Jer 33:26 indicates the harmonization of ancestor traditions, a practice observable in the Persian period.34 The commentaries tend to place Jer 33:14–26, a passage

31  In the words of Lai, “The convergence and difference between the three texts [Isa 56:1–8; 60:7; 66:18–21] reflect the efforts of Trito-Isaiah tradents in negotiating the ethnic and religious identity in the context of Persian Yehud.” Lai, “Reconceiving” (see n. 28), 286. 32  J. Blenkinsopp, Isaiah 1–39 (AB 19; New Haven, CT: Yale University Press, 2000), 226. 33  Two and a half centuries after the composition of the Ezra narrative, the concept of the holy seed figures prominently in texts such as Jubilees (16:25–26; 25:3), where it describes a kinship unit, and in Aramaic Levi (16–18; 22–25), where it refers to priesthood. On holy seed in these and related texts, see A. S ivertsev, Households, Sects and the Origins of Rabbinic Judaism (JSJSup 102; Boston: Brill, 2005), 54, 70, 81–84. 34  See Edelman et al., Opening (see n. 5), 152.

Holy Seed

537

not found in the LXX, in the fifth century BCE, and William Holladayʼs dating it to the end of the fifth century35 places the text close in time to the Ezra narrative. I now return to the question, where might the writer of the Ezra narrative encounter the term seed, ‫ ?זרע‬Asked this way, the question implies that the writer of Ezra 9:2 was actively searching for a term to describe the golah community. In fact, it is more likely that the term ‫ זרע‬found this writer, inasmuch as ‫ זרע‬was commonplace in Persian period texts such as Trito-Isaiah and Jer 33. It is no leap, for example, from the “blessed seed” of Trito-Isaiah to Ezraʼs “holy seed,” although there are differences between the two concepts, as has been noted. There is no support for the conclusion that the writer of Ezra took the word seed, ‫זרע‬, from the H legislation on agriculture, since the use of seed as a metaphor for a social group was more widespread. Given the broad use of seed, it is somewhat striking that Lev 20:22–26 does not use ‫ זרע‬in this way and so does not privilege the Judeans living in the land of Judah to the degree that the Ezra writer does. 4.2 How Is D Related to the Concept of Holy Seed in Ezra? The second question is how is D material related to the concept of the holy seed incorporated into this portion of the Ezra narrative? The response will involve further comparison of Ezra 9:1–5 and Lev 20:22–26. Ezraʼs prohibition of intermarriage clearly shows the influence of D. For example, in Ezra 9:1, the list of the eight nations is based on Deut 7:1; 23:3; and in Ezra 9:2, the statement “they have taken some of their daughters as wives for themselves and for their sons” replicates the formulaic prohibition against intermarriage in Deut 7:3 (“do not give your daughter to his son or take his daughter for your son”). The reference to Israel as a holy seed (9:2) is “just as” the command in Deut 7:6 that Israel separate from other nations (“because you are a holy nation to Yahweh your God”).36 Furthermore, ideas and language found in Ezra 9:1–2 recur later in Ezra 9 (9:11–12, 14) to reinforce the connection to D, and indeed there are D 35  W. L. H olladay , Jeremiah 2: A Commentary on the Book of the Prophet Jeremiah, Chapters 26–52 (Hermeneia; Minneapolis: Fortress, 1989), 230. Mark Leuchter observes that Jer 33:14–26 closes with references to the Deuteronomistic tradition and the pentateuchal narratives: “We should note that the reference to both the Pentateuchal and Deuteronomistic language in the same rhetorical breath suggests that the author/compiler of Jer 33:14–26 was working at a time when a narrative spanning Genesis–2 Kings had obtained an authoritative position in Israelʼs developing canon but before the time when the Pentateuch had been separated into a distinct, self-contained work.” Leuchter describes, in fact, the period of the combined P/non-P Hexateuch, or the early fourth century, although he himself dates Jer 33:14–26 significantly earlier. The textʼs date notwithstanding, it is contemporary with the combined P/non-P Hexateuch and the Ezra narrative. In the manner of the Ezra narrative, Jer 33:14–26 uses the term seed to designate particular social groups in Persian Yehud. See M. Leuchter, The Polemics of Exile in Jeremiah 26–45 (Cambridge: Cambridge University Press, 2008), 80. 36  M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon Press, 1985), 116.

538

Richard J. Bautch

expressions throughout the chapter, beginning with the prose introduction (Ezra 9:1–5) and continuing to the end of the penitential prayer (Ezra 9:6–15).37 In all, there are nine lexical points of contact between Ezra 9:6–15 and Deuteronomic discourse. Although these points are concentrated in 9:11–12, they are attested throughout the Ezra text.38 The surfeit of D material in Ezra 9 constitutes what Reinhard Kratz calls “a new theological dogmatics that fits in well with this authorʼs writing” and reflects “the Deuteronomistic picture of history.”39 The manner in which the writer of Ezra 9 incorporates D material into his text is thoroughly Deuteronomistic. In the Persian period, there were Deuteronomistic editors of the Hexateuch (and subsequently the Pentateuch) motivated by certain concerns, such as segregating the peoples of the land away from the golah community. Thomas Römer notes that although the concern over mixed marriages is manifest in key passages of Deuteronomy and in the books that follow (the so-called Deuteronomistic History), Ezra–Nehemiah also and equally reflects this ideology.40 Ezra 9–10 are especially Deuteronomistic, based on how the writer culled parts of Deuteronomy (D) related to intermarriage and integrated them into his text. With an eye to D, we turn back to Lev 20:22–26. Jacob Milgrom plausibly suggests that an H tradent appended Lev 20:22–26 to chapter 20 based on passages found in Leviticus 18.41 Although Milgrom would date such activity earlier than the fourth century, the point is that Lev 20:22–26 is likely a secondary text and the product of redaction. The redactors in question were focused on the Holiness Code but knew other traditions, too, as evidenced by the expression “a land flowing with milk and honey” (20:24). This description of the land is never found in Priestly texts except here, though it occurs twelve times in the Hebrew Bible, with eight of those instances in Deuteronomy or Jeremiah (Exod 3:8, 17; 13:5; Num 16:13–14; Deut 6:3; 11:9; 26:9, 15; 27:3; 31:20; Jer 11:5; 32:22). Many commentators regard “a land flowing with milk and honey” in Lev 20:24 as a Dtr interpolation, and indeed, it may have been designed to harmonize Lev 20:22–26 with a broader pentateuchal Weltanschauung grounded in Deuteronomy. Such harmonizing or coordination was standard at a late stage in the Pentateuchʼs formation.42 37  Reinhard Kratz documents this linkage between D and Ezra; he also shows that Deuteronomic lexemes from Ezra 9 recur in Ezra 10 (e.g., “the peoples of the land” is attested in 9:1, 11, 14; 10:2, 11). See R. G. Kratz, The Composition of the Narrative Books of the Old Testament (trans. J. Bowden; New York: T&T Clark, 2005), 79. 38  The nine points of contact are listed with commentary in Bautch, Developments (see n. 5), 86–89. 39  Kratz, Composition (see n. 37), 78. 40  T. R ömer , The So-Called Deuteronomistic History: A Sociological, Historical and Literary Introduction (London: T&T Clark, 2005), 150, 172, 177. 41  J. M ilgrom , Leviticus 17–22: A New Translation with Introduction and Commentary (AB 3A; New Haven, CT: Yale University Press, 2008 [1st ed., 2000]), 1759–1761. 42  See Carr, “Scribal Processes” (see n. 10), 63–83.

Holy Seed

539

Importantly, Lev 20:22–26 contains no Deuteronomic language or ideas other than “a land flowing with milk and honey.” There is no mention of a holy people (much less a holy seed) that would coordinate the passage in Leviticus with, for example, Deut 7:6 or 14:2a, 21a. These verses are three motive clauses in Deuteronomy that address the instructed as a sacral collective: “You are a people holy to Yahweh your God.” Rather, “a land flowing with milk and honey” is the extent of D influence in Lev 20:22–26, despite the good fit and richer harmonization that would result were a reference to holy people (based on Deut 7:6; 14:2, 21a) included.43 Thus, the exhortation of Lev 20 does incorporate D material, namely, “a land flowing with milk and honey” (Lev 20:24), but it does not include a D-based reference to the “holy people” or to the people as holy. These facts further clarify the relationship between Lev 20:22–26 and Ezra 9:1–5, two late Persian texts promoting endogamy. Ezra 9 utilizes a wealth of D material, articulated from a Deuteronomistic point of view, to build its case for endogamy as the norm of the Israelite ethnos, the holy seed. In the case of Lev 20, a different course of action results in a text with select D material, none of which gives voice to the notion of the holy people as it is found in Deuteronomy.

5 Summary and Conclusions To recapitulate, this study has shown that the combined P/non-P Hexateuch and the Ezra narrative are contemporaneous, both dating to the first half of the fourth century. The two are comparable in other respects as well. Both are evolving constellations of literary tradition that will eventually exist as fixed units within the Bible, namely the Pentateuch and the companion books of Ezra–Nehemiah. They are especially comparable in that both incorporate in a significant way the language and ideas of the D and P sources. Textually, the combined P/ non-P Hexateuch and the Ezra narrative are siblings of a sort, and their common descent from D and P is a dominant trait that they share. Moreover, Lev 20:22–26 and Ezra 9:1–5 exist within these two textual entities and replicate their parallelism on a smaller scale. Both passages employ D and P elements to exhort the Jerusalem community to refrain from mixing with people outside the ethnos. In this vein, Ezra 9:2 addresses the community as “holy seed,” but no such language is found in Lev 20:22–26. The passages align in many ways, but not in this one.  An example of this good fit is found in the final portion of Lev 11, Lev 11:44–45, which, like Lev 20:22–26, is understood to be a late addition, literarily later than P, H, and Deuteronomy. The tradents responsible for Lev 11:44–45 correlated their text with the earlier Deut 14:3–20 and likely incorporated “You are a people holy to Yahweh your God” (Deut 14:2a, 21a) into Lev 11:44–45. See R. Achenbach, “Zur Systematik der Speisegebote in Leviticus 11 und in Deuteronomium 14,” ZABR (2011), 161–209, here 187. 43

540

Richard J. Bautch

The question of why Lev 20:22–26 never refers to holy people or a holy seed is, on the one hand, a difficult one, because any response leads to an argument from silence. But the question has value as part of a heuristic inquiry. That is, to gather and sift through the data on group identity and the moniker of holiness in this period is to engage historical scenarios and literary likelihoods that can shed light not only on Lev 20:22–26 but also on the formation of the Pentateuch in the fourth century. The paper that Oded Lipschits presented at the 2014 Jerusalem conference showed that in the second half of the Persian period the small population of Yehud was convinced of its own holiness and sought to live and write about itself accordingly. Lipschits identified this holiness group through texts in the book of Numbers, which also probe the question of a communityʼs openness to “the other” living outside of it.44 Group identity, holiness, the Pentateuch, and the Persian period – this is the constellation to explore for a greater understanding of the Persian period. There are certain explanations of why Lev 20:22–26 never refers to holy people or a holy seed that can be dismissed at this point. It cannot be the case that the redactor responsible for Lev 20:22–26 could not conceive of the expression holy seed because it is an insight that is unique to the writer of the Ezra narrative. In Judean religious literature, seed was a common metaphor for distinct social groups, such as in Trito-Isaiahʼs “blessed seed,” and the attribute of holiness could be applied to religious collectives, as in Deut 7:6; 14:2a, 21a. A more cogent explanation for the development of Lev 20:22–26 without holy-seed language is that the companion text of Ezra 9:1–5 was so rhetorically charged that holy seed became synonymous with intergroup conflict. Recall the narrative context in Ezra: one group is hoping to dismantle the family structures of its rival group. In the fourth century, holy seed had connotations of controversy that became more pronounced in later texts. Subsequent uses of holy seed, for example, in Jubilees and Aramaic Levi,45 attest to the fact that it was a loaded expression, clearly polemical. In Jubilees especially, holy seed betrayed strong ethnic sentiment among religious and social elites. But just as there is a trajectory of writers who took up the term holy seed as a means of gaining leverage over other groups and maintaining their own status quo, there may have also been an alternative practice of nonengagement with holy-seed language, Lev 20:22–26 being a prime example. The persons responsible for this text, Lev 20:22–26, issued no bans and wrote, as we might say, more inclusively.46  For example, in the case of Num 15, which features legislation on the integration of outsiders, the community responsible for the text endeavors to deal with this issue. See M. C hristian , “Openness to the Other Inside and Outside of Numbers,” in The Books of Leviticus and Numbers (ed. T. Römer; BETL 215; Leuven: Peeters/Leuven University Press, 2008), 579–608, here 605–608. Lipschitsʼs paper is not included in this volume. 45  See n. 33. 46  Regarding an alternative and more inclusive approach to Judeans who are outside the golah group associated with the figure of Ezra, see for example the school of Hexateuchal 44

Holy Seed

541

Appendix: Textual Correlations between Ezra 9–10 and Leviticus 18–22 The first terms to consider are the root ‫בדל‬, “to separate,” which appears in Ezra 9:1, and its antonym, ‫ערב‬, “to mix,” in Ezra 9:2. Ezra laments that the priests, the Levites, and the laity have not separated themselves from the people of the land and have opted instead to intermarry, thereby subjecting themselves to their spousesʼ abominations (9:11, ‫)תועבת‬. Leviticus 20:26 states in first person that the Lord has separated “you” from the peoples to be the Lordʼs own. The idea of separation is prominent in the Holiness section (see, e.g., Lev 18:6–23) and also in the material belonging to the Priestly code in the Pentateuch.47 The importance of separation for the Priestly writers comes out in the creation story of Gen 1, which uses the verb ‫בדל‬, “to separate” (Gen 1:4, 7).48 The term ‫תועבה‬, “abomination,” also punctuates the Holiness Code, where in Lev 18:26–30 it appears three times to refer to the customs of the “men of the land.” Related to the notion of abomination (‫ )תועבה‬is the uncleanness of the land, expressed in Ezra 9:11 as the result of the abhorrent practices of the peoples of the land (‫)נדת עמי הארצות‬, who filled the realm with their impurities (‫)טמאתם‬. The idea of the uncleanness of the land is a Priestly notion of inhabitants defiling their own land through their sexual offenses, such as incest (Lev 18:27). In response, the land would spew out the offenders (Lev 18:28), and only their stopping said practices would allow them to inherit and possess the land (Lev 20:24). As Sara Japhet notes, the figure of Ezra employs the idea found in Leviticus of uncleanness among the people and across the land, in tandem with Deuteronomyʼs idea of prohibiting intermarriage (from Deut 7:1–3): “In his confession, Ezra combines the two ideas: the prohibition of intermarriage was enacted because of the uncleanness of the people, which he then describes in specific but most emphatic terms.”49 redactors described in Christian, “Openness to the Other” (see n. 44). 47  On the parallels between chapters 18 and 20 of Leviticus, see commentaries such as T. Hieke, Levitikus (HThKAT; Freiburg: Herder, 2014), 805–806; Milgrom, Leviticus 17–22 (see n. 41), 1765–1768; B. A. Levine, The JPS Torah Commentary: Leviticus (Philadelphia: Jewish Publication Society, 1989), 139–140. 48  S. G rätz , “The Second Temple and the Legal Status of the Torah: The Hermeneutics of the Torah in the Books of Ruth and Ezra,” in The Pentateuch as Torah: New Models for Understanding Its Promulgation and Acceptance (ed. G. N. Knoppers and B. M. Levinson; Winona Lake, IN: Eisenbrauns, 2007), 273–287, here 274–276. The P legislation on separation from non-Israelites is highly rhetorical, but it is not univocal; Lev 18:24 allows for a resident foreigner to dwell among Israel and observe its statutes. Similarly, Deuteronomy prohibits intermarriage with women of the nations (Deut 7:1–3), but there are exceptions, such as the case of an attractive woman taken captive after the defeat of a foreign nation (Deut 21:11–14). 49  S. J aphet , “Law and ‘The Law’ in Ezra–Nehemiah,” in From the Rivers of Babylon to the Highlands of Judah: Collected Studies on the Restoration Period (Winona Lake, IN:

542

Richard J. Bautch

Another correlation occurs with respect to the issue of guilt. Ezra 9:6b reads: “Our iniquities (‫ )עונתינו‬have risen over our heads, and our guilt (‫ )אשמתנו‬is great unto the heavens.” The guilt of which Ezra speaks in 9:6b in connection with the golah communityʼs mixed marriages is incurred principally through evil deeds (Ezra 9:13) and through folly. The strongest correspondence to iniquity and guilt is found in the law of trespass against sancta (Lev 22:14–16). Leviticus 22:16a speaks of the “iniquity of guilt” as well: “They cause them to bear the iniquity (‫ )עון‬of guilt (‫ )אשמה‬by eating their sacred offerings.” The iniquity of guilt is what a person must bear if another person defiles the sanctified gifts that the first has brought to God. How might Lev 22:16a be applied to the golah community? Consider that a Judean marrying outside the golah community so compromises the group that it is as if one person had defiled anotherʼs sacred offering. The transgression results in iniquity and guilt upon the party making the offering, which in this case is the entire golah community, whose solidarity in guilt extends even to Ezra.50 Yet another point of comparison relates to acceptable unions for priestly families. Ezra 9–10 aligns with Lev 21:6–9 inasmuch as both texts place restrictions upon whom a priest may marry, and Ezraʼs ban is more extensive because it applies to priests and non-priests alike. The legislation in Leviticus applies only in certain cases; priests may not marry divorcées, victims of rape, or promiscuous women, but they may marry widows. Ezekiel, in contrast, prohibits all these forms of marriage and permits only an Israelite virgin or the widow of a priest (Ezek 44:22). Like Ezekiel, Ezra 9–10 is highly restrictive. Ezra rules out intermarriage for priests and laity alike. This difference notwithstanding, on the issue of intermarriage the combined P/non-P Hexateuch and the Ezra narrative should not be contrasted to an extreme. Willa Johnson observes: “The parallel development of the ban against intermarriage in the Holiness Code and Ezra-Nehemiah mitigate against the notion that Ezra marked or started a heretofore unprecedented ban against interethnic marriage.”51 Her description of “parallel development” between Ezra and the Holiness Code anticipates the thesis of this essay, that Ezra 9:1–5 and Lev 20:22–26 share many similarities but not the concept of holy seed, which is particular to Ezra (9:2).

Eisenbrauns, 2006), 137–151, here 146. See also Carr, Formation (see n. 1), 223. 50  A means of expiation appears in Ezra 10:19, the instruction that violators bring a guilt offering to atone for their intermarriage. On the iniquity of guilt and its expiation, see B autch, Developments (see n. 5), 75. 51  W. M. Johnson, The Holy Seed Has Been Defiled: The Interethnic Marriage Dilemma in Ezra 9–10 (HBM 33; Sheffield: Sheffield Phoenix Press, 2011), 22.

What May Be Learned from Ezra–Nehemiah about the Composition of the Pentateuch? Sara Japhet A. Introduction The “Persian period” has always had a role in the critical study of the composition of the Pentateuch. As far back as the seventeenth century, Baruch Spinoza ascribed the composition of the entire Enneateuch (Genesis–Kings) to this period.1 Critical biblical scholarship has come to regard the “Persian period,” or, more generally and less precisely, the “postexilic” period, as an overarching chronological framework for the composition of the Pentateuch, with the precise nature of this stage in the composition of the Pentateuch changing from one scholar to another, in accord with his or her specific views of the compositional act or process. Certain schools have seen the Persian period as the historical context for the composition of the entire Pentateuch, and others have ascribed to it the composition of some, most, or all of the Pentateuchal sources, whereas still others have regarded it as the backdrop for certain stages of the redactional process that ended with the final form of the Pentateuch.2 The goal of this paper is not to scrutinize all these theories, to identify their merits and shortcomings, to consider the acceptance or rejection of any one of them, or to suggest an alternative. My purpose here is to study the literary works written in the Persian period in order to answer the following questions: what can literary works composed in the Persian Period tell us about the composition of the Pentateuch,

 B. de Spinoza, Theological-Political Treatise (trans. Samuel Shirley; 2nd ed.; Indianapolis: Hackett, 2001), chapter 8. 2  For recent studies and summaries of the research on the composition of the Pentateuch see, among others, E. Zenger, Einleitung in das Alte Testament (3rd ed.; Kohlhammer Studienbücher Theologie 1/1; Stuttgart: Kohlhammer, 1998), 66–176; R. G. Kratz, The Composition of the Narrative Books of the Old Testament (trans. J. Bowden; London: T&T Clark, 2005); T. Römer and K. Schmid, eds., Les Dernières Rédactions du Pentateuque, de LʼHexateuque et de Ennateuque (BETL 203; Leuven: Leuven University Press, 2007); T. B. Dozeman et al. (eds.), The Pentateuch: International Perspectives on Current Research (FAT 78; Tübingen: Mohr Siebeck, 2011); R. Albertz, “The Current Discussion on the Formation of the Pentateuch/Hexateuch,” Beit Mikra 55 (2010), 5–38 (Hebrew). 1

544

Sara Japhet

and what theory or methodology now prevailing in the scholarly world do they support? I should preface my study with two preliminary remarks. The first is that our knowledge of the Persian period is very limited and many of its aspects are still controversial. Although much has been done in recent scholarship to unveil the historical and literary contours of this period, much is still unknown and controversial. I will therefore restrict myself to searching only for the evidence that may be culled from one literary work: the book of Ezra–Nehemiah (E-N). My second remark is that the study of E-N – like that of other aspects of the Persian period – has greatly expanded and intensified in recent decades and, like everything else in biblical studies, has produced varied results and views. Since I can work only within the framework of my own view of this book, I will present it in advance: I regard E-N as one book, a historiography. The greater part of the book consists of sources of different genres, dates, authors, and modes of composition quoted basically as they are. The author of E-N brought the sources together, arranged them according to his own historical, chronological, and theological presuppositions, and added some sections of his own.3 Thus, though I do not rule out a priori the possibility of later redactional additions to the work, I believe that such additions are rather limited. Also, though I am fully aware of the different dates, historical backgrounds, and viewpoints of the individual sources in E-N, for the purpose of this article I treat the book as a single literary work. I date the author of this book to the last part of the Persian period, perhaps toward 350 BCE.4 With these preliminary remarks in view, I can now move on to the main discussion.

3  The long literary sources are Nehemiahʼs memoirs (Neh 1:1–7:5; 11:1–2; 12:27–43; 13:4–31) and the Ezra story (Ezra 7–10; Neh 8). The rest of Neh 8–12 consists of several sources of different origins, among them lists of priests and Levites, a list of the inhabitants of Jerusalem, a list of the settlements in the province of Judah, a prayer of the Levites (see discussion below), an account of the making of the covenant, and some appendices. Of these sources, some are earlier than Nehemiahʼs memoirs and some are later. Ezra 1–6 is the latest layer of the book; the bulk of it is composed of earlier original documents. I concur with the view of several scholars that this part of the book is the work of E-Nʼs author. 4  A full discussion of these topics may be found in my previous works: S. Japhet, “Composition and Chronology in the Book of Ezra–Nehemiah,” in Second Temple Studies 2: Temple and Community in the Persian Period (ed. T. C. Eskenazi and K. H. Richards; JSOTSup 175; Sheffield: Sheffield Academic Press, 1994), 189–216 (repr. in From the Rivers of Babylon to the Highlands of Judah [ed. S. Japhet; Winona Lake, IN: Eisenbrauns, 2006], 245–267); eadem , “Periodization between History and Ideology II: Chronology and Ideology in Ezra– Nehemiah,” in Judah and the Judeans in the Persian Period (ed. O. Lipschits and M. Oeming; Winona Lake, IN: Eisenbrauns, 2006), 491–508 (repr. in From the Rivers of Babylon to the Highlands of Judah [ed. S. Japhet; Winona Lake, IN: Eisenbrauns, 2006], 416–431); see also eadem , Ezra–Nehemiah: A Commentary (forthcoming; Hebrew).

What May Be Learned from Ezra–Nehemiah?

545

B. The Footsteps of the Pentateuch in Ezra–Nehemiah It has been shown by quite a few scholars that E-N contains many echoes of the Pentateuch; some are explicit references and some are literal quotations, and others are allusions and interpretations. Of the two most important genres of pentateuchal literature, law and narrative, the lionʼs share of these echoes relate to the legal material. I will now consider the reflections in E-N, first of pentateuchal narratives and then of the law. (a) The Narrative Material The section of E-N that displays the most extensive reflection of the pentateuchal narrative is the prayer of the Levites in Neh 9.5 In recent years this long prayer has been defined as belonging to the genre of “penitential prayer,” characterized by the conceptual framework of theodicy: the glorification of Godʼs justice and the confession of the peopleʼs sins, which together serve as the point of departure for a plea for help in a dire situation.6 One of the most typical features of 5  For discussions of this prayer, see, in addition to the commentaries on E-N and the introductions to biblical literature, A. C. W elch , “The Source of Nehemiah 9,” ZAW 47 (1929), 130–137; M. G ilbert , “La Place de la Loi dans la Prière de Néhémie 9,” in De la Tôrah au Messie: Mélanges Henri Cazelles (ed. M. Carrez et al.; Paris: Desclée, 1981), 301–316; H. G. M. W illiamson , “Structure and Significance in Nehemiah 9,” Proceedings of the Ninth World Congress of Jewish Studies, Panel Sessions: Bible Studies and Ancient Near East (Jerusalem: World Union of Jewish Studies, 1985), 117–131; W. C hrostovski , “An Examination of Conscience by Godʼs People as Exemplified in Neh 9:6–37,” BZ 34 (1990), 253–261; S. Segert, “History and Poetry: Poetic Patterns in Nehemiah 9,” in Storia e Tradizioni di Israel: Festschrift A. Soggin (ed. D. Garronne and F. Israel; Brescia: Paideia, 1991), 255–265; G. A. Rendsburg, “The Northern Origin of Nehemiah 9,” Biblica 72 (1991), 348–366; J. V ermeylen , “The Gracious God, Sinners and Foreigners: How Nehemiah 9 Interprets the History of Israel,” in History and Identity: How Israelʼs Later Authors Viewed Its Earlier History (ed. N. Calduch-Benages and J. Liesen; Berlin: de Gruyter, 2006), 77–114; G. Kugler, “Present Affliction Affects the Representation of the Past: An Alternative Dating of the Levitical Prayer in Nehemiah 9,” VT 63 (2013), 605–626; and in particular, M. J. Boda, Praying the Tradition: The Origin and Use of Tradition in Nehemiah 9 (BZAW 277; Berlin: de Gruyter, 1999). 6  See Boda, Praying the Tradition (see n. 5), 28–32, and the different definitions of this genre given in the appendix, 198–202. See also the collection of essays in M. J. Boda et al. (eds.), Seeking the Favor of God (EJL 22; 3 vols.; Atlanta: Society of Biblical Literature, 2006–8). For a broad survey of the concept and literary expressions of theodicy in the Bible and its world, see A. L aato and J. C. de M oor (eds.), Theodicy in the World of the Bible (Leiden: Brill, 2003); for the concept of theodicy in E-N specifically, see S. Japhet, “Theodicy in Ezra–Nehemiah and Chronicles,” in Theodicy in the World of the Bible (ed. A. Laato and J. C. de Moor; Leiden: Brill, 2003), 429–445. It should be noted that the book of Chronicles does not include penitential prayers and the very concept of confession of the sins of earlier generations is completely foreign to its religious outlook. See S. Japhet, “The Relationship

546

Sara Japhet

such prayers is the recounting of Israelʼs past, which aims to demonstrate Godʼs justice on the one hand and his endless compassion for his people on the other. The prayer in Neh 9 is of interest from many perspectives, but I will restrict my attention to the question at hand: Does the prayerʼs historical survey reflect the pentateuchal narrative, and if so, how? Our attention is immediately attracted by the structure of the prayer:7 1. Verse 5: Opening: Call to praise God 2. Verse 6: Creation of the world 3. Verses 7–8: Abraham 4. Verses 9–11: Servitude in Egypt, exodus, and crossing the Sea 5. Verses 12–23: Traveling through the wilderness: Verse 12: Pillars of cloud and fire Verse 13: Theophany at Sinai Verse 14: Receiving good laws, particularly the Sabbath Verse 15: Provision of bread and water Verses 16–21: Rebellion in the wilderness (including the sin of the     golden calf) and forgiveness Verse 22: Conquest of the kingdoms east of the Jordan Verse 23: Multiplication of the people 6. Verses 24–25: Conquest of Canaan 7. Verses 26–31: Review of life in the land: rebellion and forgiveness 8. Verses 32–37: The present situation and implied plea for deliverance Even a casual glance at this structure makes it clear that the historical survey follows – with some minor changes – the general outline of the pentateuchal narrative. After the call to bless God, the survey begins with a declaration of the monotheistic principle and then moves on to recount the creation of the world and the choosing of Abraham; it continues through the account of servitude in Egypt and the exodus and goes on to relate the crossing of the sea and the travel in the wilderness, including the giving of the law at Sinai. The survey then mentions the conquest of the kingdoms east of the Jordan and ends with the conquest of Canaan. The prayer then presents a general overview of life in the land. The author of the prayer selected certain elements from the pentateuchal story and omitted others, quoted certain passages literally and paraphrased others, and used all these devices to serve his own goals. Studied on its own, the order of the events presented in the prayer may lead one to the conclusion that the author of the prayer was acquainted with the between Chronicles and Ezra–Nehemiah,” in From the Rivers of Babylon to the Highlands of Judah [ed. S. Japhet; Winona Lake, IN: Eisenbrauns, 2006], 177–178. 7  The structure of the prayer is presented somewhat differently by various scholars. See, among others, Vermeylen, “The Gracious God” (see n. 5), 77–105. These views do not affect the general conclusions that follow in this article.

What May Be Learned from Ezra–Nehemiah?

547

tradition that underlies the Pentateuchʼs structure and course of events rather than with the actual text of the Pentateuch.8 Such a conclusion would be wrong; but the clarification of this point necessitates that we make a detailed study of the text of the prayer. I will do this for several selected passages. (1) Nehemiah 9:6: The Creation of the World You alone are the Lord. You made the heavens, the highest heavens and all their host, the earth and everything upon it, the seas and everything in them. You keep them all alive and the host of heaven prostrate themselves before You.9

The opening of this prayer, and in particular the combination of the monotheistic declaration with the description of the creation of the world, is almost a literal quotation of 2 Kgs 19:15 (= Isa 37:16): “You alone are God of all the kingdoms of the earth.10 You made the heaven and the earth.” Although it is possible that by this time the declaration of the monotheistic principle had already become a conventional statement, it is not very common in the Bible; the formulation here almost certainly draws on this earlier prayer tradition. The narrative framework, which begins with the story of creation, is similar to the framework of the Pentateuch itself, where the primeval history serves as an introduction to and setting for the history of Israel. This point of beginning is of significance, as it diverges from the more common tendency of biblical historical surveys to avoid the creation story and start at other points in history.11 The prayer presents the universe as composed of three spheres: the heavens and their hosts, the earth and everything on it, and the seas and everything in them. This is an expanded quotation of Exod 20:11: “For in six days the Lord made heaven and earth and seas, and all that is in them” (see also Ps 146:6). The same tripartite division of the universe is reflected in Gen 1, albeit in a somewhat 8  Such a view is reflected in the title of Bodaʼs book, Praying the Tradition. Boda intends the term tradition to include both oral and written traditions and thus circumvents the pressing question of the prayerʼs relationship to the actual text of the Pentateuch. Boda, Praying the Tradition (see n. 5), 3–19. 9  In general, I follow the English translation of the NJPS but deviate from it as the need arises. I point out these deviations. 10  See also Deut 4:35; 2 Kgs 19:19 (= Isa 37:20). For other formulations of this declaration, see Deut 6:4 as well as Isa 43:1–11; 44:6, 10; 45:5–7, among others. 11  Ps 105 begins with Abraham; Ps 106 begins with the exodus and the wilderness; Ps 78 begins with the crossing of the sea and the traveling in the wilderness and then goes back to the plagues in Egypt; and Dan 9 features a general survey in which only one specific historical event is mentioned, namely, the exodus (Dan 9:15). The connection between creation and historical survey is found only in two cases: Jer 32:17–21: “Ah, Lord God! You made heaven and earth with Your great might and outstretched arm. [. . .] You displayed signs and marvels in the land of Egypt with lasting effect. [. . .] You freed Your people Israel from the land of Egypt [. . .]”; and note also Ps 136:1–6: “Praise the Lord; for He is good, His steadfast love is eternal. [. . .] Who made the heavens with wisdom. [. . .] Who spread the earth over the water.”

548

Sara Japhet

different order: the land and everything on it (vv. 11–12, 24–25, 26–28), the heaven and all its hosts (vv. 8, 14–18), and the seas and the air (vv. 20–22). This basic structure, in the wake of Gen 1, is elaborated in our text by the addition of a few more elements, derived from other traditions. The definition of the heavens is enlarged with the addition of the “highest heavens” (Heb.: ‫)שמי השמים‬, and the term “hosts” is used for the heavenly bodies. The view that heaven is built of two stories – the heavens seen by the human eye and the “highest heavens” above them – is expressed in the Bible only six times, all in Deuteronomic texts or texts influenced by them (Deut 10:14; 1 Kgs 8:27 [= 2 Chr 6:18]; 2 Chr 2:5; Ps 148:4). The term “host of heaven” (Heb.: ‫ )צבא השמים‬is quite common, but the emphasis of this passage is that “the host of heaven” is submissive to the Lord. Rather than being the object of prostration (compare Zeph 1:5; also Ezek 8:16), it prostrates itself to the Lord, who is “God alone.” Another emphasis peculiar to the prayer is the idea that God not only created the world but also provides for it. The praise “You keep them all alive” summarizes the message of Ps 104. (2) Nehemiah 9:7–8: Abraham You are the Lord God who chose Abram, who brought him out of Ur of the Chaldeans and changed his name to Abraham. (8) Finding his heart true to You, You made a covenant with him to give the land of the Canaanite, the Hittite, the Amorite, the Perizzite, the Jebusite and the Girgashite – to give it to his descendants. And You kept Your word, for You are righteous.

The close affinity of these verses to Gen 15 is evident and has been acknowledged by all the scholars who have dealt with them.12 Following the rhetorical repetition of “You are the Lord,” which alludes to the monotheistic principle, the third colon in v. 7 paraphrases Gen 15:7: “I am the Lord who brought you out from Ur of the Chaldeans,” with adjustment to the person of the speaker. “Ur of the Chaldeans” is mentioned in the Bible only twice more, in the description of the family of Terah (Gen 11:28, 31). The description of Godʼs act in “bringing out [Abram] from Ur of the Chaldeans” is peculiar to Gen 15:7 and the prayer. However, this quotation from Gen 15 is being interpreted here within the conceptual framework of Deuteronomy, as indicated by the use of the verb “to choose.” The idea of “choosing” or “chosenness” is characteristic of Deuteronomy and the Deuteronomistic literature in general;13 and yet, the application of the concept of “choosing” to Abraham is peculiar to this prayer. In Deuteronomy, 12  See, among others, J. M. Myers, Ezra, Nehemiah (AB 14; Garden City, N.Y.: Doubleday, 1965), 167; H. G. M. Williamson, Ezra, Nehemiah (WBC 16; Waco, TX: Word, 1985), 313; J. Blenkinsopp , Ezra–Nehemiah: A Commentary (OTL; Philadelphia: Westminster, 1988), 304. 13  See Deut 4:37; 7:6, 7; 14:2; 18:5; 21:5, and cf. the phrase “in the place that the L ord will choose,” which appears in Deut about twenty times. See, among others, M. Weinfeld, Deuteronomy and the Deuteronomic School (Oxford: Clarendon, 1972), 324, 327.

What May Be Learned from Ezra–Nehemiah?

549

God “loved the patriarchs” and “swore to them” but “chose” their offspring: “Yet it was to your fathers that the Lord was drawn in His love for them, so that He chose you, their lineal descendants, from among all people” (Deut 10:15; see further 4:37; 7:8). The prayer is the only place in the Bible where Abraham himself is described as having been chosen!14 In verse 8, the phrase “and you found his heart faithful before you” (NRSV; ‫ )ומצאת את לבבו נאמן לפניך‬is a paraphrase of Gen 15:6, “and because he put his trust in the Lord, He reckoned it to his merit” (‫)והאמן ביהוה ויחשבה לו צדקה‬. (The similarity between the phrases is obscured by the differences in the translations). Defining Abraham by the root ‫ אמ״ן‬is peculiar to these two texts.15 The end of the verse, “You made a covenant with him to give the land of the Canaanite [. . .] to give it to his descendants,” is another paraphrase, this time of Gen 15:18: “On that day the L ord made a covenant with Abram, saying, ‘To your offspring I assign this land.’ ” Two matters should be pointed out here. First, the making of a covenant with Abraham is related twice in the Pentateuch: in Gen 15:18 and later in Gen 17:2, 4, 7. The terminology used in the prayer is ‫ – כרת ברית‬the same term as in Gen 15:18 (and the most common term for establishing a covenant), whereas in Gen 17 the terminology varies: “I will establish My covenant between Me and you” (‫ ;ואתנה בריתי‬v. 2); “this is My covenant with you” (‫ ;הנה בריתי אתך‬v. 4); “I will maintain My covenant between Me and you” (‫ ;והקמתי את בריתי‬v. 7). The second point is that in Gen 15, the land is promised to Abrahamʼs descendants, as in the prayer, whereas in Gen 17:8, the promise of land is to both Abraham himself and his descendants: “I assign the land you sojourn in to you and your offspring to come, all the land of Canaan, as an everlasting holding” (see also Gen 13:15; and in Gen 12:7; 24:7; 26:4; 48:4, ‫לזרעך‬, “to your offspring”). The list of the six peoples of the land is also found in Gen 15:20 (with the same omission of the Hivites) – although Gen 15 also lists four additional peoples who never appear in the other conventional lists – in order to bring the number to ten (the Kenites, the Kenizzites, the Kadmonites, and the Rephaim).16 The next sentence in v. 8, along with v. 9, effects a transition from the story of Abraham to that of the servitude in Egypt – skipping the other patriarchs and Jo For the significance of this idea, see R. Rendtorff, “Nehemiah 9: An Important Witness of Theological Reflection,” in Tehillah le-Moshe: Biblical and Judaic Studies in Honor of Moshe Greenberg (ed. M. Cogan et al.; Winona Lake, IN: Eisenbrauns, 1997), 111–117. 15  The only other biblical persons characterized as “faithful/trustworthy to the L ord ” are Moses (Num 12:7: “My servant Moses; he is trusted throughout My household”) and Samuel (1 Sam 13:20: “that Samuel was trustworthy as a prophet of the Lord”). David is also characterized as “trusted,” but in relation to Saul (1 Sam 22:14: “But who is there among all your courtiers as trusted as David?”). 16  This difference can be explained as either an omission from v. 8, in order to make the list conform more closely to the more conventional list of nations, or as addition to Gen 15:19–21, in order to raise the number of nations to ten. 14

550

Sara Japhet

seph. It begins with an apologetic statement: “And You kept Your word, for You are righteous,” followed by “You took note of our fathersʼ affliction in Egypt,” etc. At first glance this sequence may seem to be a deviation from the storyline of Gen 15, but this would be a misunderstanding. The jump to the servitude in Egypt, together with the apologetic sentence preceding it, is in strict adherence to the continuation of the narrative in Gen 15. After the making of the covenant, God reveals to Abraham that “your offspring shall be strangers in a land not theirs, and they shall be enslaved and oppressed four hundred years; but I will execute judgment on the nation that they shall serve, and in the end they shall go free with great wealth” (15:13–14). Nehemiah 9:8 offers an interpretation of the fulfillment of this prediction through its direct transition from Abraham to the servitude in Egypt, described very concisely as “our fathersʼ affliction.” The emphasis of the prayer in Neh 9 is on the fulfillment of Godʼs words, because God “is righteous.” Through all these small alterations and adaptations, the recounting of Gen 15 becomes a meaningful interpretation that serves the purpose of the prayer; it emphasizes both Godʼs justice and his compassion for his people. So far I have ignored one short phrase in v. 7: “And changed his name to Abraham.” At first sight this statement seems completely out of place, as it has nothing to do with the prayerʼs topic and ideas, but its function and purpose become clear at a second look. The function of this statement is to introduce into the prayer an obvious allusion to Gen 17, the only place in the Pentateuch where the changing of Abrahamʼs name is narrated and explained: “As for Me, this is My covenant with you: You shall be the father of a multitude of nations. And you shall no longer be called Abram, but your name shall be Abraham, for I make you the father of a multitude of nations” (Gen 17:4–5). By the addition of the short reference to Abrahamʼs change of name, the prayer combines the two covenants with Abraham – related in Gen 15 and 17 – into a single covenant. Although the sources of the texts in Gen 15 and 17 are still debated in scholarly circles,17 the testimony of the prayer is unequivocal: the author of the prayer in Neh 9 was familiar not merely with the tradition of the covenant with Abraham but with the actual text of Genesis, which included both chapter 15 and chapter 17, with their literary differences. He quoted them, paraphrased them, and proposed his own interpretation.

17  That Gen 17 belongs to the Priestly layer of the Pentateuch is generally accepted, but the attribution of Gen 15 is strongly debated: Does it belong to J? Or J with Deuteronomic touches? Or is it of an altogether unidentified origin? See, among others, J. Ha, Genesis 15: A Theological Compendium of Pentateuchal History (BZAW 181; Berlin: de Gruyter, 1989); J. C. G ertz , “Abraham, Moses und der Exodus: Beobachtungen zur Redaktion Geschichte von Gen 15,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 63–81.

What May Be Learned from Ezra–Nehemiah?

551

(3) Nehemiah 9:9–11: Servitude in Egypt, Exodus, and Crossing the Sea You took note of our fathersʼ affliction in Egypt, and heard their cries at the Sea of Reeds. (10) You performed signs and wonders against Pharaoh, all his servants and all the people of his land, for You knew that they acted presumptuously toward them. You made a name for Yourself that endures to this day. (11) You split the sea before them; they passed through the sea on dry land, but You threw their pursuers in the depths, like stone into the raging waters.

After the concise reference to the servitude in Egypt, the prayer moves on to the exodus from Egypt and the crossing of the sea. The text is constructed in the same mode as before: quotation / paraphrase / allusion and combination of the Pentateuchʼs different texts dealing with these topics. “You took note of our fathersʼ affliction in Egypt” (‫ )ותרא את עני אבתינו במצרים‬is a paraphrase of Exod 3:7: “I have marked well the plight of My people in Egypt” (‫ראה ראיתי את עני‬ ‫ ;עמי אשר במצרים‬the similarity is obscured by the differences in translation); “the cry at the sea” alludes to “The Israelites cried out to the Lord” (Exod 14:10), with linguistic allusions to Exod 3:7, 9. The ten plagues that God inflicted on the Egyptians are not detailed in the prayer but rather summarized by the phrase “signs and wonders,” which recurs mainly in the Deuteronomic texts and those under their influence. Verse 10 thus presents a combined quotation of two verses from Deuteronomy: “The L ord displayed before our eyes great and awesome signs and wonders (‫)אותת ומפתים‬ against Egypt, against Pharaoh and all his household” (Deut 6:22; NRSV); and “For all the signs and wonders that the L ord sent him to perform in the land of Egypt, against Pharaoh and all his servants and his entire land” (Deut 34:11; NRSV).18 One should note the close resemblance of the prayer to Jer 32:20, which combines the “signs and wonders” with the “name”: “You showed signs and wonders in the land of Egypt [. . .] and have made yourself a name (that continues) to this very day” (NRSV).19 The crossing of the sea is described in detail (v. 11) in three statements that quote or paraphrase pentateuchal texts, particularly from Exodus. Explicit linguistic allusions may be seen in the description of the crossing with the root “split” (‫ ;בק״ע‬Exod 14:21 etc.), the passage through the sea on dry land and the use of the root ‫( עב״ר‬Num 33:8; see also Ps 78:13), the description of walking “in the sea” (‫ ;בתוך הים‬Exod 14:16, 22, 27, 29; Num 33:8), and the definition of the Egyptians as “their pursuers” (Exod 14:4, 8, 9, 23; 15:9). The description of the Egyptiansʼ end is an almost literal quotation of Exod 15:5: “They went down into the depths like a stone.”

 See also Exod 7:3; Deut 29:2; Pss 78:43; 135:9, among others.  For the intermediate colon in v. 10 – “for you knew that they acted insolently” (NRSV), which expresses the distinctive theological presuppositions of the prayer – see S. J aphet , Ezra–Nehemiah: A Commentary (see n. 4). 18 19

552

Sara Japhet

(4) Nehemiah 9:12–37: The Continuation of the Survey The same system of writing – following the narrative outline of the Pentateuch and quoting or paraphrasing its statements – is carried on throughout the remainder of the prayer. Even the repetitions – the miracles of the columns of fire and cloud (vv. 12, 19) and the provision of manna and water (vv. 15, 20) – follow the pentateuchal text, where these topics are mentioned first in Exodus and then again in Numbers (Exod 13:21–22 and Num 14:13; Exod 15:22–25; 16; and Num 20:7–11; Num 11:7–9, respectively). For continuing quotations and combinations of pentateuchal texts, note the combination of Exod 13:21–22 with Deut 1:33 in v. 12, Exod 19:20 with 20:22 in v. 13, and Exod 16:3–4 with Num 20:8–11 in v. 15; the combination of elements from Deut 1:6–8 and Exod 6:8 in v. 15; the use of the phrase ‫נת״ן ראש‬ in v. 17, taken from Num 14:4 (peculiar to these two verses); the quotation from Deut 8:4 (and its similarity to 29:4) in v. 21; and so on, to the end of the prayer.20 The conclusion is straightforward: the author of the prayer was working with the complete text of the Pentateuch on his desk. Boda ascribed this feature of Neh 9 to penitential prayers in general: “Whether focusing on historical or legal traditions, Penitential Prayers treat the Pentateuch in a similar fashion. They not only assume a completed Pentateuch, but use that body of literature as a unified whole.”21 For the purpose of this article, the evidence of Neh 9 would suffice to demonstrate this point. (b) The Legal Material The legal material of the Pentateuch is reflected in E-N in several ways, and quite frequently. There are almost thirty explicit references to the Torah, which imply – through diverse modes of phrasing – a written law book. Thus, for example: “They told the scribe Ezra to bring the book of the Law of Moses which the Lord had given to (Heb.: ‫צוה‬, commanded) Israel” (Neh 8:1; NRSV).22 Similarly: “And they found written in the Law (‫ )בתורה‬which the L ord had commanded by Moses,” “that the people of Israel should live in booths [. . .] to make booths as it is written” (8:14–15). The terms “Torah,” “book,” “as it is

20  For a full presentation of all these borrowings and quotations, see B oda , Praying the Tradition (see n. 5) and Japhet, Ezra–Nehemiah: A Commentary (see n. 4). 21  B oda , Praying the Tradition (see n. 5), 73. Boda continues by quoting Williamson: “According to Williamson (1985), commenting on Ezra 9:11–12, this trend of creating a ‘mosaic’ of many passages is ‘a pointer to the emergence of a view which came to regard Scripture as a unity and hence of uniform authority.’ ” See also Boda, Praying the Tradition, 123–124, 142, and elsewhere. 22  In the translation of the NJPS, the term Torah (Heb: ‫ )תורה‬is consistently represented by “Instruction,” and the Hebrew term for “book,” ‫ספר‬, is sometimes represented by “scroll.” For the sake of clarity I have not followed these practices in this article.

What May Be Learned from Ezra–Nehemiah?

553

written,” and “they read” are employed repeatedly, along with similar expressions.23 What parts of the pentateuchal law are quoted or referred to in E-N? Most obvious, and agreed upon by all, is the acquaintance of E-N with the laws of Deuteronomy. This is demonstrated in the clearest way by the almost literal quotation – with merely a change from the second to the third person – in Neh 13:1–3: At that time they read to the people from the Book of Moses, and it was found written that no Ammonite or Moabite might ever enter the congregation of God, since they did not meet Israel with bread and water, and hired Balaam against them to curse them, but our God turned the curse into a blessing. When they heard the Torah they separated all the alien admixture from Israel.

This is an almost literal quotation, with some abbreviation, of Deut 23:4–6: No Ammonite or Moabite shall be admitted into the congregation of the Lord; none of their descendants, even in the tenth generation, shall ever be admitted (lit.: shall come) into the congregation of the Lord, because they did not meet you with food and water on your journey after you left Egypt, and because they hired Balaam son of Beor [. . .] to curse you. But the Lord your God refused to heed Balaam; instead, the Lord your God turned the curse into a blessing for you.24

Another clear example of dependence on Deuteronomy is the instruction against mixed marriages, which appears in E-N twice, paraphrased to suit the specific contexts (Ezra 9:1; Neh 10:31). Deuteronomy 7:3 instructs: “You shall not intermarry with them, do not give your daughters to their sons or take their daughters for your sons.” This instruction is quoted in Ezra 9:12: ‫בנותיכם אל תתנו לבניהם‬ ‫ובנתיהם אל תשאו לבניכם‬, paraphrased in the NJPS as “do not give your daughters in marriage to their sons or let their daughters marry your sons.” The same instruction is quoted again as part of the covenant made by the returnees: ‫לא נתן‬ ‫בנתינו לעמי הארץ ואת בנתיהם לא נקח לבנינו‬, “We will not give our daughters in marriage to the peoples of the land, or take their daughters for our sons” (Neh 10:31).25 However, Deuteronomy is not the only pentateuchal law code represented in E-N. Foremost among the other parts of the Pentateuch is the Priestly law. 23  Specifically: (1) the Torah of Moses the man of God (Ezra 3:2); the Torah of Moses which the Lord God of Israel had given (Ezra 7:6); (2) the book of Moses (Ezra 6:18, in Aramaic; Neh 13:1, in Hebrew); (3) the book of the Torah of Moses which the Lord had commanded Israel (Neh 8:1); (4) the Torah of the Lord (Ezra 7:10); (5) the book of the Torah of God (Neh 8:8); (6) the book of the Torah of the Lord their God (Neh 9:3); (7) the Torah of God (Neh 10:29); (8) the book of the Torah given by Moses the servant of God (Neh 10:30); (9) the book of the Torah (Neh 8:3); (10) Torah, the Torah, to the Torah, your Torah, in the Torah, according to the Torah (Neh 8:2, 7, 9, 13, 14; 9:14, 26, 29, 34; 10:35, 37; Ezra 10:3). 24  On E-Nʼs interpretation of this instruction see below, pp. 559–560. 25  See further on pp. 555, 558–559.

554

Sara Japhet

The celebration of the festival of Succoth (Booths, or Tabernacles) is mentioned twice in E-N; in both cases, the text specifies that the prescriptions of the festival were performed “as written.” Quite a long passage in Neh 8:13–18 describes the celebration in the time of Ezra: On the second day [. . .] [they] gathered to Ezra the scribe to study the words of the Torah. They found written in the Torah that the Lord had commanded Moses that the Israelites must dwell in booths during the festival of the seventh month, and that they must announce and proclaim throughout all their towns [. . .] “Go out to the mountains and bring leafy branches [. . .] to make booths, as it is written.” So the people [. . .] made themselves booths [. . .] The whole community made booths and dwelt in booths. [. . .] They celebrated the festival seven days and there was a solemn gathering on the eighth, as prescribed.

The festival of Succoth is mentioned in the Pentateuch by different appellations in all the legal corpora (Exod 23:16; 34:22 – ‫ ;חג האסיף‬Lev 23:33–36, 39–45 – ‫חג‬ ‫ ;הסכות‬Num 29:12–13: ‫ ;חג ליהוה‬Deut 16:13–15 – ‫)חג הסכות‬. It is also mentioned in Deut 31:10 as the date of the convocation (hakhel) in the year of remission.26 However, the commandment to dwell in booths during this festival is peculiar to the Holiness Code (Lev 23:39), along with the instruction to add to the seven days of celebration a solemn convocation on the eighth day (Lev 23:35–36, 39–43). This law stands in some tension with the law of Deuteronomy, which describes the festival as a centralized event of seven days, with no booths and no convocation: “You shall hold a festival for the Lord your God seven days, in the place that the Lord will choose” (Deut 16:15). The festival described in Neh 8 as a celebration in the time of Ezra shares most of its features with Lev 23 but also differs from that text in some respects. Nevertheless, this festival is said to have been celebrated “as written” (‫ככתוב‬, Neh 8:15) and “as prescribed” (‫כמשפט‬, 8:18); there are booths in Jerusalem and a convocation on the eighth day, according to the commandment of the Torah – that is, the Holiness Code.27 The celebration of Succoth is also mentioned in Ezra 3:4–5: “Then they celebrated the Festival of Tabernacles as is written, with its daily burnt offerings in the proper quantities, for each day as is prescribed for it, followed by the regular burnt offerings and the offerings for the new moons and for all the sacred fixed times of the Lord.” This description of Succoth is a concise summary of the sacrificial system as prescribed in Num 28–29. The expression used in Ezra 3:4, “in the proper quantities [. . .] as is prescribed for it” (‫)במספר כמשפט‬, is taken from Num 29,  It is also mentioned as “the Feast of Succoth” in Zech 14:16–19 and as “the pilgrimage” (Heb: ‫)חג‬, outside the legal corpora. 27  For the celebration described in Neh 8 as a contemporary interpretation of the text of Lev 23, see K. W. Weyde, “And They Found It Written in the Law: Exegetical Proceedings Reflected in Nehemiah 8:13–18,” in Shai le-Sara Japhet: Studies in Bible, Its Exegesis, and Its Language (ed. M. Bar-Asher et al.; Jerusalem: Bialik, 2007), 143*–163*. 26

What May Be Learned from Ezra–Nehemiah?

555

where it is repeated seven times, for each day of the festival except the first (Num 29:18, 21, 24, 27, 33, 37). The phrase is peculiar to Num 29; it appears nowhere else in the Pentateuch and only once more in the Bible – in a text later than E-N and certainly influenced by the same source (1 Chr 23:31). Another instance of allusion to the Priestly legal material, this time combined with the instructions of Deuteronomy, is provided by Ezra 9. I have already mentioned the double quotation of the law against intermarriage taken from Deut 7:3 in Ezra 9:12 and Neh 10:31. Deuteronomy supports the prohibition of intermarriage by giving an explicit explanation for it: “You shall not intermarry with them [. . .] for they will turn your children away from Me to worship other gods, and the Lordʼs anger will blaze forth against you and He will promptly wipe you out” (Deut 7:4). Thus, the prohibition of intermarriage is one among other measures taken in the difficult struggle against idolatry, which is presented in Deuteronomy as a threat to Israelʼs very survival. Ezra, too, includes in his confession a justification of the prohibition against intermarriage but explains it in altogether different terms: “The land which you are about to possess is a land unclean through the uncleanness of the peoples of the land, through their abhorrent practices with which they, in their impurity, have filled it from one end to the other. Now then, do not give your daughters in marriage to their sons” (Ezra 9:11–12). The concept of the landʼs impurity, described by the extreme and unique term ‫( ארץ נדה‬based on the common Priestly vocabulary of sexual impurity),28 is taken from Lev 18:24–25: “It is by such that the nations which I am casting out before you defiled themselves. Thus the land became defiled and I called it to account for its iniquity, and the land spewed out its inhabitants” (and again in vv. 27–29). In Lev 18, however, the iniquities that caused the land to be defiled were different forms of incest and other sexual offenses rather than idolatry. The view of the land as “defiled” and “impure” before the entrance of the people of Israel stands in stark contrast to the views of the land as a “good land,” etc., expressed in Deuteronomy and elsewhere. Ezraʼs confession thus combines the instructions of Deut 7 with the concept of impurity of the land and the urgent need to secure the inheritance of the land, taken from Lev 18. Ezra–Nehemiah also displays dependence on the laws of the Book of the Covenant, an example of which is the commitment taken on by the community in regard to firstborns, as set forth in the covenant (Neh 10:37). The phrasing of the covenant is a conflation of four pentateuchal texts in which the instruction to dedicate the firstborn is spelled out: Exod 22:28–29; Exod 13:2; Deut 12:6; and Num 18:15. The law in the Covenant Code is phrased as follows: ‫בכור בניך‬ ‫׃ כן תעשה לשרך ולצאנך‬ ‫תתן לי‬, “You shall give Me the firstborn among your sons. You shall do the same with your cattle and your flocks” (Exod 22:28–29). The instruction given in Exod 13:2 is: ‫קדש לי כל בכור פטר כל רחם בבני ישראל‬ 28  The term appears in the Bible thirty times, twenty-five of which are in texts of Priestly origin – Leviticus, Numbers, and Ezekiel.

556

Sara Japhet

‫באדם ובבהמה לי הוא‬, “Consecrate to Me every firstborn; man and beast, the first issue of every womb among the Israelites is Mine” (with further explanation in vv. 12–14). The law of Deut 12:5–6 demands: ]. . .[ ‫כי אם אל המקום אשר יבחר‬ ‫] ובכרת בקרכם וצאנכם‬. . .[ ‫תדרשו ובאת שמה והבאתם שמה עלתיכם וזבחיכם‬, “but look only to the site that the Lord your God will choose. [. . .] There you are to go, and there you are to bring your burnt offerings and other sacrifices [. . .] and the firstlings of your herds and flocks”; and the Priestly instruction specifies: “The first issue of the womb of every being, man or beast, that is offered to the Lord, is yours” (that is: the priestsʼ; Num 18:15). The phrasing of Neh 10:37 appears at first sight to constitute a tautological repetition, but it is in fact a combination of selected elements from all these sources. “Also the firstborn of our sons and our beasts, as written in the Torah” is based on Exod 22:28–29 and Exod 13:2, which specify the dedication of both human and animal firstborn. “And to bring the firstlings of our cattle and flocks to the House of our God” reflects the law in Deut 12:5–6, which mandates bringing the firstborn to “the site that the L ord [. . .] will choose,” that is, the temple. The final element, “for the priests who minister in the House of our God” (Neh 10:37), follows Numbersʼ assignment of the firstborn to the priests. Another example of this kind is the obligation to keep the seventh year. Nehemiah 10:32b reads: “We shall forgo [the produce] of the seventh year and every outstanding debt.” This concise commitment actually combines two different issues. The first statement is phrased in a very general manner, without specifying what is to be given up (supplemented by the English translations with “the produce” [NJPS], “the crops” [NRSV, v. 31b]). However, the close affinity to the phrasing of the instruction of Exod 23:10–11, “Six years you shall sow your land and gather in its yield, but in the seventh you shall let it rest and lie fallow (‫)והשביעית תשמטנה ונטשתה‬,” implies that this short statement in Neh 10 refers to relinquishing the tilling of the land. The second statement in v. 32b specifies explicitly what is to be relinquished: “and every outstanding debt” (‫)משא כל יד‬. The term for “debt” is ‫משא‬, a debt that the creditor demands from the debtor (see also Neh 5:7, 10). The phrasing of this commitment reflects Deut 15:1–2: “Every seventh year you shall practice remission of debts [. . .] every creditor shall remit the debt that he claims from his neighbor” (‫מקץ שבע‬ ‫] שמוט כל בעל משה ידו אשר ישה ברעהו‬. . .[ ‫)שנים תעשה שמטה‬. Thus the covenant combines the two laws that appear in the Pentateuch in two different places. The law in Exodus uses the verb “forgo” (‫)נט״ש‬, but its application is limited to the release of the land, while the law in Deuteronomy uses the unique phrase ‫שמוט‬ ‫ יד‬and is limited to the remission of debts. The covenant combines the two laws and presents them as two aspects of the seventh year. The conclusion seems obvious: E-N has recourse to all the legal corpora in the Pentateuch. However, E-N does not repeat the pentateuchal legal material in toto, and it is not another law code. Rather, the book displays a clear process

What May Be Learned from Ezra–Nehemiah?

557

of selection and adaptation, reflecting its own reality and aimed at serving its own purposes.29 The evidence regarding E-Nʼs reflection of the Pentateuch is straightforward and unequivocal: E-N knew the Pentateuch in its entirety – both the narrative and the legal material. One should not rule out the possibility that some changes were introduced later on into either the Pentateuch or E-N, in the process of transmission or in the ongoing process of editing, but this does not affect the conclusion that the complete Pentateuch existed at the time of the composition of E-N.

C. The Attitude of Ezra–Nehemiah to the Pentateuch The conclusions presented above necessarily lead to an unavoidable question that has been hinted at above but not explicitly discussed: What is the attitude of E-N towards the Pentateuch at large and more specifically to its laws? Is the Pentateuch regarded as “canon,” the instructions of which should be observed “to the letter”? Does E-N follow the instruction “You shall not add anything to what I command you or take anything away from it” (Deut 4:2)? An explicit statement in Neh 10:30 introduces the commitment of the community to fully adhere to the commandments of Godʼs Torah; the people “join with their noble brothers and take an oath with sanctions to follow the Torah of God, given through Moses the servant of God, and to observe carefully all the commandments of the Lord our God, His rules and laws.” Nevertheless, as already demonstrated, this adherence to the Torah is not always realized in full literal accord with pentateuchal wording. I wish to look more specifically into this matter through the discussion of a few more texts: the covenant (‫ )אמנה‬in Neh 10, a short passage in Ezra 10, and a short paragraph in Neh 13. After the long introduction (Neh 10:1–28) and the general commitment to follow the Torahʼs commandments quoted above (vv. 29–30), the text records a number of obligations to which the people commit themselves, most of which concern the maintenance of the temple and support of the clergy (Neh 10:33– 40). These commitments are prefaced by three other matters: the prohibition of mixed marriages, the command to observe the Sabbath, and the command to observe the Sabbatical Year (vv. 31–32). Viewed from the perspective of their relationship to the pentateuchal laws, the commitments display an interesting feature. Some of the obligations reflect those of the Pentateuch – the prohibition  Therefore, the use of the “argument from silence” in respect to E-N is faulty and should be avoided. Contrast, for instance, J. Wellhausen, who regarded the silence of E-N as absolute evidence for the dating of the P source and, following it, for the composition of the Pentateuch: “The Testamentum e silentio is enough” (J. Wellhausen, Prolegomena to the History of Israel [trans. J. S. Black and A. Menzies; Edinburgh: Black, 1885], 111). 29

558

Sara Japhet

of mixed marriage (v. 31), the details of the sacrificial system (v. 34, with all terms taken from the Priestly portions of the Pentateuch), and the contributions to the clergy (vv. 36–38) – but others are new, in part or in full. Among the obligations that are not mentioned in the Torah at all are the annual temple tax of a third shekel (v. 33); the prohibition of commerce on the Sabbath and the “holy days,” particularly with the “peoples of the land” (v. 33); the “wood sacrifice” and the procedure by which its provision is organized (v. 35); and the procedures for the collection of the priestly and Levitical contributions (vv. 38b–40). Nevertheless, no distinction is made between the two groups of commitments, and they are all considered “the commandments of the Lord our God” as defined by the introductory verse. One may explain the new obligations in several ways: as having followed common traditions that were not attested in the Pentateuch, as interpretations of pentateuchal instructions,30 or as complete innovations. However, one of them – the wood sacrifice – is explicitly authorized “as written in the Torah” (v. 35). This same authorization is also accorded to the commitment to dedicate the firstborn discussed above, which is certainly not a precise replication of any single law. It seems, rather, that all these obligations are conceived of as following the commandments of the Torah “as written,” although they do not reflect the precise wording of the Torah.31 Because of the prevalence of this phenomenon, one cannot explain these cases as instances of “imprecision” or reflections of another (written) Torah. The unavoidable conclusion is that the term “as written” does not refer to the literal wording of the Pentateuch but rather to its interpretation by the contemporary congregation.32 The following examples demonstrate the same procedure, while using different definitions. Ezra 10:3 is part of the episode concerning intermarriage. I have dealt with this matter in a separate article and will refer to it here only briefly.33 In Ezra 9:1–2, the people address Ezra with the complaint that the people of Israel and the priests and Levites have not separated themselves from the peoples of the 30  For the covenant of Neh 10 as a midrash, see Y. Kaufmann, History of the Religion of Israel (4 vols.; Tel Aviv: Bialik, 1960), 4:331–338 (Hebrew); D. J. A. Clines, “Nehemiah 10 as an Example of Early Jewish Biblical Exegesis,” JSOT 21 (1981), 111–117 (without the use of the term midrash). 31  Other instances in E-N of the phrasing “as written” are Ezra 3:2, 4; Neh 8:18; and Ezra 6:18 (Aramaic: ‫ ;)ככתב ספר משה‬the phrases “they found written” and “it was found written” appear in Neh 8:14 and 13:1, respectively. 32  See also S. J aphet , “Law and ‘the Law’ in Ezra–Nehemiah,” in From the Rivers of Babylon to the Highlands of Judah: Collected Studies on the Restoration Period (ed. S. Japhet; Winona Lake, IN: Eisenbrauns, 2006), 137–151. 33  S. Japhet, “The Expulsion of the Foreign Women (Ezra 9–10): The Legal Basis, Precedents, and Consequences for the Definition of Jewish Identity,” in “Sieben Augen auf Einem Stein” (Sach 3,9): Studien zur Literatur des Zweiten Temples; Festschrift für Ina Willi-Plein zum 65. Geburtstag (ed. F. Hartenstein and M. Pietsch; Neukirchen-Vluyn: Neukirchener Verlag, 2007), 141–161.

What May Be Learned from Ezra–Nehemiah?

559

land and have taken foreign women in marriage, for themselves and for their sons. Ezra 9:3–15 narrates Ezraʼs dramatic reaction and confession. Following upon these events, Shecaniah son of Jehiel, one of the congregants, speaks up: “We have trespassed against our God by bringing into our homes foreign women from the peoples of the land. But there is still hope for Israel despite this. Now then, let us make a covenant with our God to expel all these women and those who have been born to them, in accordance with the bidding of the Lord (or: according to the counsel of my lord; NRSV) and all who are concerned over the commandment of our God, and let the Torah be obeyed” (NRSV; more accurately, “let it be done/handled according to the Torah”; Heb.: ‫)וכתורה יעשה‬. (Ezra 10:2–3)

Although Shecaniah describes his proposal as “according to the Torah,” not one of the steps he recommends is mentioned in the Pentateuch, neither the expulsion of the foreign women nor the expulsion of their sons. Both stand in clear contrast to biblical legal and genealogical concepts. As I have argued elsewhere, Shecaniahʼs actual suggestion – which was not put into words explicitly – is to lower the status of these women from regular, primary wives to secondary wives, like concubines or maidservants. As a result of this change, both the expulsion of the women and the determination of the childrenʼs status by that of their mothers rather than by that of their fathers become legally possible and could be confirmed by biblical precedents and laws and by extrabiblical law. When Shecaniah describes these procedures as “according to the Torah,” he does not have in mind the literal wording of the Torah but the interpretation of its precepts. The acceptance of his proposal by Ezra and the entire congregation demonstrates that the concept of “Torah” or “Law” is for them broader than the literal phrasing of the Pentateuch and already includes its elaboration through interpretation. Nehemiah 13:1–2 is a quotation of Deut 23:4–6, as I have already mentioned.34 The Deuteronomic instruction demands that Ammonites and Moabites should not enter “the community of the L ord .” According to the narrative in Nehemiah, what the people do to adhere to this instruction differs from the precise demand in Deuteronomy: they “separated all the alien admixture from Israel”! This deviation from the demand of Deuteronomy is multifaceted. Deuteronomy refers to the offspring of two ethnic groups, Ammonites and Moabites; the injunction is justified by their past conduct towards Israel. The congregation ignores the ethnic definition and interprets the instruction to include not just offspring but the broader notion of intermixing and then enlarges the application of the verse to encompass not only mixing with “Ammonites and Moabites” but mixing with anyone outside of the community, of whatever origin. Moreover, Deuteronomyʼs instruction refers the prohibition against admission of such people to “the congregation of the L ord .” The community understands it in more proactive terms and “separate[s] all aliens from Israel.” The force of this example lies in the fact that the Deuteronomic 34

 See above p. 553.

560

Sara Japhet

demand could not be better illustrated. It is described explicitly as taken from “the book of Moses,” where it was “written,” and quoted literally. Nevertheless, the act of the people demonstrates that the injunction is not followed literally but interpreted and applied to the communityʼs needs in their own time and reality.

D. Conclusions What then does E-N contribute to the scholarly discussion of the composition of the Pentateuch? What data does it provide or what methodology does it support? We may sum up the conclusions of our discussion in the following statements: a. Ezra–Nehemiah is familiar with the entire Pentateuch and draws amply from pentateuchal texts, both narrative and legal, according to its needs. b. The Pentateuch is defined in various terms that point to its divine origin, transmission through Moses, and written form. It is Scripture – an authoritative, holy text. c. The law and commandments of the Pentateuch are regarded by the community of the Persian period as binding. The members of the community commit themselves to fulfilling this law in all its spheres and to conducting their lives “as it is written.” d. Nevertheless, the problems presented by the actual historical period, the exigencies of “real life,” and the development of faith and theology demand solutions and answers. These are met not merely by the written word of the Pentateuch but also by deviations from its precise language, in both theology and practice. e. These “deviations” from the Torah, which consist of practical and theological innovations to meet specific social circumstances, are formulated as fulfillments of the precepts of the Torah; they are based on reinterpretations of the pentateuchal text, on harmonizations of the Torahʼs different dicta, and on oral traditions and practices that existed outside the Pentateuch. f. For the study of the Pentateuch the evidence of E-N is relevant in two ways: (1) E-N attests to the completed format of the Pentateuch and so provides a terminus ad quem for the Pentateuchʼs composition. It should be ascribed to the Persian period and cannot be relegated to the Hellenistic period.35 (2) The fact that E-N reflects the use of the Pentateuch in its final form precludes its relevance for the earlier stages of the Pentateuchʼs composition. For questions regarding the precise literary act(s) or literary process(es) by which the Pentateuch was created (sources, redaction, supplementation, etc.), and the methodology that should be followed in analyzing these act(s) or process(es), the textual evidence of E-N provides no help. 35  Unless, of course, the composition of E-N is dated to the Hellenistic period – a wonderful example of circular argumentation (“begging the question”).

Part Five Evidence for Redactional Activity in the Pentateuch

Introduction Konrad Schmid In critical biblical scholarship, there is general agreement that the Pentateuch is a heterogeneous entity that is the result of a long tradition history. This is true even for scholars like Benno Jacob,1 Umberto Cassuto,2 or R. Norman Whybray,3 who deem the Pentateuch to be the work of a single author, albeit one who employed, combined, and reworked older traditions he had at hand. The mainstream of pentateuchal scholarship, however, goes beyond the assumption of just one author or compiler for the Pentateuch. Traditional and also newer source-critical approaches reckon with four basic documents that were included in the Pentateuch, although Wellhausen admitted that there were many more: Der Einfachheit wegen abstrahiere ich meistens davon, dass der literarische Process in Wirksamkeit complicierter gewesen ist und die sogenannte Ergänzungshypothese in untergeordneter Weise doch ihre Anwendung findet. J und E haben wol erst mehrere vermehrte Ausgaben (J1 J2 J3, E1 E2 E3) erlebt und sind nicht als J1 und E1, sondern als J3 und E3 zusammengearbeitet. Ähnliches gilt von JE, Dt und Q, bevor sie mit den betreffenden grösseren Ganzen vereinigt wurden.4 For reasons of simplicity, I prescind in most cases from the fact that the literary process was in fact more complex and the so-called Supplementary Hypothesis can indeed be used in a subordinate way. J and E were probably edited and augmented several times (J1 J2 J3, E1 E2 E3), and they were combined not as J1 and E1, but as J3 and E3. A similar process took place for JE, Dt, and Q before they were combined with the relevant unities.

Hugo Gressmann supported Wellhausenʼs notion and highlighted the complex nature of the textual material that stands behind J and E: 1  B. J acob , Das erste Buch der Tora: Genesis (New York: Schocken, 1934); idem , The Second Book of the Bible: Exodus; Translated with an Introduction by Walter Jacob in association with Yaakov Elman (Hoboken, NJ: Ktav, 1992); German version, Das Buch Exodus (Stuttgart: Calwer, 1997). 2  U. Cassuto, The Documentary Hypothesis and the Composition of the Pentateuch: Eight Lectures (trans. I. Abrahams; Jerusalem: Magnes, 1961). 3  R. N. Whybray, The Making of the Pentateuch: A Methodological Study (JSOTSup 53, Sheffield: Sheffield Academic Press, 1994 [1st ed., 1987]). 4  J. Wellhausen, Die Composition des Hexateuch (3rd ed.; Berlin: Reimer, 1899), 207.

564

Konrad Schmid

Zu einer befriedigenden Erklärung wird man nur kommen, wenn man JE für Redaktoren oder Sammler hält.5 The only satisfying explanation is to consider JE as redactors or collectors.

Hermann Gunkel was even more emphatic. He stated in his commentary on the book of Genesis: Die schriftliche Sammlung der Sagen ist nicht von einer Hand und in derselben Zeit geschehen, sondern von mehreren oder gar vielen in einem ganz langen Prozess. Wir unterscheiden in diesem Prozess zwei Zeitalter: das ältere, dem wir die Sammlungen des Jahvisten (J) und Elohisten (E) verdanken, dann eine spätere, durchgreifende Umarbeitung durch den sogenannten Priesterkodex (P).6 The literary collection of the tales does not emanate from one hand or one era but was achieved by some or even many hands in a very long process. We distinguish two eras in this process: the older era, in which the Yahwist (J) and the Elohist (E) were written, and a later, thoroughgoing transformation by the so-called Priestly Codex (P).

Therefore, Gunkel highlights the collective nature of J and E: “J” und “E” sind also nicht Einzelschriftsteller, sondern Erzählerschulen.7 “J” and “E” are not individual writers but schools of narrators.

In addition to the traditional sources, scholars also posited a set of redactors. In nineteenth-century scholarship, usually three of them were assumed (RJE, RJEP, RJEPD), although current Neo-Documentarians tend to assume only one: “[H]e [i.e., the redactor] is a necessary side-effect of the recognition of multiple sources in the text, not a primary feature of the theory. The theory demands a redactor, because the source were evidently combined by someone – but no more than one.”8 Neither traditional nor Neo-Documentarian scholars attribute much textual material to the redactors, whose main task was to combine and compile the documents they had before them.9 5  H. G ressmann , Mose und seine Zeit: Ein Kommentar zu den Mose-Sagen (FRLANT 1/18; Göttingen: Vandenhoeck & Ruprecht, 1913), 372; see also 368: “In vielen Fällen sind JE weiter nichts als Etiketten, die man beliebig vertauschen darf” (“In many cases, JE are no more than tags that can be arbitrarily exchanged”). 6  H. Gunkel, Genesis (6th ed.; HKAT 1/1; Göttingen: Vandenhoeck & Ruprecht, 1964), lxxx. 7  Gunkel, Genesis, lxxxv. 8  J. S. B aden , J, E, and the Composition of the Pentateuch (FAT 72; Tübingen: Mohr Siebeck, 2009), 8–9 (see also 289, 305, and the detailed description on 255–286). Similarly, B. J. Schwartz, “How the Compiler of the Pentateuch Worked: The Composition of Genesis 37,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; VTSup 152; Leiden: Brill, 2012), 263–278; idem, “Josephʼs Descent into Egypt: The Composition of Genesis 37,” in The Joseph Story in the Bible and throughout the Ages (ed. L. Mazor; Beit Mikra 55; Jerusalem: Magnes, 2010), 1–30 (Hebrew). 9  H. Donner, “Der Redaktor: Überlegungen zum vorkritischen Umgang mit der Heiligen Schrift,” Henoch 2 (1980), 1–30; reprinted as idem , Aufsätze zum Alten Testament aus vier

Introduction

565

Redaction-critical analyses of the Pentateuch posit many more hands responsible for the development of the Pentateuch.10 One of the most excessive proposals is Christoph Bernerʼs analysis of Exod 1–15.11 Berner posits that Exod 1–15 can be described as a “Prozess ‘midraschartiger’ Selbstauslegung” (7). As a result, Berner finds in Exod 1 more than nine layers (48), in Exod 2–4 more than nine layers (135), in Exod 5 more than five layers (151–152), in Exod 6–7 more than four layers (166–167), in Exod 7–9 more than fourteen layers (212–214), in Exod 9–11 more than thirteen layers (264, 266), in Exod 12 more than nineteen layers (339–342), and in Exod 13–15 more than fourteen layers (403–405). This result is further complicated by the fact that the different layers in the different textual segments cannot necessarily be identified with each other (9). Thus, there are two extremes represented in current scholarship: one hand or an abundant number of hands is responsible for the formation of the Pentateuch. Thus, the most prominent common feature of current redactional criticism on the Pentateuch is its stunning diversity. Apparently, there is a real need for pentateuchal theory to base itself onto more solid ground in this respect. Of course, it would be most welcome to base redaction-critical theories regarding the Pentateuch on external evidence.12 But as is well known, no manuscripts of the Pentateuch from the biblical period exist. Nevertheless, there are at least some hints in terms of evidence that may be evaluated diachronically, and they stem from the textual diversity of the different versions of the Pentateuch (Masoretic Text, Septuagint, Samaritan, Qumran, etc.).13 Of course, it would be too bold to claim that it is always possible to evaluate the differences between these versions in unambiguous, diachronic terms. But the different versions show that there was at least some fluidity in the process of textual transmission of the Jahrzehnten (BZAW 224; Berlin: de Gruyter, 1994), 259–285. See also Schwartz, “How the Compiler of the Pentateuch Worked,” 263–278. 10  See, e.g., C. Levin, Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993); R. G. Kratz, The Composition of the Narrative Books of the Old Testament (trans. by J. Bowden; London: Continuum, 2005). 11  C. Berner, Die Exoduserzählung: Das literarische Werden einer Ursprungserzählung Israels (FAT 73; Tübingen: Mohr Siebeck, 2010). 12  Cf. J. H. T igay (ed.), Empirical Models for Biblical Criticism (Philadelphia: Fortress, 1985); idem, “An Empirical Basis for the Documentary Hypothesis,” JBL 94 (1975), 329–342. 13  Cf. J. W. W evers , “The Interpretative Character and Significance of the Septuagint Version,” in Hebrew Bible/Old Testament: The History of Its Interpretation (ed. M. Sæbø; Göttingen: Vandenhoeck & Ruprecht, 1996), 1.1:84–107; A. Lange, Handbuch der Textfunde vom Toten Meer, Band 1: Die Handschriften biblischer Bücher von Qumran und den anderen Fundorten (Tübingen: Mohr Siebeck, 2009); G. N. K noppers , “Parallel Torahs and Inner-Scriptural Interpretation: The Jewish and Samaritan Pentateuchs in Historical Perspective,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 507–531; M. M. Zahn, “The Samaritan Pentateuch and the Scribal Culture of Second Temple Judaism,” JSJ 46 (2015), 285–313.

566

Konrad Schmid

Pentateuch and, in some instances, it is indeed possible to draw some plausible conclusions regarding some specific redactional activities in the Pentateuch.14 As the text of the Pentateuch apparently stabilized earlier than did the prophetic books and the writings,15 the textual differences are often minor in nature. Despite its modest nature, however, some evidence is nevertheless available and needs to be taken into account in reconstructing a redaction history of the Pentateuch. The contributions in this part will discuss the possible evidence for redactional activity in the Pentateuch and also provide an overview of the history of research of its latest, i.e., post-Priestly additions.

14  Cf. the approach of J. Popper, Der biblische Bericht über die Stiftshütte: Ein Beitrag zur Geschichte der Composition und Diaskeue des Pentateuch (Leipzig: Hunger, 1862). 15  Cf. T. H. L im , The Formation of the Jewish Canon (New Haven, CT: Yale University Press, 2013).

Some Empirical Evidence in Favor of Redaction Criticism Jean Louis Ska The historical books of the Old Testament in general did not come into being in as mechanical a manner as – contra Ewald – is generally believed. In the Pentateuch too, there are not two or several great historical complexes with the same subject, originally written independently of each other in such a way that the later one takes no account of the one written earlier. It is rather that sometimes smaller blocks were joined to a single core (note: or were probably also assimilated into it), as Gen. 4 is joined to Gen. 2–3; into this [core] for the first time the individual stories which until then existed in oral or written form were fitted [. . .] the whole sometimes being newly worked over as a fresh complex, perhaps in such a way that from the beginning its essential content was incorporated into the new revision, or in such a way that only the bare outlines of its plan determined this revision, thus making it possible for a later redactor to combine the old and the new – there is much to be said in favor of both possibilities. At all events, modifications of the original core and the revisions of shorter passages, changes in individual words, and minor interpolations (Gen. 3.20) are indissolubly bound up with the way in which the historical books developed, and it is difficult to find the dividing line where literary criticism ceases and textual criticism begins.1

This long quotation comes from Julius Wellhausen, who is often presented as the most prominent herald of the Documentary Hypothesis, the personality that most contributed to the triumph of the latter theory over against the Fragmentary and the Supplementary Hypotheses. There are, indeed, good reasons to think so. It is, however, possible to find in Wellhausenʼs writings positions much more nuanced than the presentation of his theory in common and classical introductions to the Pentateuch. In some of his writings, as in the above quotation taken from a short commentary on the books of Samuel, we also find affirmations very close to the Supplementary Hypothesis on the one hand and similarly close to the proponents of the more recent Redaktionsgeschichte on the other. The latter “school” or method has come under vehement attack in very recent years. First, John Van Seters cast doubts on the very existence of “redactors” and “editors” in antiquity.2 The words redactor and editor were coined during the 1  J. Wellhausen, Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Ruprecht, 1871), x–xi. 2  J. Van S eters , The Edited Bible: The Curious History of the “Editor” in Biblical Criticism (Winona Lake, IN: Eisenbrauns, 2006); cf. also idem, “Author or Redactor?” JHebS

568

Jean Louis Ska

Renaissance, in the study of ancient manuscripts, and are applied only anachronistically to the study of the composition of ancient texts. Therefore, John Van Seters pleads in favor of a theory that supposes the existence of real authors and writers at the origin of the biblical books. This is the case, especially, with respect to the Yahwist and the Deuteronomist. The other vehement attack comes from the so-called Neo-Documentarians, who plead, sometimes forcefully, on behalf of a return to the Documentary Hypothesis.3 The two main tenets of their position are the following: (1) The Documentary Hypothesis is the most satisfactory solution to the problems of the Pentateuch, since it is a literary solution to a literary problem. (2) According to this new formulation of the Documentary Hypothesis, four independent documents (J, E, D, and P) were compiled by one redactor. All historical or chronological problems are set aside since we do not have any means for solving them. In this short paper, I would like to show that there is room for more nuanced assumptions than those just exposed. In other words, Wellhausen was not far off the mark when he said that the border between text criticism and literary criticism is not easy to trace. My contention is also that the modern search for redactional additions, modifications, instances of reorganization, or reinterpretation of former versions of a text has an empirical basis.4 The comparison between the Masoretic Text (MT) and the Samaritan Pentateuch (SP), the Septuagint (LXX), and the Qumran fragments shows that copyists and scribes worked with a certain freedom. To put it simply, they did not work as our modern photocopying machines do and were not supposed to work that way. I will take a few examples from the book of Genesis and the book of Numbers.

1 Genesis 12:6: MT and LXX In the MT, Gen 12:6 reads: ‫וַ ּיַ ֲעבֹר ַא ְב ָרם ָּב ָא ֶרץ ַעד ְמקֹום ְׁש ֶכם‬, “And Abraham passed through the land until the place [called] Shechem.” The LXX translates literally but adds a few words: καὶ διώδευσεν Αβραμ τὴν γῆν εἰς τὸ μῆκος αὐτῆς, “And Abraham passed through the land to its length [. . .].” The words εἰς τὸ μῆκος αὐτῆς most probably come from Gen 13:17, an oracle of YHWH to Abram 7 (2007), 1–22, http://www.jhsonline.org/Articles/article_70.pdf (accessed 6/1/13). 3  See especially J. S. Baden, J, E, and the Redaction of the Pentateuch (FAT 68; Tübingen: Mohr Siebeck, 2009); idem, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven, CT: Yale University Press, 2012); idem, The Promise to the Patriarchs (Oxford: Oxford University Press, 2013); see also on this problem, T. B. Dozeman et al. (eds.), The Pentateuch: International Perspectives on Current Research (FAT 78; Tübingen: Mohr Siebeck, 2011). 4  Along the same lines, see S. W. Crawford, Rewriting Scripture in Second Temple Times (Grand Rapids, MI: Eerdmans, 2008).

Some Empirical Evidence in Favor of Redaction Criticism

569

after the latter had separated from Lot: “Arise, go through the land, to its length (‫)לארכה‬, to its breadth (‫)לרחבה‬, for to you I will give it.” The LXX translates this verse faithfully: ἀναστὰς διόδευσον τὴν γῆν εἴς τε τὸ μῆκος αὐτῆς καὶ εἰς τὸ πλάτος ὅτι σοὶ δώσω αὐτήν. What is the reason for the LXXʼs addition in Gen 12:6? It seems to me that it is of a juridical nature. As David Daube demonstrated some time ago, walking through a piece of land in certain circumstances means taking possession of this piece of land de iure.5 This is attested in Roman law, for instance in this passage of Justinianʼs Digest 41.2.3.1: Quod autem diximus et corpore et animo adquirere nos debere possessionem non utique ita accipiendum est, ut qui fundum possidere velit, omnes glebas circumambulet: sed sufficiet quamlibet partem eius fundi introire, dum mente et cogitatione hac sit, ut totum fundum usque ad terminem velit possidere. What we have just said, i.e., that we must acquire a property et corpore et animo [materially and intentionally], does not mean that the person that wants to acquire a piece of land must go through it in its entirety [must tread every patch of it]: it is sufficient to enter into any part of the land with the intention of acquiring the plot of land up to its limits.

In the Hebrew Bible, the clearest example, in my opinion, is to be found in Num 13, from the Priestly Writer, where God asks Moses to send twelve men, one from each tribe, to “walk all over the land” (Num 13:2, 17). This is a way of taking possession of the land de iure, and therefore we have performative language in Num 13:2, where the participle is used: “Send men to walk all over (‫ )ויתורו‬the land of Canaan, which I am giving (‫ )אני נתן‬to the people of Israel. From each tribe of their fathers you shall send a man, everyone a chief among them.”6 The explorers take possession of the land de iure when walking all over it. This is the reason why the Priestly version of the text insists on their walking through the whole land for forty days (Num 13:21). In my opinion, this is also the reason why YHWH invites Abram to go through the land in Gen 13:17. After separating from Lot, YHWH gives the land of Canaan to Abram and the latter takes possession of it de iure when walking through it, to its length and to its breadth. 5  D. D aube , Studies in Biblical Law (Cambridge: Cambridge University Press, 2008), 37–39. 6  On this text, see N. Lohfink, “Die Landverheißung in Numeri und das Ende der Priesterschrift: Zu einem rätselhaften Befund im Buch Numeri,” Studien zum Deuteronomium und zur deuteronomistischen Literatur V (SBAB 38; Stuttgart: Katholisches Bibelwerk, 2005), 273– 292; on performative language, see Joüon-Muraoka § 112f; see also A. Wagner, Sprechakte und Sprechaktanalyse im Alten Testament: Untersuchungen im biblischen Hebräisch an der Nahtstelle zwischen Handlungsebene und Grammatik (BZAW 253; Berlin: de Gruyter, 1997); M. F. Rogland, Alleged Non-Past Uses of Qatal in Classical Hebrew (SSN 44; Assen: Van Gorcum, 2003).

570

Jean Louis Ska

The LXX transfers a short part of this divine discourse into the narration in Gen 12:6 to give this passage a more juridical dimension. As soon as Abram entered the land, he walked through it and took possession of it de iure immediately. In other words, Abram entered the land as proprietor although the Canaanites were still present in the country, as is said in the same verse, in Gen 12:6b. Most probably, this note at the end of the verse brought about the intervention of the Greek translator. The LXX wants to make sure from the start that Abram is the real owner of the land in spite of the presence of Canaanite populations. The difference between Gen 12:6 MT and Gen 12:6 LXX is minimal. It is also revelatory of the translatorʼs freedom to interpret some texts in the light of others and to give them a new meaning.7

2 Genesis 12:16a: MT and SP Genesis 12:16a is part of the narrative about Abramʼs sojourn in Egypt. When Pharaoh takes Sarai into his harem, he showers gifts on Abram. The list in the MT comprises “sheep, oxen, male donkeys, male servants, female servants, female donkeys, and camels.” The Samaritan Pentateuch adds three words to this impressive list: ‫מקנה כבד מאד‬, “much livestock.” This seems redundant since Abram receives from Pharaoh almost anything that a cattle breeder may expect to possess at that time. What is the reason for this addition? These three words are present in another narrative that is set in Egypt, namely, in Exod 12:38b. When the Israelites leave Egypt after the death of the firstborn, they take with them “small cattle (‫)צאן‬, big cattle (‫)בקר‬, [i.e.,] much livestock (‫)מקנה כבד מאד‬.” The first two words are present in Gen 12:16b MT and in Exod 12:38: ‫צאן ובקר‬, “small cattle and big cattle,” i.e., “sheep and oxen.” The SP completed the list in Gen 12:16b with the three words present in Exod 12:38. This intervention of the SP can be interpreted as an attempt at harmonizing both texts. But there is perhaps more to it than that. It is possible to detect in this addition within the SP an attempt to bring out the parallelism between Abramʼs sojourn in Egypt and the exodus of the Israelites. Several features are common to Gen 12:10–20 and Exodus: Abram goes down to Egypt because of a famine (Gen 12:10); Pharaoh is struck with plagues (12:17, ‫ ;נגעים‬cf. Exod 11:1); and he “sends off” or “expels” (‫וישלחו‬, “and [Pharaoh] sent [Abram] away”) Abram,

7  For more details on this text, see J. L. Ska, “Gen 12,6 et son interprétation juridique dans la LXX,” in A Pillar of Cloud to Guide: Text-critical, Redactional, and Linguistic Perspectives on the Old Testament in Honour of Marc Vervenne (ed. H. Ausloos and B. Lemmelijn; BETL 269; Leuven: Peeters, 2014), 449–454.

Some Empirical Evidence in Favor of Redaction Criticism

571

who goes back to the promised land (12:20).8 To these features, SP adds another one, namely, that Abram acquired “much livestock” in Egypt, as it is said that the Israelites possessed “much livestock” when they left Egypt. This is well summarized by Nachmanides (Ramban) in his commentary on Gen 12:10–20: “Everything that happened to the fathers happened to the sons,” or, more specifically, “when an event happens to any one of the three patriarchs, that which is decreed to happen to his children can be understood.”9 Nachmanides himself is relying on Genesis Rabbah.10

3 Genesis 47:5: MT and LXX11 This case had already been considered by J. Wellhausen in his Composition des Hexateuchs.12 His theory was returned to later by M.-J. Lagrange in a study that was meant to be part of a commentary on the book of Genesis. That commentary was never published, however, because of difficulties with authorities of the Catholic Church.13 Genesis 47:1–12 describes in two different ways the meeting of Jacobʼs family with Pharaoh. In the first version, Joseph goes to Pharaoh and introduces five of his brothers, who ask for permission to sojourn in the land of Goshen with their flocks. Pharaoh agrees and even asks that some of them take charge of his own flocks (47:1–5a). In the second version, Jacob arrives in Egypt with his family, and Pharaoh comes to know of his arrival before Joseph does. The king of Egypt announces to Joseph that his family has arrived. Joseph introduces his father, Jacob, to Pharaoh. Pharaoh inquires about Jacobʼs age and, after the conversation, Joseph 8  See, among many others, E. Blum, Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 309–311; T. R ömer , “Exodusmotive und Ex­odus­polemik in der Erzvätererzählungen,” in Berührungspunkte: Studien zur Sozial- und Re­li­gionsgeschichte Israels und seiner Umwelt, Festschrift für Rainer Albertz zu seinem 65. Geburts­tag (ed. U. Kottsieper et al.; AOAT 350; Münster: Ugarit-Verlag, 2008), 3–20. 9  Nachmanides (Ramban; 1194–1270), in his commentary on Gen 12:6, and quoted in M. Z. B rettler , The Creation of History in Ancient Israel (London: Routledge, 1995), 48, 52–53. For the text, see Ramban (Nachmanides), Commentary on the Torah. (trans. C. B. Chavel; New York: Shilo Publishing House, 1971), 169. 10  Gen. Rab., 40:6, quoted in Brettler, Creation of History (see n. 9), 52. The first sentence quoted by Nachmanides can also be found in Midrash Tanhuma, Lekh Lekha, 9. 11  For a full treatment of this text, let me refer to F. Giuntoli, Lʼofficina della tradizione: Studio di alcuni interventi redazionali post-sacerdotali e del loro contesto nel ciclo di Giacobbe (Gen 25,19–50,26) (AnBib 154; Roma: PIB, 2003), 309–340. 12  J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (3rd ed.; Berlin: de Gruyter, 1963 [orig. pub., Berlin: Georg Reimer, 1899]), 51–52. 13  M.-J. Lagrange, “Les sources du Pentateuque,” RB 7 (1898), 10–32, here 15–16.

572

Jean Louis Ska

settles his brothers in “the best of the land,” or the “land of Rameses,” according to Pharaohʼs indications (47:5b–12). Briefly, the main differences are three: (1) In the first version, Joseph announces to Pharaoh that his father has arrived with his family; in the second, this happens the other way round, since Pharaoh announces the news to Joseph. (2) In the first version, five of the brothers ask for permission to sojourn in Egypt with their flocks; in the second, Pharaoh proposes to settle Jacob and his family in the best region of Egypt. (3) In the first version, the brothers ask to settle in the land of Goshen and Pharaoh agrees, whereas in the second, Joseph settles his family in the best part of the land (or in Rameses), according to Pharaohʼs instructions. In short, in the first version, the initiative is that of Josephʼs brothers, whereas in the second it is Pharaohʼs. Scholars defending the classical Documentary Hypothesis attribute the first version to the Yahwist and the second to the Priestly Writer. There is however another problem in the MT. The text of Gen 47:5–6 reads: And Pharaoh said to Joseph: “Your father and your brothers have come to you. The land of Egypt is before you. In the best part of the land settle your father and your brothers. Let them settle in the land of Goshen [. . .].”

This creates a serious problem of understanding. The five brothers – introduced by Joseph to Pharaoh – ask Pharaoh to let them settle in Goshen in 47:4. According to the MT, in 47:5, Pharaoh himself proposes to Joseph that he settle the whole family in the land of Goshen. Afterward, in 47:11, Joseph settles his family in the land of Rameses. The problem is double. First, the question asked by the brothers in 47:4 is answered by Pharaoh in a different scene. Second, in the same scene, the same Pharaoh grants the same brothers two different places to settle: the land of Goshen in 47:6b and the land of Rameses in 47:11. Now the Greek version of the same passage is more satisfactory. The text is somewhat longer and in a slightly different order, but the two versions of the brothersʼ arrival in Egypt are separated more neatly: Genesis 47:5–6 εἶπεν δὲ Φαραω τῷ Ιωσφη κατοικείτωσαν ἐν γῇ Γεσεμ εἰ δὲ ἐπίστῃ ὅτι εἰσὶν ἐν αὐτοῖς ἄνδρες δυνατοί κατάστησον αὐτοὺς ἄρχοντας τῶν ἐμῶν κτηνῶν ἦλθον δὲ εἰς Αἴγυπτον πρὸς Ιωσηφ Ιακωβ καὶ οἱ υἱοὶ αὐτοῦ καὶ ἤκουσεν Φαραω βασιλεὺς Αἰγύπτου καὶ εἶπεν Φαραω πρὸς Ιωσηφ λέγων ὁ πατήρ σου καὶ οἱ ἀδελφοί σου ἥκασι πρὸς σέ ἰδοὺ ἡ γῆ Αἰγύπτου ἐναντίον σού ἐστιν ἐν τῇ βελτίστῃ γῇ κατοίκισον τὸν πατέρα σου καὶ τοὺς ἀδελφούς σου [. . .]. And Pharaoh said to Joseph: “Let them settle in the land of Geshem. If you know that there are among them capable men, appoint them in charge of my livestock.” Jacob and his sons came to Joseph in Egypt and Pharaoh, king of Egypt, heard about it and Pharaoh said to Joseph: “Your father and your brothers have come to you. See, the land of Egypt is before you. Settle your father and your brothers in the best part of the land [. . .].”

Some Empirical Evidence in Favor of Redaction Criticism

573

The primary differences between the MT and the LXX are two. First, the Greek text contains a plus, namely, “Jacob and his sons came to Joseph in Egypt and Pharaoh, king of Egypt, heard about it and Pharaoh said to Joseph,” which is found at the end of v. 5. Second, the order of the text is different. We have in the LXX the sequence 5a, 6b, 5b, 6a (according to the versification of the MT). Different explanations have been proposed. Some consider that the LXX, as in other cases, clarifies and harmonizes a difficult Hebrew text. Others consider that the LXX reflects an original and more ancient Hebrew text and that the MT tried to harmonize the two conflicting versions. The second version seems to be preferable, although the MT can claim to be the lectio difficilior.14 For this and other reasons, some commentators, such as Claus Westermann for instance, prefer the MT, arguing that Gen 47:1–6 describes a very complicated audience of Joseph and his brothers with Pharaoh.15 First, Pharaoh receives Joseph, who introduces his family (v. 1). Then Pharaoh speaks to five of Josephʼs brothers about their trade and the need of pasture-grounds (vv. 2–4), in a kind of short parentheses. Verse 5 picks up where v. 1 left off, and we now have a conversation between Pharaoh and Joseph about the fate of the latterʼs family (vv. 5–6). This may be so, but the explanation is very complicated and does not solve all the problems of the text. It is not clear why we have two introductions and why Pharaoh twice offers the same solution to the same problem. Biblical narrative style is generally more straightforward. The main argument in favor of the originality of the LXX is that it is easier to understand why the MT changed the order of the verses and canceled part of the text rather than the opposite, namely, that the LXX added a sentence and rearranged the text to make it smoother. Generally the LXX works in a different way. Moreover, if one supposes that the Greek translators had the MT before their eyes, it is difficult to understand the way they managed to arrive at the Greek text we know, especially because their version is now flawless. These translators had to be geniuses to be able to recreate two parallel, complete, and competing versions of Jacobʼs arrival in Egypt. On many other occasions, the LXX reproduces the difficulties of the MT or tries to harmonize conflicting versions. It is much easier to understand the reason why the Hebrew scribes changed the original Hebrew text and canceled an evident difficulty. The reason for their intervention in this case is simple. Jacob arrives in Egypt with his whole family, 14  See, for instance, the discussion in C. Westermann, Genesis 37–50 (BKAT 1/3; Neukirchen-Vluyn: Neukirchener Verlag, 1982), 187–188, with bibliography. Westermann defends the MT, the hebraica veritas, as original, against the LXX. For a more recent bibliography, see Giuntoli, Officina (n. 11), 179, n. 57. 15  See Westermann, Genesis 37–50 (see n. 14), 187–188, following W. Rudolph, “Die Josefsgeschichte,” in Der Elohist als Erzähler: Ein Irrweg der Pentateuchkritik? (ed. P. Volz and W. Rudolph; BZAW 63; Gießen: Töpelmann, 1933), 165–166.

574

Jean Louis Ska

who will stay there for centuries. The scribes tried by all means to reconcile the brothersʼ and Pharaohʼs proposals to make sure that in both cases they settle in Goshen, the place mentioned in Gen 45:10; 46:28, 34, and later in the book of Exodus. It appears in our text, in Gen 47:1, 4, 6, 27; afterwards in Gen 50:8; and in Exod 8:18; 9:26; cf. Josh 10:41. The Priestly Writer is less precise and speaks only of “the best part of the land.” Rameses, instead, is probably a late addition. The name reappears only in Exod 1:11, as a name of a city built under the pharaoh of the oppression, and as the first stage of the journey of Israel when leaving Egypt (Exod 12:37; 33:3.5). The purpose of the addition is probably to identify the place where the first Israelites settle with the point of departure of the exodus a couple of centuries later.16 The most important point is that there are evident differences between the MT and the LXX: this means that either the copyists of the Hebrew text or the Greek translators felt free to reorganize or to modify the text. Neither copying nor translating is supposed to be a mechanical operation. In short, copying was not photocopying.

4 Numbers 10:34–36: MT and LXX This text speaks of the first stages in the wilderness after the long stay of Israel at Mount Sinai.17 When the ark leaves the camp, Moses intones a song (v. 35), and he intones another song when the ark stops (v. 36): Numbers 10:35–36 And whenever the ark set out, Moses said, “Arise, YHWH, and let your enemies be scattered, and let those who hate you flee before you.” And when it rested, he said, “Return, YHWH, to the ten thousand thousands of Israel.”

In the MT, Num 10:35–36 are enclosed in so-called inverted nuns. The meaning of these Masoretic signs is debated, but it is clear that ancient copyists were aware of the problematic position of these verses. Either the text comes from another source or it is out of place.18 The same signs exist in Greek manuscripts 16  On this point, see Giuntoli, Officina (see n. 11), 203–208, for whom the use of Rameses in Gen 47:11 is an anachronism. 17  For an extensive treatment of this passage, see, among others, T. B. D ozeman , “The Midianites in the Formation of the Book of Numbers,” in The Books of Leviticus and Numbers (ed. T. Römer; BETL 215; Leuven: Peeters, 2008), 261–284, here 272–275. For the problem of sources, see the survey on p. 275; for a late date, see, among others, R. Achenbach, Die Vollendung der Tora: Studien zur Redaktionsgeschichte des Numeribuches im Kontext von Hexateuch und Pentateuch (BZABR 3; Wiesbaden: Harrassowitz, 2003), 202. 18  See, among others, B. A. Levine, Numbers 1–20: A New Translation with an Introduction and Commentary (AB 4; Garden City, NY: Doubleday, 1993), 317–318. For the possible mean-

Some Empirical Evidence in Favor of Redaction Criticism

575

and are called περιγραφή, παραγραφή, or ἀντίσιγμα, i.e., inverted sigma.19 In Greek manuscripts, these signs were intended to suggest that the text marked in this way had to be removed because it was interpolated. This is also what ancient Jewish commentators noticed, as is attested in Sifre Numbers 84: “When the ark was to set out [. . .]” There are dots above and below [this pericope] to indicate that this was not its correct place. Rabbi says, “This is because the pericope at hand constitutes a scroll unto itself.” [. . .] Rabbi Simeon says, “In the written version there are dots above and below [this pericope] to indicate that this was not its correct place.” And what ought to have been written instead of this pericope? “And the people complained in the hearing of the Lord” (Nm 11:1ff.).20

Ancient Jewish commentaries were already aware of possible later additions to or interpolations into biblical texts, and this quotation is an unmistakable example of rabbinic critical sensitivity announcing what is today called redaction criticism. As for Rabbi Simeon, the original text did not contain these verses, and the reader had to skip from Num 10:34 to Num 11:1. Modern exegetes take the same line and pinpoint other cases of the same phenomenon. This example justifies the basic principle of redaction criticism, but surely not its excesses. In this field as in others, one should remember the old Latin saying, abusus non tollit usus, i.e., in a free translation, “the abuses of a method do not condemn the sensible use of this method.” There is another problem when comparing MT with LXX. The order of verses in this text is different, and this confirms our doubts about the origin of the song in Num 10:35–36.21 These verses are preceded, in the MT, by a short notice about a three-day journey in the wilderness (10:33) and by another notice in v. 34: “And the cloud of YHWH was over them by day, whenever they set out from the camp.” In the LXX this latter verse is found after v. 36, namely after the two songs. ing of the inverted nuns, see S. Z. Leiman, “The Inverted nuns at Numbers 10:35–36 and the Book of Eldad and Medad,” JBL 93 (1974), 348–355; E. Tov, Textual Criticism of the Hebrew Bible (2nd ed.; Minneapolis, MN: Augsburg Fortress Press, 2001), 54–55. These scribal dots were mistakenly understood as inverted nuns. 19  For a more complete explanation, see S. L ieberman , Hellenism in Jewish Palestine: Studies in the Literary Transmission, Beliefs and Manners of Palestine in the I Century B.C.E.–IV Century C.E. (Texts and Studies of the Jewish Theological Seminary of America 18; New York: The Jewish Theological Seminary of America, 1950), 38–43, quoted by T ov , Textual Criticism (see n. 16), 54. 20  See J. Neusner, Sifré to Numbers: An American Translation and Explanation (Atlanta: Scholars Press, 1986), 80, quoted by Tov, Textual Criticism (see n. 16), 55. See also b. Shabbat 115a–116a. 21  On this problem see, among others, Tov, Textual Criticism (see n. 16), 339, for whom “the sequence in G*, referring to the Ark, comes immediately after v. 33, where the Ark is also mentioned in this connection, is probably more natural.”

576

Jean Louis Ska

Which is the more logical or – and this is a different question – the more original text? It seems, at first sight, that the order of verses is more logical in the LXX than in the MT. Numbers 10:33 describes a three-day journey of Israel in the wilderness, guided by the ark of the covenant. It is more natural to explain immediately afterwards what happens every time the ark sets out or comes to rest, since it is the first time the Israelites are on the move after their long stay at Mount Sinai, and this happens according to the instructions received previously in Num 9:15–23. Numbers 10:35–36 explains what Moses has to do when departing and when arriving and what he did this time in connection with the ark. The introduction of the song explicitly mentions the ark (v. 35a). The notice about the presence of the cloud in Num 10:34 MT is less important since it does not speak of the ark. For this reason, the order of the verses in the LXX seems to be preferable because it is more logical. How is the order of verses in the MT to be explained? It seems plausible to see the connection in the use of the verb ‫נסע‬, “to move (the camp),” present twice in 10:33 and once in 10:34. This is indeed a key word in this context. Besides this, we find the word ‫יום‬, “day,” used in different forms in these two verses as well.22 Be that as it may, we can draw another conclusion from this short analysis. Most probably, the song was a well-known melody, since we find it in Ps 68:2. The redactors of the book of Numbers found it fitting to introduce it here, in the context of the first stage of Israelʼs journey through the wilderness after the long stay at Mount Sinai. But the redactors of the MT and the Vorlage of the LXX made different choices. In the MT, the stylistic links were more important, whereas the logic of the text and its content seem to have prevailed in the Vorlage of the LXX. Again, copyists and/or translators enjoyed a certain freedom. This is another empirical example of slight changes between texts and a rationale for the view that this phenomenon is neither rare nor impossible in biblical literature. There are definitely many examples of this kind, but these are sufficient, it seems to me, to illustrate the point that Wellhausen, whom I quoted at the beginning of this article, wanted to drive home.23 These are empirical examples 22  Tov , Textual Criticism (see n. 16), 339–340, mentions other examples where the sequence of verses or chapters is different in MT and LXX: In LXX the order of 1 Kgs 20 and 21 is reversed; Solomonʼs blessing in 1 Kgs 8:12–13, at the dedication of the temple, follows v. 53 in LXX; Jer 23:7–8 appears after v. 40 in LXX. The most conspicuous example is, of course, the order of chapters in the central part of Jeremiah, the oracles against the nations. In the MT, they are found in Jer 46–51, whereas they appear after chapter 25 in LXX. There are many discussions on this topic. See a summary in G. Fischer, Jeremia: Der Stand der theologischen Diskussion (Darmstadt: Wissenschaftliche Buchgesellschaft, 2007); P.-M. Bogaert, “De la vetus latina à lʼhébreu pré-massorétique en passant par la plus ancienne Septante: Le livre de Jérémie, exemple privilégié,” RTL 44 (2013), 216–243. 23  For examples of this kind in the book of Exodus, see J. E. Sanderson, An Exodus Scroll from Qumran: 4QpaleoExodM and the Samaritan Tradition (HSS 30; Atlanta: Scholars Press,

Some Empirical Evidence in Favor of Redaction Criticism

577

of the practices of scribes and copyists when dealing with sacred texts. We have good reasons, therefore, to believe that the same phenomena can be found in the Hebrew Bible itself.24 Moreover, again according to Wellhausen, the dividing line between textual criticism and literary criticism is often fluid.

1986). For studies on this topic, see, among many others, A. Schenker and P. Hugo (eds.), Lʼenfance de la Bible hébraïque: Lʼhistoire du texte de lʼAncien Testament à la lumière des recherches récentes (MdB 52; Geneva: Labor et Fides, 2005); A. Schenker (ed.), The Earliest Text of the Hebrew Bible: The Relationship between the Masoretic Text and the Hebrew Base of the Septuagint Reconsidered (SBL Septuagint and Cognate Studies Series 52; Leiden: Brill, 2003); idem , Anfänge der Textgeschichte des Alten Testaments: Studien zu Entstehung und Verhältnis der frühesten Textformen (BWANT 194; Stuttgart: Kohlhammer, 2011); E. Ulrich, The Dead Sea Scrolls and the Origins of the Bible (Studies in the Dead Sea Scrolls and Related Literature; Grand Rapids, MI: Eerdmans, 1999); E. Tov , The Text-Critical Use of the Septuagint in Biblical Research (Jerusalem: Simor, 1997); idem, The Greek and Hebrew Bible: Collected Essays on the Septuagint (SVT 72; Leiden: Brill, 1999); idem, Scribal Practices and Approaches Reflected in the Texts Found in the Judean Desert (STDJ 54; Leiden: Brill, 2004); idem , Hebrew Bible, Greek Bible, and Qumran: Collected Essays (TSAJ 121; Tübingen: Mohr Siebeck, 2008). 24  For more information in this field, see the classic work by J. H. Tigay (ed.), Empirical Models for Biblical Criticism (Philadelphia, PA: University of Pennsylvania Press, 1985) and, from a different perspective, the more recent work by D. M. Carr, Writing on the Tablet of the Heart: Origins of Scripture and Literature (Oxford: University Press, 2005).

The Pentateuch A Compilation by Redactors Christoph Levin The Starting Point There can be no literary-critical analysis without more or less consciously presuming a certain literary-historical synthesis. My own presumption is that the Hebrew Bible was by its very nature the traditional religious text of the Jewish community of the Second Temple period. Not only was it transmitted and received as the holy Scripture of Second Temple Judaism, but it in fact came into being with precisely this status. The text had its beginnings in the holdings of the royal archives in Jerusalem as texts that happened to survive the conquest of the city and the destruction of the archives of the court and the temple. These sources consisted of relatively slim collections of prophetic sayings, cult poetry, and wisdom sayings; collections of law such as the core of the Covenant Code and the core of Deuteronomy; the synchronistic excerpt from the annals of the kings of Israel and from the annals of the kings of Judah that later provided the historical scaffolding for the books of Kings; and cycles of stories, mostly about heroes and kings, that had been passed down at the court and are now found in the books of Judges, Samuel, and Kings. One may also include among these texts the stories about the fortunes of certain families that were transmitted within the patriarchal narratives in the book of Genesis. After the downfall of the monarchy in Judah, two redactional compilations came into being as a way of coming to terms with the new situation. As Martin Noth stated: “The whole of the historical tradition in the Old Testament is contained in a few large compilations. These works have collated and systematized the extremely diverse material of traditional tales and historical reports and enclosed them in a framework, determined in each case by their own particular concerns.”1 One of these compilations, known today as the Deuteronomistic History, was the historical work that now constitutes the nucleus of the books of Joshua, Judges, Samuel, and Kings. It was compiled and written 1  M. N oth , The Deuteronomistic History (trans. J. Doull et al.; JSOTSup 15; Sheffield: Sheffield Academic Press, 1981), 1.

580

Christoph Levin

in Judah and propagandizes the return of the Davidic monarchy.2 Another of these compilations was the history forming the basis of Genesis, Exodus, parts of Numbers, and the close of Deuteronomy. In the tradition of biblical studies, this work is referred to as the Yahwistʼs history (J). Its literary shape has been uncovered in more recent times by means of redactional criticism. It was probably written in Babylon, as it grapples with the experience of the exile and the dispersion.3 Later on, a third compilation, usually referred to as the Priestly code (P) in modern scholarship, repeated the outline of the Yahwistʼs history.4 Because it depends heavily on the books of the Prophets in their edited form, above all on the book of Ezekiel, it cannot have been written before the middle of the fifth century, probably in Jerusalem. At an unknown later date, the Priestly code was combined with the Yahwistʼs history in order to present one single, unified history of the early times of the people of God, beginning with the creation of the world. The compilation of P and J is what forms the subject of the Documentary Hypothesis. Ever since this hypothesis was first published by Jean Astruc in 1753, it has provided the basis of modern pentateuchal criticism.5 One century later, in 1853, Hermann Hupfeld concluded: The discovery that the Pentateuch was composed of different sources or documents is indisputably not only most important for the understanding of the historical books of the Old Testament, with serious consequences for the concept of theology and history, but also one of the most secure discoveries in the whole field of critique and literary history. It shall maintain its hold and cannot be undone by anything as long as there is something like critique, that is, a sense of what is equal and what is contradicting, what is similar and what is dissimilar.6

 T. V eijola , Die ewige Dynastie: David und die Entstehung seiner Dynastie nach der deuteronomistischen Darstellung (AASF 193; Helsinki: Suomalainen Tiedeakatemia, 1975), rediscovered that the intention of the Deuteronomistic History was to favor the monarchy. 3  See C. Levin, Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993); and in shorter form, idem , “The Yahwist: The Earliest Editor in the Pentateuch,” JBL 126 (2007): 209–230; repr. in C. Levin, Re-Reading the Scriptures: Essays on the Literary History of the Old Testament (FAT 87; Tübingen: Mohr Siebeck, 2013), 1–23. 4  For recent research on P, see the overview by E. Otto, “Forschungen zur Priesterschrift,” TRu 62 (1997), 1–50; for the “Grundschrift” of P, see esp. T. P ola , Die ursprüngliche Priesterschrift: Beobachtungen zur Literarkritik und Traditionsgeschichte von Pg (WMANT 70; Neukirchen-Vluyn: Neukirchener Verlag, 1995). 5  J. Astruc, Conjectures sur les mémoires originaux dont il paroit que Moyse sʼest servi pour composer le Livre de la Genèse (Brussels: Fricx, 1753). 6  H. H upfeld , Die Quellen der Genesis und die Art ihrer Zusammensetzung (Berlin: Wiegandt & Grieben, 1853), 1 (my translation). 2

The Pentateuch

581

The Literary-Historical Process The compilation of the sources J and P is usually seen as the final redaction (R) of the Pentateuch. It would be more accurate, however, to designate the work as the redaction J/P, because it in fact formed a very early step in comparison with the literary history of the Pentateuch as a whole. The combined J/P-document was not the outcome of the Pentateuchʼs literary development but rather its basis. The body of text reached its current size during the Persian period. This process of growth did not come to an end until the beginning of the Hellenistic era. The guiding concern was to relate the transmitted text, which was understood as the word of God, to the changing conditions in the ongoing history of Judaism. This concern found expression through an inner-textual interpretation that overlaid the already-existing material with commentary and increasingly expanded the text. In this regard, the Hebrew Bible is to a great extent its own interpretation – we might say, a great midrash. The hypothesis that best does justice to this literary fact is the Supplementary Hypothesis. In understanding the literary history of the Bible, the Supplementary Hypothesis is the standard hypothesis that can be applied nearly everywhere. Generally speaking, the literary growth followed no rules. This was appropriate: a sacred text is not “made”; it is handed down from tradition and interpreted only for the needs of the present time. Redactional interventions such as organization and rearrangement of the written material were the exception.

The Conditions of the Compilation Regarding the compilation that first provided the literary basis of what eventually became the Pentateuch, it is fundamental that the sequence of events as it is presented in the outline of the Bible was not a matter of course. The entire narrative rests on a number of individual compositions. In the non-Priestly text, six separate narrative groups can be distinguished: (1) the primeval history, which presents the origin of the world and humankind; (2) the history of the patriarchs, Abraham, Isaac, and Jacob; (3) the story of Joseph and his brothers; (4) the narrative of Moses in Exod 2–4; (5) the history of the Exodus of the Israelites from Egypt and their wanderings through the desert in Exod 12–Num 20, to which the death of Moses may also have belonged (Deut 34*); and (6) the story about the seer Balaam (Num 22–24). All of these compositions are independent literary entities of individual origin and with a multiform previous history. All of them bear traces of editorial activity and literary growth. The diversity of the material indicates that it was only at a later stage that these groups were linked to form the continuous narrative we have today. It

582

Christoph Levin

is clear that the material was not randomly compiled but rather intentionally selected and assembled. This is evident from the fact that the story of the Pentateuch does not pertain to history but to prehistory. As Julius Wellhausen stated: “When the subject treated is not history but legends about pre-historic times, the arrangement of the materials does not come with the materials themselves, but must arise out of the plan of a narrator. [. . .] From the mouth of the people there comes nothing but the detached narratives. [. . .] To weave them together in a connected whole is the work of the poetical or literary artist.”7 We must, however, replace Wellhausenʼs “narrator” with “editor” or “redactor” on the basis of the literary nature of the former narrative compositions. In pentateuchal research of the mid-twentieth century, scholars explored the possibility that the sequence followed a traditional pattern, which supposedly had its Sitz im Leben in the memoria used in the cult. To a certain degree it is obvious – as Gerhard von Rad has stated – that “a gigantic structure such as this, the whole conforming to one single plan, does not grow up naturally of its own accord. How could such heterogeneous materials as those embraced by the Yahwist have cast themselves in this form of their own accord!”8 The problem is identified appropriately here, but the tradition-based historical solution that von Rad proposed has proved simply to be untenable. The credal formulae, above all Deut 26:5–9, which, it was assumed, provided the original structure for this pattern, have been proved definitively to be late summaries.9 They do not precede the redactional compilations; they presuppose them and are inconceivable without them. The cohesion of the whole can be explained only in light of redactional history. “J should be considered the product of redactional work that elaborated on older narrative cycles.”10 As a consequence, the literary identity of the Yahwist is to be seen in a different way from that of former scholarly tradition. The Yahwist is no longer a narrator or collector of oral tradition, as Hermann Gunkel11 and von Rad saw 7  J. Wellhausen, Prolegomena to the History of Israel (trans. J. S. Black and A. Menzies; Edinburgh: Black, 1885), 296. 8  G. von R ad , “The Form-Critical Problem of the Hexateuch,” in The Problem of the Hexateuch and Other Essays (trans. E. W. Trueman Dicken; London: SCM Press, 1984 [1st ed., 1966]), 1–78, here 52. 9  See W. Richter, “Beobachtungen zur theologischen Systembildung in der alttestamentlichen Literatur anhand des ‘kleinen geschichtlichen Credo,’ ” in Wahrheit und Verkündigung: Festschrift M. Schmaus (ed. L. Scheffczyk et al.; 2 vols.; Munich: Schöningh, 1967), 1:175–212; B. S. Childs, “Deuteronomic Formulae of the Exodus Traditions,” in Hebräische Wortforschung: Festschrift W. Baumgartner (ed. B. Hartmann; VTSup 16; Leiden: Brill, 1967), 30–39; N. L ohfink , Theology of the Pentateuch: Themes of the Priestly Narrative and Deuteronomy (trans. L. M. Maloney; Minneapolis: Augsburg Fortress, 1993), 265–289. 10  J. L. Ska, Introduction to Reading the Pentateuch (trans. P. Dominique; Winona Lake, IN: Eisenbrauns, 2006), 144. 11  H. Gunkel, Genesis (trans. M. E. Biddle; Macon, GA: Mercer University Press, 1997).

The Pentateuch

583

him12 and as John Van Seters still regards him.13 Rather, he is to be seen as an editor who dealt with written sources. We can grasp the editorʼs intention and his editorial technique by the selection of sources he made and by the peculiarities of his language, which enable us to clearly distinguish the editorial additions from the preeditorial literary level.14 It is more interesting here to consider the consequences for the Documentary Hypothesis, which derive from the fact that the making of the document J was a distinctive editorial step in the history of transmission. To produce documents like this depended on special preconditions. It is therefore very probable that the Priestly code took its outline from the Yahwistʼs history, whereas the so-called E source never existed at all, as has been shown by Wilhelm Rudolph: The reservations against source criticism, which I could not help thinking about more and more, do not turn toward the separation of sources in general: It seems to me unshaken and unshakeable that J is the oldest and P the youngest source of the Pentateuch. In my eyes, it is rather the way of separating the sources that would not do, that is, the method of sorting out verse after verse in every section of the text and attributing the single verses or even half verses or quarter verses to different written sources.15

It should be remembered that Jean Astruc, who first proposed the Documentary Hypothesis, had only two sources for the book of Genesis and ten fragments.

The So-Called Elohist (E) The Elohist (E) was first introduced – at least de facto – by Karl David Ilgen, who differentiated between Eliel Harischon and Eliel Haschscheni within the source that used the name elohim.16 The hypothesis was later elaborated by Hermann Hupfeld while he was clearing up the outline of Grundschrift, which we now, after the redating by Karl Heinrich Graf, Abraham Kuenen, and Julius Wellhausen, call the Priestly code (P).17 When it became evident to Hupfeld that  G. von R ad , Genesis (trans. J. H. Marks; rev. ed.; OTL; Philadelphia: Westminster, 1972). 13  J. Van Seters, Prologue to History: The Yahwist as Historian in Genesis (Louisville, KY: Westminster/John Knox Press, 1992); idem, The Life of Moses: The Yahwist as Historian in Exodus–Numbers (CBET 10; Kampen: Kok Pharos, 1994); idem , The Edited Bible: The Curious History of the “Editor” in Biblical Criticism (Winona Lake, IN: Eisenbrauns, 2006). 14  See Levin, Der Jahwist (see n. 3); idem, “The Yahwist: The Earliest Editor in the Pentateuch” (see n. 3). 15  W. R udolph , “Die Josefsgeschichte,” in Der Elohist als Erzähler: Ein Irrweg der Pentateuchkritik? (ed. W. R udolph and P. V olz ; BZAW 63; Gießen: Töpelmann, 1933), 143–183, here 145 (my translation). 16  K. D. Ilgen, Die Urkunden des ersten Buchs von Moses in ihrer Urgestalt: Zum bessern Verständnis und richtigern Gebrauch derselben (Halle: Hemmerde & Schwetschke, 1798). 17  Hupfeld, Die Quellen der Genesis (see n. 6). 12

584

Christoph Levin

a remarkable quantity of the elohim texts did not belong to this older Elohist, it seemed only a matter of course to attribute it to a source of its own, the younger (and later, after the redating, older) Elohist. Paul Volz and Wilhelm Rudolph, however, rightly declared that the conjectured Elohist was an error in Pentateuchal criticism.18 The existence of this source has been supposed only because the division of the sources worked so perfectly in the case of the documents J and P, particularly in the primeval history, that scholars were inclined to regard the Documentary Hypothesis as a general rule. As an example, I quote Rudolf Smend Sr.: “In the history of the Hexateuch this remarkable process (i.e., the joining of sources) was continually repeated.”19 But the truth is that the combination of formerly independent sources has been an absolute exception within the literary history of the Bible. It can only have taken place under very special conditions. If it is applied several times or to more than two sources, the Documentary Hypothesis becomes absurd. Some texts formerly attributed to E in fact prove to be pre-editorial sources of the Yahwistʼs history. This is true of the stories of the founding of the cult sites related in Gen 28 and Exod 3. In these cases, the Yahwistic part that can easily be identified by its use of the divine name ‫ יהוה‬does not form an independent parallel version but an editorial addition to the older literary tradition that uses the term ‫אלהים‬.20 The Abraham stories in Gen 20–22, in contrast, are fragments of a Genesis midrash comparable to the Genesis Apocryphon from Qumran. These three chapters relate to J and P as well. They were inserted into the text of what is now the book of Genesis after the two sources had already been unified for a long time.21 Other E texts are late additions mostly focusing on the piety and behavior of the patriarchs. This is true of large parts of the Joseph story, for example. They unite to form an edition with a distinct theological profile, which I propose to call the righteousness edition.22 Additions of this purpose are to be found in most narratives of the Tanak. They serve to bring fate and behavior into accord with the Jewish belief in the unwavering righteousness of God. The story of Joseph and his brothers was an especially suitable subject for this purpose, because righteousness was already a distinguishing feature of the  Rudolph and Volz, Der Elohist als Erzähler (see n. 15).  R. Smend, Die Erzählung des Hexateuchs (Berlin: Reimer, 1912), 343 (my translation). 20  For Gen 28, see L evin , Der Jahwist (see n. 3), 216–220; for Exod 3 see idem , “The Yahwist and the Redactional Link between Genesis and Exodus,” in A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (ed. T. B. Dozeman and K. Schmid; SBLSymS 34; Atlanta: Society of Biblical Literature, 2006), 131–141; repr. in idem , Re-Reading the Scriptures (see n. 3), 83–93. 21  See Levin, Der Jahwist (see n. 3), 172–180. 22  See C. Levin, “Righteousness in the Joseph Story: Joseph Resists Seduction (Genesis 39),” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 223–240; repr. in idem, Re-Reading the Scriptures (see n. 3), 65–82. 18 19

The Pentateuch

585

original narrative. It provided the perfect matrix to show how God was guiding human fate according to his purpose to perform righteousness under each and every circumstance.

The Documentary Hypothesis and the Other Hypotheses The Documentary Hypothesis is by no means sufficient for uncovering the literary history of the Pentateuch. Its weaknesses were observed soon after its development in the eighteenth century. The Documentary Hypothesis can only be maintained if it is combined with the Supplementary Hypothesis. This was emphatically underlined by Abraham Kuenen: “The redaction of the Hexateuch assumes the form of a continuous diaskeue or diorthosis, and the redactor becomes a collective body headed by the scribe who united the two works [. . .] into a single whole, but also including the whole series of his more or less independent followers.”23 Also in the Pentateuch, not differently from anywhere else in the Old Testament, the Supplementary Hypothesis is the most appropriate way to understand the process of literary growth. Of special importance is the fact that the two sources J and P were joined at an early stage in comparison with the process of textual growth of the Pentateuch as a whole and that their compass was restricted. In the primeval history, where the Documentary Hypothesis works best, I found that about seventy of 299 masoretic verses – about one quarter of the present text – had been added after the combination of the two sources.24 In the patriarchal narratives, the extent of late textual additions is much greater. Many texts are not part of one of the documents but were added later, such as Gen 12:10–20;25 14; 15;26 20–22; 23;27

 A. K uenen , An Historico-Critical Inquiry into the Origin and Composition of the Hexateuch (trans. P. H. Wicksteed; London: Macmillan, 1886), 315. 24  See C. Levin, “Die Redaktion RJP in der Urgeschichte,” in Auf dem Weg zur Endgestalt von Genesis bis II Regum: Festschrift H.-C. Schmitt (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 15–34; repr. in C. L evin , Verheißung und Rechtfertigung: Gesammelte Studien zum Alten Testament II (BZAW 431; Berlin: de Gruyter, 2013), 59–79. 25  See C. L evin , “Abraham in Ägypten (Gen 12,10–20),” in “Vom Leben umfangen”: Ägypten, das Alte Testament und das Gespräch der Religionen; Gedenkschrift M. Görg (ed. S. Wimmer and G. Gafus; ÄAT 80; Münster: Ugarit-Verlag, 2014), 109–121. 26  See C. L evin , “Jahwe und Abraham im Dialog: Genesis 15,” in Gott und Mensch im Dialog: Festschrift für O. Kaiser (ed. M. Witte; BZAW 345/I; Berlin: de Gruyter, 2004), 237–257; repr. in idem, Verheißung und Rechtfertigung (see n. 24), 80–102. 27  See C. L evin , “Abraham erwirbt seine Grablege (Genesis 23),” in “Gerechtigkeit und Recht zu üben” (Gen 18,19): Festschrift für E. Otto (ed. R. Achenbach and M. Arneth; BZABR 13; Wiesbaden: Harrassowitz, 2010), 96–113; repr. in idem , Verheißung und Rechtfertigung (see n. 24), 103–123. 23

586

Christoph Levin

34;28 38;29 48; 49; and most of the promises to the patriarchs,30 some of the genealogies, and many additions and comments in nearly every part of the present text. In the books of Exodus through Numbers, the late texts are so predominant that, from at least Exod 19 onward, it becomes difficult to identify the narrative strands of the two sources at all – which, nevertheless, do exist. One reason for present doubts about the Documentary Hypothesis is that the size of the two sources J and P has been greatly overestimated in previous scholarship. Moreover, the formerly separate documents did not remain unscathed when copied into the framework of the combined J/P document. There are a number of gaps. The Priestly code was preserved in the primeval history, but in the patriarchal narratives most of its text is lost. In the text of the Yahwistic history, there are also a number of gaps, even if they are smaller. As a result, none of the two documents is suited to serve as the one and only basic document (Grundschrift) of the whole composition. The Pentateuch does not hang on a single thread but on a cord plaited together from two strands. This cord makes it possible for the work as a whole to avoid falling apart when one of the two threads is torn or missing. Seen in this way, the fact that the former documents cannot be fully reconstructed does not shake the Documentary Hypothesis. On the contrary, it makes it even more probable.

Toward the Present Shape of the Pentateuch Even in the case of the Torah, it would seem reasonable that a work that evolved from one single redaction would, as a literary unit, have constituted one single scroll. “The fact that the biblical books, and even the smallest of them, were kept from the outset on separate scrolls is [. . .] a conclusive proof of the basic rule, that each complete work was to be written on its own scroll.”31 The extent of todayʼs text of the Pentateuch, however, exceeds by far the compass of any scroll known to us. The Isaiah scroll from Qumran could accommodate no 28  See C. Levin, “Dina: Wenn die Schrift wider sich selbst lautet,” in Schriftauslegung in der Schrift: Festschrift für O. H. Steck (ed. R. G. Kratz et al.; BZAW 300; Berlin: de Gruyter, 2000), 61–72; repr. in C. Levin, Fortschreibungen: Gesammelte Studien zum Alten Testament (BZAW 316; Berlin: de Gruyter, 2003), 49–59. 29  See C. Levin, “Tamar erhält ihr Recht (Genesis 38),” in Diasynchron: W. Dietrich (ed. T. Naumann and R. Hunziker-Rodewald; Stuttgart: Kohlhammer, 2009), 279–298; repr. in idem , Verheißung und Rechtfertigung (see n. 24), 124–143. 30  See C. Levin, “Die Väterverheißungen: Eine Bestandsaufnahme,” in The Post-Priestly Pentateuch: New Perspectives on its Redactional Development and Theological Profiles (ed. F. Giuntoli and K. Schmid; FAT 101; Tübingen: Mohr Siebeck, 2015), 125–143. 31  M. Haran, “Book-Size and the Thematic Cycles in the Pentateuch,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für R. Rendtorff (ed. E. Blum et al.; Neukirchen-Vluyn: Neukirchener Verlag, 1990), 165–176, esp. 166.

The Pentateuch

587

more than a quarter of todayʼs Torah. From this we must conclude that the first redactional form of the combined J/P document was considerably shorter than the present text. This again is a strong argument in favor of the Supplementary Hypothesis. Without gradual literary growth, the compass of the Pentateuch as we have it would be inconceivable. This pertains to the shaping of the books of the Pentateuch as well. It is highly unlikely that the overall historiographical concept came into being only through the subsequent linking of books to one another, which were for the most part already independent. The individual books did not precede the literary growth; they are its outcome. Scrolls that had become too large were split up. Thus the separation of the Pentateuch into single books is secondary, with the probable exception of the division between Numbers and Deuteronomy. The “cross-referencing at the seams of the books reflects the division of the material into books. Therefore, it is clear that the book subdivisions [. . .] must be older than the last redactional shaping of the text.”32 Again, in relation to the literary process as a whole, the division of the Pentateuch into five books had already been made early on and was presumably not made for all books at the same time. The separated books presented the matrix for further growth. This provided the opportunity to add new sheets and to enlarge the separated single scrolls. In this way, the individual books came to have their individual shapes, at least to a certain degree.33 This is most obvious in the case of the book of Genesis. The literary growth eventually came to an end. From the Hellenistic period onward, the text gradually crystallized into fixed form. It is evident from the Septuagint that the Pentateuch took the shape it has today at the time of its translation into Greek. According to the Letter of Aristeas, this took place in the middle of the third century BCE. From then on, the process of interpreting and actualizing the religious tradition moved from the canonical books to writings outside of the collection that eventually became the Hebrew Bible. Regarding the shape of the Pentateuch that has come down to us, the statement of Erhard Blum proves to be true: “There is no final redaction.”34

 K. S chmid , Genesis and the Moses Story: Israelʼs Dual Origins in the Hebrew Bible (trans. J. D. Nogalski; Siphrut 3; Winona Lake, IN: Eisenbrauns, 2010), 27. 33  See C. L evin , “On the Cohesion and Separation of Books within the Enneateuch,” in Pentateuch, Hexateuch, or Enneateuch: Identifying Literary Works in Genesis through Kings (ed. T. B. Dozeman et al.; Ancient Israel and Its Literature 8; Atlanta: Society of Biblical Literature, 2011), 127–154; repr. in idem, Re-Reading the Scriptures (see n. 3), 115–141. 34  E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 380 (my translation). 32

Post-Priestly Additions in the Pentateuch A Survey of Scholarship Konrad Schmid One of the coincidences of the history of scholarship of the Pentateuch is that the main epochs since Johann Gottfried Eichhorn fall into periods of about one hundred years. Eichhornʼs Einleitung in das Alte Testament offered the first version of a two-source theory,1 approximately a century later Julius Wellhausen formulated his version of the Documentary Hypothesis,2 and then again about a century later in quick succession the works of John Van Seters,3 Hans Heinrich Schmid,4 and Rolf Rendtorff appeared.5 Some have interpreted and continued to interpret these latter works as markers of a “crisis” in pentateuchal studies, but they actually indicate the collapse of an obsolete theoretical model for the Pentateuch, at least in the view of a rather significant strand of Hebrew Bible scholarship.6 The advances of the most recent period of pentateuchal scholarThis is an updated and reworked English translation of K. Schmid, “Von der Dia­skeuase zur nachendredaktionellen Fortschreibung: Die Geschichte der Erforschung der nachpriesterschriftlichen Redaktionsgeschichte des Pentateuch,” in The Post-Priestly Pentateuch: New Perspectives on Its Redactional Development and Theological Profiles (ed. F. Giuntoli and K. Schmid; FAT 101; Tübingen: Mohr Siebeck, 2015), 1–18. My thanks to Peter Altmann for taking care of the translation. 1  J. G. Eichhorn, Einleitung in das Alte Testament (Leipzig: Weidmann, 1783). 2  J. Wellhausen, Die Composition des Hexateuch und der erzählenden Bücher des Alten Testaments (3rd ed.; Berlin: Reimer, 1899). 3  J. Van S eters , Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975). 4  H. H. Schmid, Der sogenannte Jahwist (Zürich: Theologischer Verlag, 1976). 5  R. Rendtorff, Das überlieferungsgeschichtliche Problem des Pentateuch (BZAW 147; Berlin: de Gruyter, 1977). 6  See the still-groundbreaking work of E. Blum, Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 1: “What is often called a ‘crisis’ should instead be greeted as a newly discovered openness, as a chance to question in a self-critical manner and without pressing too quickly toward ‘conclusions’ in the form of dialogue-stopping hypotheses, and, moreover, to call into question the method and perspectives that are taken for granted.” Cf., e.g., J. S. Baden, J, E, and the Composition of the Pentateuch (FAT 72; Tübingen: Mohr Siebeck, 2009), 10: “This study attempts some much-needed steps to pull biblical scholarship back from the precipice of the perceived ‘crisis’ in pentateuchal criticism.”

590

Konrad Schmid

ship, beginning with this latest break, have at the same time made it possible to accord increased attention to the sections of text within the Pentateuch that are evidently later than the Priestly document. It is not novel to reckon with material in the Pentateuch that, on the one hand, is not source material, and, on the other, concerns the expansion of the Pentateuch in its final literary stage. Nevertheless, the dominant strand of scholarship since Wellhausen has accepted that the literary history of the Pentateuch essentially ended with the combination of P and JE. Furthermore, it has accepted that this combination basically provided no impetus for the production of further expansions within the Pentateuch.7 Wellhausen, for example, held with regard to the process of the combination of JE and P: The activity of the redactor consists primarily in the skilled interweaving of the sources, in which he left their content as unabridged and the wording and order of the narrative as unchanged as possible. However, he was not always able to proceed without intrusions of his own. Sometimes he made additions, at times to overcome a contradiction or to cover over a seam.8 7  As significant exceptions – each within the framework of a completely different theoretical approach to the Pentateuch – mention should be made here of the works of B. Jacob (Das erste Buch der Tora: Genesis [New York: Schocken, 1934]) and R. N. Whybray (Introduction to the Pentateuch [Grand Rapids, MI: Eerdmans, 1989]). Both understand the Pentateuch as the work of one author who at the same time worked as the editor of all the traditional material, which he worked into the Pentateuch. 8  W ellhausen , Composition (see n. 2), 2; all translations are mine unless otherwise noted. Wellhausen reproduces here a perspective that has basically been adopted since Astruc (J. A struc , Conjectures sur les mémoires originaux dont il paroit que Moyse sʼest servi pour composer le Livre de la Genèse [Brussels: Fricx, 1753]). According to Astruc, Genesis consists of two sources that Moses as the redactor reworked, somewhat like a “harmonie des Evangelistes” (A struc , Conjectures, 525), in order to bring them together (see M. W itte , Die biblische Urgeschichte: Redaktions- und theologiegeschichtliche Beobachtungen zu Gen 1,1–11,26 [BZAW 265; Berlin: de Gruyter, 1998], 2–3; J. C. G ertz, “Jean Astruc and Source Criticism in the Book of Genesis,” in Sacred Conjectures: The Context and Legacy of Robert Lowth and Jean Astruc [ed. J. Jarick; LHBOTS 457; New York: Continuum, 2007], 190–203). Quite similarly, J. G. E ichhorn described a redactor who had a “holy respect” for the sources he set out to combine “without first filing down, changing, or tinkering with their expression” (Einleitung in das Alte Testament: Vierte Original-Ausgabe, I–V [Göttingen: Rosenbusch, 1823–1824], 99). A concise portrayal of the view of the final redactor of the Pentateuch in the eyes of nineteenth century scholars appears in W itte , Urgeschichte, 1–16. It is remarkable that it was also possible for pentateuchal scholarship before Wellhausen to combine the final redaction of the Pentateuch with the insertion of broad sections of text. For example, F. B leek (Einleitung in das Alte Testament: Vierte Auflage nach der von A. Kamphausen besorgten dritten bearbeitet von J. Wellhausen [ed. J. Bleek und A. Kamphausen; Berlin: Reimer, 1878], 124–125), identified the Deuteronomist as the final redactor of the Pentateuch, who was also responsible for Lev 26: “Solely our Deuteronomy was undoubtedly written from the beginning as an addition and expansion of the older historical work, which was the same one that the Jehowistic redactor of the first four books of our Pentateuch received; and quite probably the author of Deuteronomy was also the final redactor of the entire Pentateuch, through whom the work

Post-Priestly Additions in the Pentateuch

591

Nonetheless, Wellhausen did not see the composition of the Pentateuch as completely finished with the process that combined JE and P. Following Popper and Kuenen, Wellhausen reckoned with a diaskeuast,9 who also edited the Pentateuch further after this combination: “The diaskeuast, with whom Kuenen correctly reckoned, already had the combined composition JE+Q in front of him and subjected it to further editing.”10 On a related matter, it is important to note that Wellhausen emphasized on multiple occasions, in particular in Composition, that “the literary process was in reality more complicated, and the so-called Supplementary Hypothesis is still applicable in a subordinate role.”11 With regard to the Priestly document, Wellhausen speaks of “secondary and tertiary levels,”12 though it appears that he hardly considered the alternative that such “secondary and tertiary” elements could simply be additions to P itself rather than to the Priestly document combined with JE. The term diaskeuast, employed by Wellhausen, did not, however, originate with Kuenen.13 In its application to the biblical texts,14 and in the less common

obtained the extent and configuration in which it now appears to us. Because this was then combined with a new copy of the preexisting work, it can be considered that individual [texts] in the previous books were also changed or interpolated by his hand. But in my view this is only probable with regard to the earlier [. . .] discussed section Lev 26:2–45, whose entire tone and character are unmistakably very similar to the speech of Deuteronomy.” Wellhausen allowed these explanations to remain in the fourth edition of the Einleitung, which he edited, but he added, in §§ 81–87, a further section entitled “Progress of Hexateuchal Criticism since Bleekʼs Death.” 9  R. G. K ratz , “Wellhausen,” TRE 35:527–536, here 529, explains diaskeuast as “Fort­ schreibung [expansion], in the manner understood today following Walther Zimmerli.” 10  Wellhausen, Composition (see n. 2), 329. 11  W ellhausen , Composition (see n. 2), 207. On this, see S. G esundheit , Three Times a Year: Studies in Festival Legislation in the Pentateuch (FAT 82; Tübingen: Mohr Siebeck, 2012), 233, n. 16, with references to similar declarations on pp. 135, 137, 178–179, 192. 12  Wellhausen, Composition (see n. 2), 178, n. 1. 13  On the relationship between Wellhausen and Kuenen, see R. S mend , “Kuenen und Wellhausen,” in Abraham Kuenen (1828–1891): His Major Contributions to the Study of the Old Testament (ed. P. B. Dirksen and A. van der Kooij; Leiden: Brill, 1993), 113–127, here 125; idem , “The Work of Abraham Kuenen and Julius Wellhausen,” in From Modernism to PostModernism (The Nineteenth and Twentieth Centuries), vol. 3 of Hebrew Bible/Old Testament: The History of Its Interpretation (ed. M. Sæbø; Göttingen: Vandenhoeck & Ruprecht, 2013), 424–453. 14  It appears prominently in F. S chlegel (e.g., Geschichte der Poesie der Griechen und Römer [Berlin: Unger, 1798], 170) and plays an important role in classical philology; see A. G räfenhan , Geschichte der Klassischen Philologie im Alterthum (Bonn: 1843), 264: “With regard to Homeric poems, these are interpolated by the diaskeuasts not only in individual verses but also through insertion of larger sections, which the old grammarians have in part uncovered with great acumen.”

592

Konrad Schmid

form diaskeue, it goes back to Julius Popper,15 who appears in Wellhausenʼs Composition as “the learned rabbi.”16 In 1862, Popper published a work on Exod 25–40,17 in which he argued that “the second part [namely the report of the Tent of Meeting, i.e., Exod 35–40] is a later sprouting trunk of our description [namely, Exod 25–29].”18 Popper compares Exod 25–31 with 35–40 and portrays the “amplifications” in Exod 35–40 as interpretations: “it is almost the traditional material, but the content itself that appears here in its fuller and more correct form has been interpreted and developed more richly and precisely.” 19 In contrast, he places great value on the text-critical differences between in the Samaritan Pentateuch and the Septuagint and combines these phenomena with the latest expansions in the Pentateuch. As a result, he interprets the “additions and interpolations in the Samaritan Pentateuch as the extensions and degener15  Not, however, exclusively; see J. Van Seters, The Edited Bible: The Curious History of the “Editor” in Biblical Criticism (Winona Lake, IN: Eisenbrauns, 2006), 237, n. 173. Wellhausen was also familiar with the term from O. Thenius, Die Bücher Samuels erklärt (2nd ed.; Leipzig: S. Hirzil, 1864) (which refers to the masoretische Diaskeuase). On Popper, see R. HaCohen, Reclaiming the Hebrew Bible: German-Jewish Reception of Biblical Criticism (Berlin: de Gruyter, 2010), 137–141. 16  W ellhausen , Composition (see n. 2), 146. Wellhausen offers a longer discussion of Popper in his revision of Bleekʼs Einleitung (B leek , Einleitung in das Alte Testament [see n. 8], 155–156): “His investigation comes to the conclusion that the descriptions of the building of the tabernacle (Exod 35–49) and the dedication of the priests (Lev 8–10) are later than the prescriptions concerning the two matters in Exod 25–31 – actually they only reach the form in which we now have them considerably later than the Babylonian captivity. It is written in a quite circular fashion; furthermore, the question itself is complicated because the author does not build his hypothesis only on the Masoretic, but also on the Samaritan, text and especially the Greek translation. [. . .] The earliest layer is not – as was generally accepted before him – made of one piece but was the product of a long, continuous diaskeue, as Popper calls it, in which the scribes following Ezra were the last to lay a hand upon it. [. . .] It is very regrettable that the German scholars investigating Composition after the time of Popper have gained no insight from his book or at least excused themselves from the study of it: only Geiger (Jüd. Zeitschr. für Wiss. und Leben I S. 122ff.) and Graf (die Geschichtsbücher des A. T. S. 86–87) constitute exceptions. This is totally unjust. I must be very mistaken in my estimation of Popperʼs book if it does not go on to exercise considerable influence.” 17  J. P opper , Der biblische Bericht über die Stiftshütte: Ein Beitrag zur Geschichte der Composition und Diaskeue des Pentateuch (Leipzig: Hunger, 1862). 18  Popper, Der biblische Bericht (see n. 17). On Popper, see also H. Utzschneider, Das Heiligtum und das Gesetz: Studien zur Bedeutung der sinaitischen Heiligtumstexte (Ex 25–40; Lev 8–9) (OBO 77; Fribourg: Universitätsverlag, 1988), 30; Gesundheit, Three Times a Year (see n. 11), 232–234; T. Römer, “ ‘Higher Criticism’: The Historical and Literary-Critical Approach – with Special Reference to the Pentateuch,” in From Modernism to Post-Modernism (The Nineteenth and Twentieth Centuries), vol. 3 of Hebrew Bible/Old Testament: The History of Its Interpretation (ed. M. Sæbø; Göttingen: Vandenhoeck & Ruprecht, 2013), 393–423, here 421–422. 19  Popper, Der biblische Bericht (see n. 17), 147; see also Utzschneider, Heiligtum (see n. 18), 31.

Post-Priestly Additions in the Pentateuch

593

ations from the pentateuchal diaskeuast.”20 In any case, Popperʼs book marks the first comprehensive evaluation of the post-Priestly interpretative work in the Pentateuch. Popper still views the Priestly document as the oldest source of the Pentateuch, however, even though he sees the process of its extensions continuing to the year 260 BCE.21 In fact, Popper also basically views the redaction-historical developments of the Pentateuch in terms of a model of decadence, though he recognizes the importance of these late “amplifications”: The period of actual creative composition thus lay far in the past [in relation to] all these reshapings and changes, so we have here also nothing more than stylistic and copyediting activities, with the mere formal finishing of the received material concerning its diaskeuastic revisions and arrangements. Yet in this period the Pentateuch receives its formal shape, which should provide scholarship with a much larger field of activity than one usually is inclined to accept.22

Not only Wellhausen but also Kuenen read Popper. Kuenen adopted the term diaskeue from Popper: The redaction of the Hexateuch, then, assumes the form of a continuous diaskeue or diorthosis, and the redactor becomes a collective body headed by the scribe who united the two works [. . .] into a single whole, but also including the whole series of his more or less independent followers.23

A consensus was then established that on the one hand emphasized the final compilerʼs passivity and lack of originality and on the other granted the presence of still-later expansions or a post–final-redaction incorporation of specific individual source texts.24 This position appears, for example, in Gunkelʼs commentary on Genesis. He holds to the notion of a final redactor, whom he views more as a collector, like the authors of the source documents: Pʼs attempt to suppress the old failed; and a pious hand brought about a combination of JE and P. This final collection took place with extraordinary fidelity, especially to  Popper, Der biblische Bericht (see n. 17), 67.  Popper, Der biblische Bericht (see n. 17), 7. 22  Popper, Der biblische Bericht (see n. 17), 73. 23  A. K uenen , An Historico-Critical Inquiry into the Origin and Composition of the Pentateuch (trans. Ph. H. Wicksteed; London: Macmillan, 1886), 315; see also idem , De Godsdienst van Israel II (Harlem: Kruseman, 1870), 265–266. 24  The acceptance of the sporadic appearance of post-Priestly portions of text in the Pentateuch can also be found in some contributions from more recent, though also more traditionally oriented, pentateuchal scholars in the German-speaking sphere, e.g., H. Seebass, “Gehörten Verheißungen zum ältesten Bestand der Väter-Erzählungen?,” Bib 64 (1983), 189–210, here 207, n. 48; W. H. Schmidt, Exodus (BK 2/1; Neukirchen-Vluyn: Neukirchener Verlag, 1988), 318; P. W eimar , Untersuchungen zur Redaktionsgeschichte des Pentateuch (BZAW 146; Berlin: de Gruyter, 1977), 142; idem , Die Berufung des Mose: Literaturwissenschaftliche Ana­ly­se von Exodus 2,23–5,5 (OBO 32; Fribourg: Universitätsverlag, 1980), 16, 287, 321; S. Mittmann, Deuteronomium 1:1–6:3 literarkritisch und traditionsgeschichtlich untersucht (BZAW 139; Berlin: de Gruyter, 1975), 56. 20 21

594

Konrad Schmid

P; its author attempted wherever possible to prevent any granule of P from falling to the ground.25

What did this final collector himself formulate? Gunkel draws a modest conclusion: We are able to infer very little in Genesis that with more or less certainty came from his hand: a few harmonizing glosses or fills like 10:24; 15:7, 8, 15; 27:46; in 35:13, 14; further retouching in 6:7; 7:7, 22, 23; and further 7:3a, 8, 9; then the distinction between Abram and Abraham, and Sarai and Sarah also in J and E, among others.26

After the combination of JE and P, Genesis, or the Pentateuch, respectively, had essentially received its shape. However, several later additions can be identified. With this the activity of the redactors in Genesis is basically finished. But in individual [details] the work [diaskeuase] on the text continued much longer. We see small revisions in chapter 34 and in the numbers in the genealogies, in which the Jewish and Samaritan versions and the text of the Greek translation deviate from one another. Larger expansions and revisions still took place in 36 and 46:8–27; the final insertion is the narrative of Abrahamʼs victory over the four kings, a “midrash”-like legend from the latest period.27

In his History of Pentateuchal Traditions (1948), Martin Noth emphasized even more strongly that the combination of JE and P meant the completion of the Pentateuch: The formation of the entire Pentateuch, that is, the insertion of the J narrative that had been expanded by numerous E elements into the literary frame of P, is no longer of great relevance in terms of tradition history. It concerns a purely literary process that brought in neither new points of view in the editing of material nor interpretations of the material. It instead only amounts to a process of addition and is only important insofar as its result was the completed Pentateuch, as it comes to us as an extant entity from which we can extract the earlier stages through more or less certain analysis.28

This naturally does not categorically exclude that individual texts of the Pentateuch could first have been added after the combination of P and JE, but this possibility is limited to a few pieces that have no relation to one another; they are of neither great compositional nor theological importance. This conviction appears in various standard introductions to the Old Testament from the post–World War II period, which maintain that the Pentateuch essentially consists of the composition of source documents, though it also contains individual pieces that were first added after this decisive compositional step. For example, Otto Eissfeldt identified Gen 14 and Gen 18:22b–33, as well as certain  H. Gunkel, Genesis (HKAT 1/1; Göttingen: Vandenhoeck & Ruprecht, 1901), xcix.  Gunkel, Genesis (see n. 25), xcix. 27  Gunkel, Genesis (see n. 25), xcix. 28  M. N oth , Überlieferungsgeschichte des Pentateuch (Stuttgart: Kohlhammer, 1948), 267–268. 25 26

Post-Priestly Additions in the Pentateuch

595

legal texts, as post-Priestly additions.29 Georg Fohrer similarly discussed the possibility of post-Priestly origins for individual texts outside the source layers such as Gen 14; Exod 15:1–19; Exod 19:3b–8; and Deut 32 – especially Exod 15:1–19 and Deut 32. He sees Gen 14 as an absorbed source text, and he dates Exod 19:3b–8 to the late monarchic period.30 It was only in the late 1970s – especially because of the effects of the “crisis”31 in pentateuchal scholarship perceived in the immediately preceding years – that the introductory literature to the Old Testament began to include programmatic statements on the necessity of placing more value on the work of the combination of the traditional pentateuchal sources as a literary-historical procedure with an importance all its own.32 For example, Rudolf Smend writes: The theological and intellectual-historical status of this combination [of the source documents] forbids evaluating it in passing and dedicating interest more or less solely to the works extracted through the separation of sources that the redactor had before him. It would be valid to apply and extend the results of the “Documentary Hypothesis” much more strongly in the reverse sense.33

Smend specifically points out that during the combination of P and JE, one should also expect the insertion of new portions of text: The school that we call R did not limit itself to the joining of the received source documents. They instead subjected the new whole to a further redaction. [. . .] This is one of the most important domains for the Supplementary Hypothesis.34

In this judgment, Smend explicitly follows Kuenen, who conceived of the redactor as a “collective whose head is the one that combined the two [. . .] sources into a whole, to whom, however, in addition a number of more or less independent successors belonged.”35 Smend explicitly pointed out the long-term 29  O. E issfeldt , Einleitung in das Alte Testament unter Einschluß der Apokryphen und Pseud­epigraphen sowie der apokryphen und pseudepigraphenartigen Qumran-Schriften (3rd ed.; NThG; Tübingen: Mohr Siebeck, 1964), 280. 30  G. F ohrer , Einleitung in das Alte Testament (Heidelberg: Quelle & Meyer, 1965), 202–206. 31  See above, n. 6. 32  The dictum from H.-W. Hertzberg is frequently cited: “still missing is the book about the author of it all, the final redactor” (“Ist Exegese theologisch möglich?,” in Beiträge zur Traditionsgeschichte und Theologie des Alten Testaments [ed. H.-W. Hertzberg; Göttingen: Vandenhoeck & Ruprecht, 1962], 101–117, here 111; see J. C. Gertz, Tradition und Redaktion in der Exoduserzählung [FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000], 13). 33  R. S mend , Die Entstehung des Alten Testaments (ThW 1; Stuttgart: Kohlhammer, 1978), 38. 34  Smend, Die Entstehung (see n. 33), 46. 35  A. K uenen , Historisch-kritische Einleitung in die Bücher des alten Testaments hin­ sichtlich ihrer Entstehung und Sammlung, Bd. I/1 (Leipzig: J. C. Mathes, 1887), 302; cf. Gertz, Tradition (see n. 29), 16.

596

Konrad Schmid

development of the legal texts in the Torah, which should also be evaluated from this perspective. In addition to the works of Van Seters, Schmid, and Rendtorff, there was also the often-forgotten contribution by Eberhard Ruprecht from 1974, which is important for the history of scholarship. Before these other authors, Ruprecht developed an exemplary, complex analysis of the post-Priestly growth of the original Priestly version of Exod 16: The oldest literary level is the Priestly narrative in verses 1–3, 6–7, 9–27, 30, 35a. The next layer is a Deuteronomistic redaction in 15:25b–26; 16:4–5, 28–29, 31–32. It already has the combination of JE and P before it. The latest layer is an addition in Priestly style in verses 33–34. [. . .] This is quite a striking conclusion because such a stratification has not yet been observed in pentateuchal scholarship.36

Though Wellhausen had already suspected post-Priestly exegesis in the context of Exod 16:4–5, he went on to conclude that instead an older JE fragment had been worked into this location: “Could it then be that the diaskeuast originally divined this from the report of Q and placed it in verses 4–5? Such critical insight cannot really be expected from him.”37 This argument is not actually based on the evidence; it arises from a particular prior understanding of what a diaskeuast can or cannot do.38 Whereas Ruprecht presented a limited textual example of the phenomenon of post-Priestly redactional activity, D. J. A. Clinesʼs volume The Theme of the Pentateuch develops a completely different approach.39 His book is noteworthy in that it is the first time that the specific redactional outlook of the present form of the Pentateuch is investigated. Nonetheless, though Clinesʼs monograph was synchronically rather than redaction-historically oriented,40 he still explicitly formulated the investigation as a search for the “final redactor.”41 Yet the classical conception of a compiler that only intervenes minimally is quite operative in Clinesʼs work: “First, the final redactor can, with minimal interference, re-shape the total impact of his material [. . .]. Secondly, it is not necessary to posit that the shape of the final work was intended by the redactor.”42 36  E. Ruprecht, “Stellung und Bedeutung der Erzählung vom Mannawunder (Ex 16) im Aufbau der Priesterschrift,” ZAW 86 (1974), 269–307, here 302. 37  Wellhausen, Composition (see n. 2), 329. 38  R. Smend, Die Erzählung des Hexateuch auf ihre Quellen untersucht (Berlin: Reimer, 1912), 149, also recognizes the work of a diaskeuase in Exod 16. 39  D. J. A. C lines , The Theme of the Pentateuch (2nd ed.; JSOTSup; Sheffield: Sheffield Academic Press, 1997). 40  It is noteworthy that the search for the redactional outlook of the present Pentateuch is also treated as a synchronic question in more recent scholarship; see, e.g., E. W. Nicholson, The Pentateuch in the Twentieth Century: The Legacy of Wellhausen (Oxford: Oxford University Press, 1997), 249–268. 41  Clines, Themes (see n. 38), 102. 42  Clines, Themes (see n. 38), 25.

Post-Priestly Additions in the Pentateuch

597

Clines identifies the “theme of the Pentateuch” as follows: The theme of the Pentateuch is the partial fulfilment – which implies also the partial non-fulfilment – of the promise to or blessings of the patriarchs. The promise of blessing is both the divine initiative of the world where human initiatives always lead to disaster, and are an affirmation of the primal divine intentions for humanity.43

With regard to the historical location of the redactor, Clines remains quite vague. He locates the composition of the Pentateuch in the Babylonian exile.44 At the same time as Clinesʼs volume, the first publication in the German-speaking sphere actually dedicated to the discussion of the redactor as such appeared, namely H. Donnerʼs study.45 In contrast to Clines, Donner interpreted the “redactor” as neither an actively nor a passively creative spirit but rather a compiler whose task consisted of “combining the literarily existing texts with one another in such a way that a new text emerged from multiple texts with the preservation of everything essential and with as few omissions as possible.”46 The redactors were “not inventors of great theological designs, but precritical scribal exegetes.” Donnerʼs study was significant in that it was completely focused on the phenomenon and the inner logic of redaction.47 It was limited, however, by a preconception stated at the outset of the essay of the limited nature of what a redactor could be conceived as doing, namely, compiling existing textual materials as passively as possible. Donnerʼs understanding is supported by his choices about which texts to investigate. He begins with a discussion of Exod 14 rather than texts that evince productive redactional interpretations of preexisting material. Erhard Blum, in his 1990 Studien zur Komposition des Pentateuch, was the first in the history of scholarship to accord substantial attention to post-Priestly textual growth in the Pentateuch.48 This is true even though he separates this  Clines, Themes (see n. 38), 30.  Clines, Themes (see n. 38), 103: “As is generally assumed, the redaction of the Pentateuch took place in Babylonia.” See also – and more emphatically – 104: “The Pentateuch, even if it was composed after 539 BCE, is still an exilic work.” 45  H. Donner, “Der Redaktor: Überlegungen zum vorkritischen Umgang mit der Heiligen Schrift (1978),” in Aufsätze zum Alten Testament aus vier Jahrzehnten (ed. H. Donner; BZAW 224; Berlin: de Gruyter, 1994), 259–285. 46  Donner, “Der Redaktor” (see n. 45), 262–263; see the summary on 283–285. 47  Donner, “Der Redaktor” (see n. 45), 282. 48  See n. 6. At the same time, J. Hughes, Secrets of Times: Myth and History in Biblical Chronology (JSOTSup 66; Sheffield: Sheffield Academic Press, 1990), 93, points out a “postpriestly revision” of the royal chronology. On the suggestion of a Hasmonaean redaction in Gen 5 – the chapter that provides the basis for calculating universal time (see K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels in den Geschichtsbüchern des Alten Testaments [WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999], 19–22) – see R. Hendel, “A Hasmonean Edition of MT Genesis? The Implications of the Editions of the Chronology in Genesis 5,” HBAI 1 (2012), 448–464. 43 44

598

Konrad Schmid

stage from the two earlier primary steps in terms of classification. The first main section of his book is introduced as “Part One: The Pre-Priestly Composition (KD).”49 This is followed by “Part Two: The Priestly Composition (KP),”50 and the final section receives the heading “Outlook: On the Way to the Canonical ‘Final Form.’ ”51 Even though Blum generally connects the essential aspects of the formation of the Pentateuch with Pʼs incorporation into the pre-Priestly texts, he still emphasizes the programmatic importance of the post-Priestly development: Even if we might see the shaping of the principal Priestly composition as the decisive formative step on the way to the canonical Torah, this is by no means to be equated with its “final form.” The shaping of tradition continues considerably. For one – there can hardly be controversy about this – within the Priestly circles “its” material was perpetuated, actualized, etc., for a longer period of time. Furthermore, however, there were also more or less punctiliar revisions/additions from the circle that was in the broadest sense made up of “Deuteronomistic” tradents.52

Blum refers initially to the work of E. Ruprecht on Exod 1653 but also to that of N. Lohfink on Exod 15:25b, 26.54 Blum himself identifies further post-Priestly expansions in the “Aaron components” of Exod 4 and the related texts in Exod 18;55 in the texts of what he calls the “Josh 24 redaction,”56 which begins in Gen 35:1–7 and is carried on by the series of declarations in Gen 50:25b, 26; Exod 13:19; and Josh 24:32;57 as well as in the “Malʾak redaction,”58 with which one “can reckon with Exod 14:19a; 23:20ff.*; (32:34aβ) 33:2, 3b*, 4; 34:11–27, and Judg 2:1–5.”59

 Blum, Komposition des Pentateuchs (see n. 6), 7.  Blum, Komposition des Pentateuchs (see n. 6), 219. 51  Blum, Komposition des Pentateuchs (see n. 6), 361. 52  Blum, Komposition des Pentateuchs (see n. 6), 361. 53  Blum, Komposition des Pentateuchs (see n. 6), 361. See also Ruprecht, “Stellung und Bedeutung” (see n. 36). 54  N. Lohfink, “ ‘Ich bin Jahwe, dein Arzt’ (Ex 15,26): Gott, Gesellschaft und menschliche Gesundheit in einer nachexilischen Pentateuchbearbeitung (Ex 15,25b.26),” in “Ich will euer Gott werden”: Beispiele biblischen Redens von Gott (ed. N. Lohfink; SBS 100; Stuttgart: Katholisches Bibelwerk, 1981), 11–73. 55  Blum, Komposition des Pentateuchs (see n. 6), 362. 56  B lum , Komposition des Pentateuchs (see n. 6), 363. See also idem , “Die literarische Verbindung von Erzvätern und Exodus: Ein Gespräch mit neueren Endredaktionshypothesen,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 119–156, here 125, as well as M. Witte, “Die Gebeine Josefs,” in Auf dem Weg zur Endgestalt von Genesis bis II Regum (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 139–156. 57  Blum, Komposition des Pentateuchs (see n. 6), 363. 58  Blum, Komposition des Pentateuchs (see n. 6), 377. 59  Blum, Komposition des Pentateuchs (see n. 6), 363. 49 50

Post-Priestly Additions in the Pentateuch

599

In his 1984 Komposition der Vätergeschichte, Blum was still quite hesitant to identify post-Priestly portions of text, except for texts like Gen 1460 and the series connected with Josh 2461 in which Blum deduced a clear relationship to KP. Blum only later designated the D texts in Genesis, such as Gen 24 (which he assigned to D in 1984), as post-Priestly.62 Christoph Levinʼs work accentuates the post-Priestly redactional activities even more clearly than Blum did, calling them “post–end-redactional.” Levin continues to view the most important redactional step in the process of the composition of the Pentateuch as the – easily misunderstood63 – “end redaction,” which combined J and P. This process points to the shared form and equal value of J and P.64 Nevertheless, it is still the case that [t]he combination of the Yahwist and the Priestly documents does not even come close to bringing the history of the Pentateuch to a close. One of the surprises that has resulted from our investigation is the considerable extent of the text that was added after the end redaction. The established designation “end redaction” cannot be taken literally. This is not solely the case for the legal portions of the books of Exodus through Numbers. The movements concerned with theodicy, those that are moralizing, and the wisdom tendencies in the narratives of Genesis were largely added after the combination of the Yahwist and the Priestly documents. Those layers that the Documentary Hypothesis has to this point attributed to the “Elohist” are also included. The non-Priestly promises to the ancestors, to the extent that they do not come from a Yahwistic redactor, were also first inserted subsequent to the end redaction. Also post–end-redactional are the system of the twelve tribes, the apologetic in the cycle of the Egyptian plagues, and a large part of the wilderness traditions. The promise of the land and the conquest of the land also receive their present meaning quite late. The current Pentateuch to a great degree first came about through post–end-redactional additions.65  E. Blum, Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukir­ chener Verlag, 1984), 462–464, n. 5. 61  Blum, Komposition der Vätergeschichte (see n. 60), 44–61. 62  Blum, “Verbindung” (see n. 56), 140–144. 63  In light of the presence of “post–end-redactional” texts in the Pentateuch, the use of the designation “end redaction” for the process of the combination of J and P is a contradiction in terms. See also B lum , Komposition des Pentateuch (see n. 6), 380: “There is no ‘end redaction.’ ” Similarly, C. F revel , in Einleitung in das Alte Testament (ed. E. Zenger and C. Frevel; 8th ed.; Stuttgart: Kohlhammer, 2012), 148. Gertz explains the terminological use as follows: “To the degree to which the merging of the two text complexes [i.e., the Priestly and the non-Priestly] is reckoned with as the process that definitively marked the formation of the present canonical text, it should be called the end redaction. The term end redaction would, then, not be understood as the process responsible for the text-critically mediated ‘final text,’ and the use of the term does not contain a presupposition about whether it is to be understood as addressing the combination of independent narrative works or one of the two entities as an end-redactional editorial layer” (Tradition [see n. 32], 10). 64  C. Levin, Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993), 437. 65  Levin, Der Jahwist (see n. 64), 440–441. See also idem, Das Alte Testament (Munich: Beck, 2001), 81–85. 60

600

Konrad Schmid

This final statement is especially applicable in the context of more recent scholarly contributions that place the literary combination of the non-Priestly ancestral and exodus traditions only after the composition of the Priestly document.66 This post-Priestly redaction already had a model for its activity in the Priestly document, which unquestionably presupposed the combination of the ancestors and the exodus and from this perspective can be reckoned as the first proto-Pentateuch in terms of its content.67 The two recognizable narratives in Exod 1–14 (15) were combined in the end redaction. This is characterized first by the fact that it attempts to preserve its sources, to the degree that they can be recognized, as completely as possible and according to their wording.

The end redaction attempts, at the same time, “to combine the preexisting narratives to an overarching view through its own expansions.”68 In reality, Jan Christian Gertz apportions extensive original formulations to the end redaction (e.g., Exod 1:8–10; 3:12–15*, 18–20; 4:1–17, 21–23, 27–31; 5:1–6:1*; 8:21–27; 9:15–23*; 10:21–27; 12:18–20, 42–51), whereas the “expansions to the end-redactional text”69 are identified as rather small in scale (the most extensive passages according to Gertz are Exod 1:1–6 and Exod 12:14–1770). 66  Cf. A. de P ury , “Le cycle de Jacob comme légende autonome des origines dʼIsraël,” in Congress Volume: Leuven, 1989 (ed. J. A. Emerton; VTSup 43; Leiden: Brill, 1991), 78–96; Schmid, Erzväter (see n. 45); Gertz, Tradition (see n. 32), 381–388; E. Otto, Das Deuteronomium im Pentateuch und im Hexateuch (FAT 30; Tübingen: Mohr Siebeck, 2000); idem , Mose: Geschichte und Legende (Munich: Beck, 2006); idem , Das Gesetz des Mose (Darmstadt: Wissenschaftliche Buchgesellschaft, 2007); J. L. Ska, Introduction to Reading the Pentateuch (Winona Lake, IN: Eisenbrauns, 2006), 196–202; critiqued by, e.g., T. B. Dozeman, “The Commission of Moses and the Book of Genesis,” in A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (ed. T. B. Dozeman and K. Schmid; SBLSymS 34; Atlanta: Society of Biblical Literature, 2006), 107–129; J. Van Seters, “The Patriarchs and the Exodus: Bridging the Gap between Two Origin Traditions,” in The Interpretation of Exodus (ed. R. Roukema; CBET 44; Leuven: Peeters, 2006), 1–15; H.-C. S chmitt , “Erzväter- und Exodusgeschichte als konkurrierende Ursprungslegenden Israels – Ein Irrweg der Pentateuchforschung,” in Die Erzväter in der biblischen Tradition (ed. A. C. Hagedorn and H. Pfeiffer; BZAW 400; Berlin: de Gruyter, 2009), 241–266; G. I. Davies, “The Transition from Genesis to Exodus,” in Genesis, Isaiah and Psalms (ed. K. J. Dell et al.; VTSup 135; Leiden: Brill, 2010), 59–78. 67  See esp. A. de P ury , “Pg as the Absolute Beginning,” in Les dernières rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (ed. T. Römer and K. Schmid; BETL 203; Leuven: Peeters, 2007), 99–128; repr. in Die Patriarchen und die Priesterschrift/Les Patriarches et le document sacerdotal: Gesammelte Studien zu seinem 70. Geburtstag/Recueil dʼarticles, à lʼoccasion de son 70e anniversaire (ed. A. de Pury; ATANT 99; Zürich: Theologischer Verlag, 2010), 13–42. 68  Gertz, Tradition (see n. 32), 391. 69  Gertz, Tradition (see n. 32), 394–396. 70  Gertz, Tradition (see n. 32).

Post-Priestly Additions in the Pentateuch

601

Markus Witte draws a comparable picture for Gen 1–11. He also views the end redactor as textually productive, on the one hand taking over preexisting texts in his primeval history71 and on the other inserting a series of his own formulations into the text that he arranges.72 The post-Priestly texts of the Pentateuch (and Hexateuch) receive special attention from and are accorded particular importance by Eckart Otto and Reinhard Achenbach, even though they do not clearly define the content and diachronic order of either the post-Priestly Pentateuch or the Hexateuch redactions: The conceptions of the primary Dtr redactor of Deuteronomy (DtrD) and the Dtr author who combined Deuteronomy and the book of Joshua (DtrL) play key roles in the postPriestly conceptions of the Pentateuch and Hexateuch. The Pentateuch and Hexateuch were developed from Deuteronomy with the integration of the Priestly document. [. . .] This complex redactional process need in no way be attributed to only one author. It is more likely to have been the result of a school of tradition. This is supported by insertions in the spirit of the Hexateuch redaction into the Pentateuch redaction, which itself is later than the Hexateuch redaction, which react to the Pentateuch and vice versa.73

At the same time, the Hexateuch and Pentateuch redactions are conceptually quite different in outlook: The author of the Hexateuch presents a broad Israelite understanding of the land of Israel that includes the inhabitants of the ancient state of Israel with its center in Shechem (Josh 24). In the Pentateuch redaction, the reduced Judean diaspora perspective connected to Ezra prevails. It represses the salvific importance of the land in favor of the Torah. “Israel” is the place where the law is fulfilled. The written Torah assumes the role of the Mosaic mediation of the law so that only law that can be reckoned Mosaic is accorded authority. After the conclusion of the Sinai pericope, controversial laws like those concerning the inheritance rights of daughters (Num 27:1–11) as well as the controversial rationales for the responsibilities of priests, Levites [. . .] and the laity (Lev 10; Num 16–18*) are clarified post-redactionally in Leviticus and Numbers until the Pentateuch is finally closed and no longer expanded in the late fourth century BCE.74

According to Achenbach, the book of Numbers as a whole is a literary product of the Pentateuch and Hexateuch redactions as well as a further theocratic editing. 71  E.g., Gen 6:1–4 and various lists; see Witte, Urgeschichte (see n. 8), 334. For a postPriestly date for Gen 6:1–4, see W. Bührer, “Göttersöhne und Menschentöchter: Gen 6,1–4 als innerbiblische Schriftauslegung,” ZAW 123 (2011), 495–515. 72  E.g., Gen 2:9b–15 as well as the numerous redactional elements in the flood and tower of Babel narratives; see Witte, Urgeschichte (see n. 8), 334. 73  O tto , Das Deuteronomium im Pentateuch und Hexateuch (see n. 66), 243–244. See also idem, “Pentateuch,” RGG (4th ed.) 6:1089–1102, here 1097: “Key to the compositional history of the Pentateuch is the combination of Deut and P into a Pentateuch. The tensions between their different theologies spur the process of scribal mediation between them in the postexilic period through a hexateuchal and a pentateuchal redaction [. . .] whose result is the Pentateuch.” 74  Otto, “Pentateuch” (see n. 73), 1101.

602

Konrad Schmid

[T]he pre-Dtr material of the book of Numbers is from then on only accessible as part of the Hexateuch; its preservation comes as a result of a Hexateuch redaction. As a result an abundance of traditions are preserved that the Deuteronomists had omitted.75

Thomas Römer similarly remarks: Apparently these texts could at that point only be admitted to the book of Numbers because the other four books had already achieved proto-canonical status. In reality, Lev 26:46, or as the case may be Lev 27:43, clearly close the Sinai revelation. [. . .] Afterward, the Sinai pericope was expanded by Num 1–10, and this lengthening was added with a new superscription.76

The approach taken by Achenbach and Römer to the book of Numbers corresponds, mutatis mutandis, to the results yielded by Christophe Nihanʼs analysis of the book of Leviticus. He still finds portions of the original Priestly layer in Leviticus and allocates Lev 1–3; 8–9; and 11–15 to this layer, the end of which he detects in Lev 16.77 However, it is especially in the Holiness Code (Lev 17–26) that post-Priestly expansion becomes palpable, and it should be interpreted in connection with the formation of the first edition of a Pentateuch: H was actually conceived from the beginning in the prospect of a synoptic reading of the so-called “legal” codes comprised within the Torah/Pentateuch. This observation, as well as the nature of the process of inner-biblical exegesis reflected in H, demonstrates that the composition of this code should be related to a first edition of the Pentateuch, as suggested by E. Otto. [. . .] Against E. Otto the classical observation of Hʼs distinctive terminology and theology indicates that H is not simply the work of a “pentateuchal redactor”, but rather a distinct priestly scribal school, as argued by I. Knohl and J. Mil-

 R. Achenbach, “Numeri und Deuteronomium,” in Das Deuteronomium zwischen Penta­ teuch und Deuteronomistischem Geschichtswerk (ed. E. Otto and R. Achenbach; FRLANT 206; Göttingen: Vandenhoeck & Ruprecht, 2004), 23–134, here 124. 76  T. Römer, “Das Buch Numeri und das Ende des Jahwisten: Anfragen zur ‘Quellenscheidung’ im vierten Buch des Pentateuchs,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 215–231, here 223; cf. also R. G. Kratz, “The Pentateuch in Current Research: Consensus and Debate,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 31–61, here 54: “In Numbers we have very few traces of a pre-Priestly tradition. This, however, does not mean that the book as a whole can be declared a post-Priestly document.” See further H. Seebass, “Pentateuch,” in TRE 26:185–209, here 203, for whom Numbers is essentially post-Priestly. A good overview of the discussion is offered by C. Frevel, “The Book of Numbers: Formation, Composition, and Interpretation of a Late Part of the Torah; Some Introductory Remarks,” in Torah and the Book of Numbers (ed. C. Frevel et al.; FAT 2/62; Tübingen: Mohr Siebeck), 1–37. 77  C. Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2006), 20–68. On this discussion, see also J. L. Ska, “Le récit sacerdotal: Une ‘histoire sans fin’?” in The Books of Leviticus and Numbers (ed. T. Römer; BETL 215; Leuven: Peeters, 2008), 631–653. 75

Post-Priestly Additions in the Pentateuch

603

grom, to which the editing of the priestly document in Gen 1–Lev 16 was entrusted at the time of Pʼs inclusion in the Pentateuch.78

Nihan works out a line of thought that was developed earlier by I. Knohl, who found numerous H insertions in the Pentateuch and saw H as the end redaction of the Pentateuch.79 Eckart Otto presented quite a similar argument, “that the Holiness Law can be traced back to the authors of the Pentateuch redaction.”80 In a detailed study of Lev 25:44–46,81 Levinson shows how this text draws both on the manumission law in the Covenant Code and on Exod 1:13f. (P) and thus makes the argument that Lev 25 is post-P. In accordance with these contributions to scholarship, the most recent pentateuchal discourse emphasizes the quantitative and qualitative importance of the final stages of the redaction history.82 The most detailed, and in some ways excessive, reconstruction of the post-Priestly redaction history of the Pentateuch applied to the reconstruction of a specific section was presented in 2010 by Christoph Berner, on Exod 1–15. Berner attributed by far the largest portion of the exodus narrative to post-Priestly expansions, which, however, were generally unconnected with one another, focused on small amounts of text, and originated parallel to one another.83 There are, however, also radically opposing voices. The most critical objection surely comes from Joel Baden, who considers the supposition of a  Nihan, From Priest l y Torah (see n. 76), 616–617.  Cf. I. Knohl, The Sanctuary of Silence (Minneapolis: Fortress, 1995, 59ff); idem, “Who Edited the Pentateuch?” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 359–367. 80  E. Otto, “Innerbiblische Exegese im Heiligkeitsgesetz Leviticus 17–26,” in Die Tora: Studien zum Pentateuch: Gesammelte Aufsätze (ed. E. Otto; BZABR 9; Wiesbaden: Harrassowitz, 2009), 46–106, here 95, n. 235. 81  B. M. L evinson , “The Birth of the Lemma: The Restrictive Reinterpretation of the Covenant Codeʼs Manumission Law by the Holiness Code,” JBL 124 (2005), 617–639. 82  E.g., S ka , Introduction to Reading (see n. 66), 217–229; J. C. G ertz , ed., Grundinformation Altes Testament: Eine Einführung in Literatur, Religion und Geschichte des Alten Testaments (4th ed.; UTB 2745; Göttingen: Vandenhoeck & Ruprecht, 2010), 214–217; D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2012), 216: “our present Pentateuch is, in large part, a product of a Priestly oriented conflation of the P and non-P documents along with late Priestly expansions of various non-P texts” (e.g., Gen 2:4a). Carr offers an entire chapter on these final layers of the composition of the Pentateuch: “The Final (Reconstructible) Stage of the Formation of the Pentateuch” (Carr, Formation, 215–217); cf. E. Zenger and C. Frevel, Einleitung in das Alte Testament (see n. 63), 148–160; T. Römer et al., eds., Einleitung in das Alte Testament: Die Bücher der Hebräischen Bibel und die alttestamentlichen Schriften der katholischen, protestantischen und orthodoxen Kirchen (Zürich: Theologischer Verlag, 2013), 155–164. 83  C. Berner, Die Exoduserzählung: Das literarische Werden einer Ursprungserzählung Israels (FAT 73; Tübingen: Mohr Siebeck, 2010) (see also the review by K. S chmid , ZAW 123 [2010], 292–294). 78 79

604

Konrad Schmid

redactor to be a necessary but quite marginal element of the source theory that he defends: [H]e is a necessary side-effect of the recognition of multiple sources in the text, not a primary feature of the theory. The theory demands a redactor, because the sources were evidently combined by someone – but no more than one.84

Whatever one might conclude about the conditions of the redactor within the framework of a scholarly project on “refining the Documentary Hypothesis”85 that maximizes the role of supposed source texts, Julius Wellhausen cannot be claimed as the ancestor of this perspective. In a letter to Adolf Jülicher, he holds that “the Supplementary Hypothesis [has] validity” and that “the hypothesis of a mechanical mosaic [is] ludicrous.”86 The post-Priestly redaction history of the Pentateuch is truly a complex domain, and the editing process depends on numerous presuppositions about the nature of the composition of the Pentateuch as a whole. Nonetheless, recent scholarship has shown that this area can and should be investigated further.87

84  Baden, J, E, and the Redaction (see n. 6), 8–9; see also 289, 305, as well as the detailed description on 255–286. See also the critical response by D. M. Carr to Baden in http://www. bookreviews.org/pdf/7801_8507.pdf, accessed 10/13/14. B. J. Schwartz similarly determines the function of the redaction based on the example text of Gen 37; see B. J. Schwartz, “How the Compiler of the Pentateuch Worked: The Composition of Genesis 37,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; VTSup 152; Leiden: Brill, 2012), 263–278; idem , “Josephʼs Descent into Egypt: The Composition of Genesis 37,” in The Joseph Story in the Bible and throughout the Ages (ed. L. Mazor; Beit Mikra 55; Jerusalem: Magnes, 2010), 1–30 (Hebrew). 85  See J. S. B aden , The Composition of the Pentateuch: Refining the Documentary Hypothesis (New Haven, CT: Yale University Press, 2012). 86  J. W ellhausen , “Brief vom 8.11.1880,” in Briefe (ed. R. Smend; Tübingen: Mohr Siebeck, 2013), 78. 87  This field of investigation would expect and would include the description of complex conceptions of how the redaction history relates to the formation of the Pentateuch as Torah. See the initial reflections in K. S chmid , “Der Abschluss der Tora als exegetisches und historisches Problem,” in Schriftgelehrte Traditionsliteratur: Fallstudien zur innerbiblischen Schriftauslegung im Alten Testament (ed. K. Schmid; FAT 77; Tübingen: Mohr Siebeck, 2011), 159–184. A disciplined inclusion of the post-Priestly texts of the Pentateuch is offered by T. Römer, “Der Pentateuch,” in Die Entstehung des Alten Testaments (ed. R. Smend et al.; 5th ed.; ThW 1/1; Stuttgart: Kohlhammer, 2014), 53–166.

Part Six The Integration of Preexisting Literary Material in the Pentateuch and the Impact upon Its Final Shape

Introduction Joel S. Baden From the earliest days of critical scholarship, it has been recognized that there are some parts of the Pentateuch that existed independently as literary units before they were introduced into the pentateuchal text in which they are now found. This was most evident in the larger poetic units: the Song of Jacob in Gen 49, the Song of the Sea in Exod 15, and the Song of Moses in Deut 33. The mismatch between these passages and their contexts, as well as their form- and tradition-critical distinctiveness, made them obvious examples of preexisting poetry that had been chosen by a pentateuchal author or editor as appropriate to a given narrative situation. Though historically they may well have preceded the composition of the narratives in which they are now embedded, in their canonical setting these poems operate mostly in the style of commentary on their pentateuchal contexts: looking back on and poeticizing a life or an event. Remove them and nothing fundamental is lost, on the level of plot, or theme, or theology. Other passages, however, also came to be seen by some as similarly of independent origin but with far more impact on the Pentateuch as a whole. The sacrificial legislation of Lev 1–7 is one such example – indeed, for many scholars, the entire block of cultic material at the heart of the Priestly writings in Leviticus was often considered to have been some sort of preexisting Priestly ritual corpus, whether as a single mass or as a variety of individual torot. There has been nearly universal consensus that the Covenant Code, or at least some early version thereof, had its own literary history as an independent legal text before it was taken up as part of the narrative of divine revelation in the wilderness. And quite often the Decalogue, too, is put into this category: a fundamental statement of Israelite law or belief that must have been circulating prior to being treated as the first words spoken by Yahweh to the entire nation of Israel. In these cases, there is no doubt that the insertion of the preexisting material had a decisive influence on the final form of the Pentateuch – which, after all, is traditionally viewed as the lawbook and ritual handbook par excellence in both Judaism and (less favorably) Christianity. And yet the very reason that scholars were able so cheerfully to isolate these passages as independent (and secondary) was that these texts did not disturb the overall flow of the narrative, the arc leading from Abraham to Moses, from Canaan to Egypt and back again, the

608

Joel S. Baden

Heilsgeschichte that was for so long seen as the raison dʼêtre of the Pentateuch from its earliest days to its latest interpretation. In more current research, taking its cues from the seminal work of Martin Noth but beginning in earnest with that of Rolf Rendtorff and Erhard Blum, the situation has changed radically. It is often the case today that scholars identify major sections of even the basic pentateuchal narrative as being of independent literary origin. The Jacob cycle, or the Moses story, or the Joseph novella – these and many others have been taken as having existed quite on their own, not only without the surrounding and overarching pentateuchal narrative arc but indeed without any need for it at all. In other words, as the Documentary Hypothesis, with its notion of four contemporaneous, independent, and self-standing “mini-Pentateuchs,” has been set aside in favor of a single Pentateuch with a far more complex process of growth, the number of independent and self-standing “partial-Pentateuchs” has dramatically increased. As there is no discernible significant genre distinction between, for example, the Jacob cycle and the rest of the pentateuchal narrative in which it is embedded, these new views require a certain amount of extra explanation and rationalization. In order to be convincing, certain questions must be addressed. On what grounds is the given passage (or collection of passages) a self-contained unit, that it might have ever existed independently? By what literary mechanisms was it incorporated into a broader framework? What was its Sitz im Leben when independent? Why was it believed to be appropriate to its new literary position? And, if applicable: What did the “pentateuchal” text look like before this independent piece was added to it, and what purpose did it serve? These are questions that are quite easily answerable when directed toward something like the song of Jacob in Gen 49, for example, or the ritual laws of Lev 1–7. And indeed they were answered readily in those cases – these are not new questions. But they press with increasing intensity as the number, scope, and generic diversity of posited preexisting independent units expands. In the end, the possibility that substantial parts of the Pentateuch as we know it, as it has been known for more than two thousand years, may in fact have been independent literary works that have been stapled and shuffled and redacted into their respective places can be somewhat disarming and perhaps alarming. It raises the question of what, exactly, this Pentateuch of ours is. Is it a happenstance collection of bits and pieces? Is its shape a matter of intent? Which part is the core – the originally independent units, or the glue that binds them together? The essays that follow participate in this broader conversation. Tackling texts from the poetic to the legal to the narrative, from a single chapter to quite a substantial chunk of the Pentateuchʼs most central material, they aim to answer, as best as possible, the kinds of questions and concerns raised above.

Noncontinuous Literary Sources Taken Up in the Book of Exodus Rainer Albertz There is near consensus that there is some preexisting literary material integrated into the book of Exodus. Many scholars would include the Songs of Moses (Exod 15:1–18*) and of Miriam (v. 21) or the Book of the Covenant (20:22–23:19). Only the extent of that material and its range are disputed. Although Hermann Gunkel has shown that in the book of Genesis the patriarchal story included smaller narrative compositions (Sagenkränze) such as the Abraham-Lot and the Jacob-Esau stories,1 a thesis extended by Erhard Blum from the oral to the literary stage of text formation,2 most scholars regard the book of Exodus as part of a much longer narrative thread running from the birth until the death Moses (Exod 2–Deut 34) or even to the conquest of the promised land in the book of Joshua. Hugo Gressmannʼs attempt to distinguish smaller compositions, which he called “Moses legends,” “Jethro-Moses legends,” or “Kadesh legends,” has failed.3 Even Blum confined himself to the reconstruction of large compositions in this area, namely, a late Deuteronomistic (KD) and a Priestly one (KP), and only tentatively suggested a vita Mosis behind them.4 Preparing my commentary on the book of Exodus,5 however, I noticed that this book was also composed of several smaller narrative, poetic, legislative, and cult programmatic units, which had probably existed in written form before they were integrated into larger compositions.6 Because of the limited space in  See H. Gunkel, Genesis (7th ed.; HK 1/1; Göttingen: Vandenhoeck & Ruprecht, 1966), li–liii, 214–217, 291–293. 2  See E. Blum, Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 273–361; 461–477. 3  See H. G ressmann , Mose und seine Zeit: Ein Kommentar zu den Mosesagen (FRLANT 18; Göttingen: Vandenhoeck & Ruprecht, 1913), 386–392. 4  See E. Blum, Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 216–218; Blum was skeptical as to whether such older Vorlagen could be reconstructed, since their beginnings and conclusions are missing. 5  See R. A lbertz , Exodus, Band I: Ex 1–18 (ZBK 2/1; Zürich: Theologischer Verlag, 2012); idem, Exodus, Band II: Ex 19–40 (ZBK 2/2; Zürich: Theologischer Verlag, 2015). 6  See R. A lbertz , “Die vergessene Heilsmittlerschaft des Mose: Erste Überlegungen zu einem spätexilischen Exodusbuch,” EvTh 69 (2009), 443–459; idem , “The Late Exilic Book of Exodus (Exodus 1–34*): A Contribution to the Pentateuchal Discussion,” in The 1

610

Rainer Albertz

this extended volume, I would like to restrict myself to the narrative material of the non-Priestly texts. From this corpus, three examples representing a low, middle, and high level of text formation will be presented.

1 The Beginning of a Political Moses Story In Exod 2 we encounter a well-structured narrative sequence, which can be called a political Moses story, since Moses, having been saved after his birth and brought up by a daughter of Pharaoh (vv. 1–10), killed an Egyptian officer, who tortured a Hebrew compulsory worker (vv. 11–15). But what could have been a signal for an uprising failed. Moses was denounced and persecuted by Pharaoh and had to flee to Midian, where he found accommodation at the house of Jethro (vv. 16–22). The midwife scene in Exod 1:15–22* probably also belongs to this story, since it provides it with the reason for the mortal threat to the baby Moses.7 The preexisting character of this narrative can be seen in the abrupt introduction of the midwives in v. 15 and its suddenly narrowed perspective, which was only secondarily applied to the main topic of the chapter (vv. 9–12), the increase of the people, through the insertion of vv. 20b–21a.8 Since the days of the Greek translators it has often been observed that the political Moses story, which is interrupted by a Priestly passage (2:23aβ–25) and the non-Priestly vocation story (3:1–4:18) after 2:23aα, originally found its continuation in Exod 4:19–20a.9 After the death of Pharaoh, who had persecuted Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 243–256; R. A lbertz , “Der Beginn der vorpriesterlichen Exoduskomposition (KEX): Eine Kompositions- und Redaktionsgeschichte von Ex 1–5,” ThZ 67 (2011), 223–262. For the first part of my textual reconstruction of KEX and its integrated sources, see Albertz, Exodus I (see n. 5), 19–21. Even the Priestly editor used older sources in Exod 25–31. 7  I have shown in Albertz, “Beginn” (see n. 6), 227–229, that Exod 2:1–10 cannot stand on its own. 8  Note the resumptive repetition of Exod 1:17 in v. 21a. 9  The Septuagint repeats the sentence about the death of the king of Egypt (Exod 2:23aα) in 4:19. That both verses represent an original literary connection was already observed by B. D. E erdmans , Alttestamentliche Studien III: Das Buch Exodus (Gießen: Töpelmann, 1910), 8; W. R udolph , Der ‘Elohist’ von Exodus bis Josua (BWANT 68; Stuttgart: Kohlhammer, 1938), 6–7; M. Noth, Überlieferungsgeschichte des Pentateuch (2nd ed.; Stuttgart: Kohlhammer, 1948), 103; Blum, Studien (see n. 4), 20–22; C. Levin, Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993), 327–329; K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 188–190; J. C. G ertz , Tradition und Redaktion in der Exoduserzählung: Untersuchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000), 254–256.

Noncontinuous Literary Sources Taken Up in the Book of Exodus

611

Moses, the deity intervenes in order to bring the liberator back to his oppressed people in Egypt. The brief scene in which Zipporah saves her husband from a divine attack (4:24–26) may belong to the same story. In any case, the political Moses story ends somewhere in chapter 4 and is thus incomplete. Many details that we would expect after what has been narrated – how the new pharaoh reacted to Mosesʼs return, how Moses found new trust among his companions, whether he received support from the Egyptian princess who had brought him up, and what role Zipporah played during Israelʼs liberation – are no longer told in the present book of Exodus. After Exod 4:20a or 26, there is clearly a severe discontinuity in the Exodus narration. The exposition of the political Moses story is also lacking. Thus, it constitutes a noncontinuous literary source. The result can be interpreted in a traditional-historical way: the composer of the pre-Priestly book of Exodus, be it J or E or the author of the exodus composition (REX), took up an older source about Moses in a selective way. In order to use it for his larger composition, he cut off its beginning and its end. Moreover, he amplified it with an explicit vocation story (3:1–22*; 4:18), where he foretold all the events that would happen in place of those expected: After the negotiations with Pharaoh had failed, YHWH himself would transfer the fight for Israelʼs liberation from the political to the theological arena (3:18–22). Thus, what can clearly be identified as an intrusion into the political Moses story10 creates the literary links to the following plague narrative. Although we can distinguish two older narrative layers in Exod 1–4*, the literary unity and continuity on the level of the Exodus composer (REX) can be maintained.11

 In order to save the vocation story Exod 3–4* for J and E, M. Noth, Das zweite Buch Mose: Exodus (4th ed.; ATD 5; Göttingen: Vandenhoeck & Ruprecht, 1966), 21–22, neglected his own insight in N oth , Überlieferungsgeschichte (see n. 9), 103, that Exod 2:15–23aα and 4:19–20a constitute an original literary connection. For similar reasons, C. Berner , Exoduserzählung: Das literarische Werden der Ursprungslegende Israels (FAT 73; Tübingen: Mohr Siebeck, 2010), 56–59, explicitly questioned this connection in order to reconstruct an original narrative thread of the first chapters of the book of Exodus, including chapters 3–4. But none of his arguments are convincing. 11  There is no need to assign the vocation story (Exod 3:1–22*) to a much later literary layer; see A lbertz , Exodus I (see n. 5), 67–77; and see T. R ömer , “Exodus 3–4 und die aktuelle Pentateuchdiskussion,” in The Interpretation of Exodus: Studies in Honour of Cornelis Houtman (ed. R. Roukema; CBET 44; Leuven: Peeters, 2006), 65–79, here 68–70, pace E. O tto , “Die nachpriesterliche Pentateuchredaktion im Buch Exodus,” in Studies in the Book of Exodus: Redaction – Reception – Interpretation (ed. M. Vervenne; BETL 126; Leuven: Leuven University Press, 1996), 61–111, here 101–111; Schmid , Erzväter (see n. 9), 186–209. 10

612

Rainer Albertz

2 The Body of the Plague Narrative Chapters 7–12 of Exodus consist almost entirely of the extended plague narrative. In its non-Priestly stage of formation, this narrative consisted of a structured sequence of six (three times two) increasing divine sanctions against Pharaoh accompanied by Mosesʼs more and more dramatic negotiations with him, with the seventh sanction, the killing of the Egyptian firstborn, forcing Pharaoh so far to his knees that he accepted Israelʼs emigration.12 The plague narrative has probably already integrated the story about the negotiations on relief from compulsory labor in Exod 5, to which it refers back several times (7:14, 16; 8:23, 25).13 Thus, at its full length it relates how YHWH took over the struggle with Pharaoh after the efforts on a human level had failed. This extended, well-structured narrative composition, however, does not constitute the main compositional level of the pre-Priestly book of Exodus. That becomes clear from its end, where its final motif, that the Israelites were driven out from Egypt so quickly that they had no time to prepare their food for the journey in a proper way (Exod 12:33–34, 39), is interrupted by two compositional links. The first one, in vv. 35–36, which recounts that the Israelites even earned farewell presents from their Egyptian neighbors, looks back to 11:2–3 and 3:20–21. The second, which mentions Israelʼs first wandering from Rameses to Succoth (12:37, 38b), looks forward and introduces the pre-Priestly itineraries of the journey to Mount Sinai (13:20; 15:22, 27; 16:1aα; 17:1b; 19:2aα). Thus, these verses were inserted by the pre-Priestly Exodus composer, however he is designated (J, E, JE, KD, or REX), and the extended plague narrative belongs to the preexisting literary material of the book of Exodus as well. Its final motif, that the first food of freedom turned out to be unleavened (12:39), probably originally led to the foundation of the Matsot feast, which is lost in the present text of the book.14 Its beginning, too, before chapter 5, was replaced by the 12  The seven stories within Exod 7:14–12:39* constitute a well-structured composition, as shown by B. J acob , Das Buch Exodus (ed. S. Mayer; Stuttgart: Calwer Verlag, 1997), 174–183, and M. G reenberg , “The Redaction of the Plague Narrative,” in Near Eastern Studies: Studies in Honor of William Foxwell Albright (ed. H. Goedicke; Baltimore: Johns Hopkins University Press, 1971), 243–352, here 244–250. There is no reason to attribute the plague stories of Exod 9–10 to a later or even the final redaction (thus, Gertz, Tradition [see n. 9], 74–188, and Berner, Exoduserzählung [see n. 10], 168–250) if one has accepted that P was not only an author of his own passages but also operated as an editor of the received pre-Priestly material. See Albertz, Exodus I (see n. 5), 163–174. 13  For the disputed evaluation of Exod 5, see A lbertz , “Beginn” (see n. 6), 248–257; Albertz, Exodus I (see n. 5), 102–106. The literary bridge from Exod 5:3–19 to the beginning of the plague story in 7:14–15 is still visible in 5:20–21; note the rare phrase niṣṣab liqrāʾt, which only appears in these two passages of the Hebrew Bible. 14  The original etiological conclusion of the plague narrative seems to be supplanted by the Matsot regulations of the late Deuteronomistic editor (Exod 13:1–10), where it perhaps

Noncontinuous Literary Sources Taken Up in the Book of Exodus

613

redactor with a newly formulated introduction in Exod 1:9–12 and the political Moses story, as already shown.15 Thus, the plague story, too, is an incomplete and noncontinuous literary source within the book of Exodus. It can be compared to one of the reused and reworked subunits of the Gilgamesh Epic, to which David Carr refers as an example of book formation in ancient Near Eastern culture.16

3 The Final Scene of the Non-Priestly Exodus Narrative There is a broad consensus that the non-Priestly exodus narrative comes to its end in Exod 32–34, before the Priestly report on the construction of the tabernacle revealed in Exod 25–31 filled the rest of the book in chapters 35–40. Many scholars also accept that Exod 32–34 concludes that narrative, inasmuch as Israelʼs covenant with YHWH at Mount Sinai (19–24), which was broken by Israelʼs veneration of the golden calf (32), is renewed in chapter 34 because of Mosesʼs numerous intercessions (32:11–12, 29–33; 33:12–17; 34:9), Israelʼs repentance (33:4, 6), and Godʼs overwhelming compassion (34:6–7).17 According to my own research, the main problem treated in these chapters is that YHWHʼs extreme closeness to Israel, which had been established by the original covenant – even laymen could take a meal in proximity to the deity (24:9–11) – became a lethal threat after Israelʼs apostasy. Now, just a moment of YHWHʼs presence in Israelʼs midst would kill the people (33:3, 5). Thus, God decides to stop accompanying the Israelites in order to cede them a chance of survival. Moses, however, persuades God not to do this (33:12–17; 34:9). In the end, Moses himself, whose face shone because of his encounter with the divine (vv. 29–32*), turned out to be Godʼs representative, who would portray the divine presence among the people in a less dangerous form. As YHWHʼs mediator of salvation, he would guarantee the renewed covenant for the future.18 If this interpretation is correct, the non-Priestly exodus narrative did not need any continuation but would refer directly to the present and future of its glimmers through in vv. 3–4, a passage that constitutes a kind of doublet to the following admonition speech (vv. 5–10) and differs from it with regard to the number. 15  See the compositional link in Exod 5:5, which secondarily introduces the topic of the peopleʼs increase from 1:9–12. 16  See D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011), 40–48. 17  See the research report of K. S chmid , “Israel am Sinai: Etappen der Forschungs­ge­ schich­te zu Ex 32–34 in seinen Kontexten,” in Gottes Volk am Sinai: Untersuchungen zu Ex 32–34 und Dtn 9–10 (ed. M. Köckert and E. Blum; Veröffentlichungen der Wissen­schaft­li­ chen Ge­sell­schaft für Theologie 18; Gütersloh: Gütersloher Verlagshaus, 2001), 9–40, and especially Blum, Studien (see n. 4), 45–72. 18  See Albertz, “Heilsmittlerschaft” (see n. 6), 450–459; idem, “Late Exilic Book” (see n. 6), 248–253; idem, Exodus II (n. 5), 321–324.

614

Rainer Albertz

contemporary audience, probably of the late exilic generation. That Exod 34:32 actually constituted the end of a literary composition can be supported by several additional observations: First, the striking motif of Mosesʼs shining face is never taken up again in the rest of the Pentateuch; the reference to Mosesʼs splendor (hôd) in the late Priestly text Num 27:20 is a vague allusion at best. Second, although the theological motif of YHWHʼs and Mosesʼs presence in the midst of the people, so central in Exod 33–34 (33:3, 5, 12–17; 34:9–10), is known in Num 11:20–21, it has lost any significance and is not further developed. Third, all passages of Exod 32–34 that explicitly refer to future events – the promised land (33:1b), its conquest (33:2), and the future revelations to Moses at the tent of meeting (33:7–11) – are clearly later insertions into the original narrative context. The same is true of the appendix 34:33–35, which intends to integrate Mosesʼs shining face into a repeatable revelation praxis.19 Fourth, Exod 32–34 does not really prepare for the departure from Mount Sinai. YHWHʼs order for departure is located too early, in 33:1a. The topic is used here only for dealing with the problem of further divine company (vv. 3–6, 12–17).20 Exodus 34 has nothing to do with Israelʼs wanderings; the extraordinary miracles, which are mentioned in v. 10, refer not to them but to Mosesʼs transfiguration. Unsurprisingly, Julius Wellhausen regarded this chapter as totally out of place.21 Even more problems are raised by the non-Priestly passages that appear again after the bulk of the Priestly Sinai texts from Num 10:29 onward. They do not really follow Exod 34. To mention just one example: Joel Baden, who is very skillful at reconstructing smooth narrative threads, recently made the suggestion that Num 10:29 directly followed Exod 34:27 in the Yahwistic document.22 If one reads these verses one after another, however – “YHWH spoke to Moses: ‘Write these words down, because in accordance with these words I will make a covenant with you and with Israel.’ And Moses spoke to the Midianite Hobab ben Reuel, Mosesʼs father-in-law: ‘We are setting out for the place that YHWH has promised to give us. Come with us, and we will  See Albertz, “Heilsmittlerschaft” (see n. 6), 454–456; idem, “Late Exilic Book” (see n. 6), 245–247. Exod 33:1b, 7–11; 34:33–35 can be attributed to D and 33:2 to the Malʼak redaction (MalR), as elaborated by Blum, Studien (see n. 4), 365–378. 20  Noth, Exodus (see n. 10), 208, stated that Exod 33 reflects the topic of “the presence of God in the midst of the people”; he was rightly followed by T. B. Dozeman, Commentary on Exodus (ECC; Grand Rapids, MI: Eerdmans, 2009), 718. 21  See J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (4th ed.; Berlin: de Gruyter, 1963), 334. Among his postscripts, he even stated that Exod 34, which he regarded as the Yahwistic version of the Elohistic theophany story in Exod 19–20, “has been so to speak thrown in the junk room.” Like E. Blum, Textgestalt und Komposition: Exegetische Beiträge zu Tora und Vordere Propheten (FAT 69; Tübingen: Mohr Siebeck, 2010), 157–176, I regard Exod 34:11–27 as an intrusion of MalR. 22  J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (New Haven, CT: Yale University Press, 2012), 79. 19

Noncontinuous Literary Sources Taken Up in the Book of Exodus

615

deal generously with you [. . .]’ ” – the result is not really convincing. Mosesʼs reaction appears to be somewhat strange, if not impolite. Instead of following the divine command, he starts a conversation with someone else. The Midianites do not play any role in Exod 32–34. They are not even mentioned after Exod 18, before the Sinai theophany started. Thus, Num 10:29–32 is better understood as a redactional frame around the Sinai pericope, together with Exod 18. Whereas Jethro went home after having given Moses substantial help (Exod 18:27), his son Hobab is invited to join Israel in the hope that he can help orient the people (Num 10:29, 32). According to my model, both passages can be assigned to the Hexateuch redactor (HexR),23 who presupposes even later Priestly texts. Thus, it is more likely that Num 10:29–32 was written for its present context.24 The next unit, Num 10:33–36, which starts with an implicit subject, “They set off from the mountain of YHWH for a three-day walk, while the ark of the covenant of YHWH moved before them [. . .] to find them a place of rest,” followed by the ark sayings, could perhaps be connected to Exod 34:32 syntactically, but it does not fit with regard to content, because neither in Exod 32–34 nor in the pre-Priestly exodus narrative is the ark even mentioned. Many scholars have racked their brains to understand why or where the non-Priestly report about the construction of the ark got lost,25 but without any convincing results. In my view, it is much easier to admit that in our received text such a report can only be found in the Priestly revelation of the tabernacle (Exod 25:10–22; cf. 40:20–21; Num 3:31; 4:5). This would mean that the non-Priestly verses Num 10:33–36 already presuppose the Priestly tabernacle report of the book of Exodus. Syntactically, they could have followed the Priestly subscript to the Sinai revelation in Lev 26:46 or 27:34.26 The suggestion that the non-Priestly passages of the book of Numbers come from a later date than most of those in the book of Exodus is supported by additional observations. The first narrative in Numbers, the combined quail and elders story in Num 11:4–34,27 refers back to the tent of meeting outside the  In defense of ascribing Exod 18 to the HexR, see Albertz, Exodus I (see n. 5), 296–314.  See the brief transitional remark “Then” or “When they set off” in Num 10:28b, after the Priestly report about the order of departure in 10:11–28a. 25  See, for example, W ellhausen , Composition (see n. 21), 93; B. Baentsch, Exodus– Leviticus–Numeri (HAT 1/2; Göttingen: Vandenhoeck & Ruprecht, 1903), 274; Blum, Studien (see n. 4), 135–141; Baden, Composition (see n. 22), 276, n. 127. 26  In both subscripts, bənê yiśrāʾēl, the missing subject of Num 10:33, are mentioned. 27  The two topics of Num 11:4–34, feeding the people quails and the inspiration of the seventy elders, cannot be separated into two different literary layers. Both are dependent upon the lament of the people in vv. 4–17. The thematic term ʾāsap, “to collect, to assemble,” which is used six times, appears in the context of both topics (for collecting the quails, see vv. 22, 32[2×]; for assembling the elders, see vv. 16, 24; and for their return, see v. 30). The phrase kol-hāʿām hazzê is used with regard to both topics in vv. 11, 12, 13, 14. That the quail material is also formulated by the late Deuteronomistic author can be seen from v. 10, where he uses a 23 24

616

Rainer Albertz

camp (vv. 10, 16–17, 24b–29) and thus to the very passage that constituted a later insertion into Exod 33 (vv. 7–11).28 Moreover, it explicitly refers back to the priestly manna story of Exod 16, a chapter in which, according to many scholars, no pre-Priestly stratum can be found.29 Within the quail and elders story, one passage is intended to provide the reader with information about the manna (Num 11:7–9). This passage exhibits several verbal allusions, especially to the Priestly portions of Exod 16. In Num 11:8, only the rare verb lāqaṭ, “to collect,” is used, which functions as a key word in the Priestly story (see Exod 16:16–18, 21, 26–27; secondarily used in v. 5); apart from Exod 16 it appears only once more in the Pentateuch (Gen 31:46). The rather strange idea of the Priestly story that the manna can be cooked (bāšal; Exod 16:23) is taken up in Num 11:8. It has a narrative function only in the context of Exod 16, where it emphasizes the wonderful character of the Sabbath meal. Numbers 11:9 mentions that the manna came down in Israelʼs camp together with the dew, which recalls Exod 16:13b–14. This motif, which seems to be superfluous information in the context of Num 11, has great significance for the process of the discovery of the manna depicted in Exod 16.30 Thus, there are good reasons for concluding that Num 11:7–9 is literarily dependent upon the Priestly formulation of the manna story in Exod 16. Since the passage aims to remind readers of previous events in Israelʼs salvation history, it functions as an explicit literary link to the book of Exodus and makes sense as the product of an author who intended to supplement this history with a new chapter. Thus, the passage cannot be regarded as secondary in the combined quail and elders story; Num 11:4–34 altogether is of a post-Priestly date. We have to conclude that the literary continuity between the book of Exodus (along with Leviticus) and the phrase from Exod 33:8, 10, “everybody at the entrance of his tent,” the very unit to which he refers in the verses in which he depicts how the seventy elders got a share of Mosesʼs spirit (see Num 11:17, 25 and Exod 33:5, 9–10). 28  The affiliation of Exod 33:7–11 to the late Deuteronomistic composition KD was shown by Blum, Studien (see n. 4), 61–62; 76–88. Meanwhile, the secondary position of the passage in its context and its post-Priestly dating has been demonstrated by several scholars; see R. Albertz, “Ex 33,7–11, ein Schlüsseltext für die Rekonstruktion der Redaktionsgeschichte des Pentateuch,” BN 149 (2011), 13–43. 29  See E. Ruprecht, “Stellung und Bedeutung der Erzählung vom Mannawunder (Ex 16) im Aufbau der Priesterschrift,” ZAW 86 (1974), 267–307, here 271–291; Blum, Studien (see n. 4), 146–152; Albertz, Exodus I (see n. 5), 262–265, and others. Here I have shown that the Priestly manna story of Exod 16* is located at its original position in the work of the first Priestly editor (PB1). Thus, the suggestions of J. S. Baden, “The Original Place of the Priestly Manna Story in Exodus 16,” ZAW 122 (2010), 491–504, seem to be superfluous. According to my research, the non-Priestly passages of the chapter (Ex 16:4–5, 28–29) were added by MalR. 30  The fact that only Exod 16:31 and Num 11:7–8 compare the appearance of manna to coriander seed and mention its taste (ṭaʿam) does not help for determining the direction of literary dependence, because the former passage seems to be a later Priestly addition (PB3) to the manna story, where the influence might have run in the opposite direction.

Noncontinuous Literary Sources Taken Up in the Book of Exodus

617

book of Numbers was not established until a later date, that is, at a post-Priestly redactional stage. In my model, Num 10:33–36; 11:1–35; and Exod 33:7–11 belong to the late Deuteronomistic redaction D, which corresponds more or less to Blumʼs KD but has to be dated later than the first two Priestly layers.31 All these observations provide good reasons for concluding that Exod 34 not only shows the final scene of a narrative sequence but also constitutes the end of a literary composition, which was originally not continued. What looks like a continuation in the book of Numbers does not fit and seems to belong to later literary layers. Therefore, I speak of a specific Exodus composition (KEX) containing most of the non-Priestly passages of the book within the wide range from Exod 1:9 to 34:32. This composition was not completely preserved, either; its exposition was lost in the redactional work at the beginning of the book.32 Whether some concluding remarks are missing at the end is not certain. Thus, a large but noncontinuous literary unit seems to have existed in the book of Exodus even at a higher level of text formation, where many scholars tend to assume continuous literary documents or comprehensive redactional compositions.

4 A Brief Conclusion The observation that the Exodus composition probably came to its conclusion in Exod 34:32 has an important methodological consequence. It suggests that the discontinuity of the narrative flow is not only restricted to the preexisting material taken up by the composer of larger works but can also be found at the level of larger compositions themselves. This means that it cannot be taken for granted any longer that those sources of the Pentateuch that are situated in a higher compositional layer, whether they are designated J, E, JE, P or otherwise, constitute literary units throughout the entire range of the pentateuchal or hexateuchal narrative. How far such sources actually extended has to be carefully investigated in each case, as has already been done in the case of the preexisting material at lower levels of text formation.

31  See R. A lbertz , “Das Buch Numeri jenseits der Quellentheorie: Eine Redaktionsgeschichte von Num 20–24,” ZAW 123 (2011), 171–183; 336–347, here 336–338; idem , “Schlüsseltext” (see n. 28), 13; 37–39. According to my research, it was the D redaction that connected the Priestly edited books of Gen–Lev (Triteuch) with the already existing Dtr book of Deut by creating the first parts of the book of Numbers. 32  The first verse that can attributed to the Exodus Composition (KEX), Exod 1:9, lacks its subject, which – in the present text – is already named in 1:8, a passage introduced by the HexR together with vv. 1b, 5b–6; for a more detailed discussion, see my analysis in Albertz, “Beginn” (see n. 6), 227–238.

The Story of the Gadites and the Reubenites (Numbers 32) A Case Study for an Approach to a Pentateuchal Text Itamar Kislev Given the Pentateuchʼs lack of textual uniformity, scholars rule out its composition by a single author. As is well known, there are three principal approaches to the question of the lack of uniformity of the pentateuchal text and its composition. The first is the Documentary Hypothesis, which attributes the text of the Torah to the interweaving of several, usually four, independent documents, each of which includes the entire story of Israelʼs beginning.1 A second approach is the Supplementary Hypothesis, which ascribes the lack of uniformity in the Pentateuch to additions made to its basic core by different authors over the course of the generations. The third principal view is the Fragmentary Hypothesis, which proposes that lack of uniformity in the pentateuchal text ensues from the compilation and organization of many fragments from different sources.2 Since we cannot determine with certainty the method used by the compilers of the Pentateuch, independent examination of each biblical unit rather than a priori determination of the authorial method employed in the whole Pentateuch is called for. In my view, in quite a few cases, the Documentary Hypothesis provides a more satisfactory explanation for the formation of the text. But this hypothesis does not always provide the best explanation for the lack of uniformity in the text. Our method must proceed from text to theory and not vice versa. Taking Num 32 as an example, I propose a fundamental perspective that can, in my opinion, contribute to deciding between the approaches in some cases, though I do not presume to provide a detailed comprehensive analysis of the narrative.

 It should be noted that the well-known Documentary Hypothesis of Graf-Kuenen-Wellhausen, as well as its reformulation by Baruch Schwartz and those who accept his conception, are only special applications of this fundamental approach. 2  For a brief survey of these three basic approaches and their expressions in biblical scholarship through the middle of the twentieth century, see, e.g., O. Eissfeldt, The Old Testament: An Introduction (trans. P. R. Ackroyd; 3rd ed.; New York: Harper and Row, 1965), 160–170. 1

620

Itamar Kislev

Numbers 32 tells the story of the Israelite settlement in Transjordan at the time of Moses, at the end of the period of the wandering in the wilderness. I focus only on vv. 1–38, which treat the approval granted to the Gadites and the Reubenites to settle this region. At the end of the chapter (vv. 39–42), we find three anecdotes about three heroes who occupied territories in northern Transjordan and settled there. As many researchers have realized, this passage should be considered an independent fragment joined to the chapter.3 Before proceeding to our discussion, I would like to make a preliminary comment. Throughout the main story (vv. 1–38), just the Gadites and the Reubenites appear, whereas the half-tribe of Manasseh is mentioned only in v. 33. This suggests that the appearance of the half-tribe of Manasseh in this verse is secondary.4 3  E.g., A. R ofé , “Clan Sagas as a Source in Settlement Traditions,” in “A Wise and Discerning Mind”: Essays in Honor of Burke O. Long (ed. S. M. Olyan and R. C. Culley; BJS 325; Providence, RI: Brown University, 2000), 191–203, here 193–194; B. A. L evine , Numbers 21–36 (AB 4B; New York: Doubleday, 2000), 478. In his discussion Rofé tries to reconstruct the original formulation of the fragment. It seems that he is correct in thinking that vv. 41–42 were preserved in their original form (except the necessary splitting of the word ‫ חותיהם‬into ‫[ חות הם‬the villages of Ham]). In v. 39 a redactor changed the subject of the verse from Machir himself to his sons by adding the word ‫ בני‬and altering the original singular verb ‫ וילך‬to the plural ‫וילכו‬. The verb ‫ וילכדה‬is punctuated as plural but its original meaning is the singular, as evidenced by the last verb in the verse, ‫ויורש‬, which appears in the singular (in SP all the verbs in the verse are altered to the plural); cf. G. B. Gray, Numbers (ICC; Edinburgh: T&T Clark, 1903), 438–439. Many scholars considered v. 40 in its entirety to be a redactorʼs work; e.g., J. Wellhausen , Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (3rd ed.; Berlin: Reimer, 1899), 115. It seems possible, however, that the words ‫ וישב בה‬at the end of the verse are part of the original formulation. Cf. recently O. Artus, “Numbers 32: The Problem of the Two and a Half Transjordanian Tribes and the Final Composition of the Book of Numbers,” in Torah and the Book of Numbers (ed. C. Frevel et al.; FAT 2/62; Tübingen: Mohr Siebeck, 2013), 367–382, here 370, who considers v. 40 to be part of the earlier stratum of the chapter. According to him, v. 40 came immediately after v. 33. This view, however, is impossible, since v. 33 and v. 40 are not fully consistent and therefore cannot be part of the same stratum and cannot be sequential. 4  Cf. S. E. Loewenstamm, From Babylon to Canaan: Studies in the Bible and Its Oriental Background (Jerusalem: Magnes, 1992), 112. Some scholars thought the entire verse as secondary; e.g., Gray, Numbers (see n. 3), 432. It seems, however, that the researchers who opined that only some additions were added to the verse are right. Many of them deemed the reference to the tribes names and the conquered territories secondary; e.g., B. B aentsch , Exodus – Leviticus – Numeri (HKAT; Göttingen: Vanderhoeck und Ruprecht, 1903), 669; L. S chmidt , “Die Ansiedlung von Ruben und Gad im Ostjordanland in Numeri 32,1–38,” ZAW 114 (2002), 497–510, here 504. It stands to reason, however, that the mention of Gad and Reuben is original, since the order Gad-Reuben-Manasseh is very rare and appears only twice (Josh 18:7; 2 Kgs 10:33), in comparison to the regular order, Reuben-Gad-Manasseh, which appears twenty-one times. The occurrence of this irregular order arouses our suspicions and leads us to think that the reference to Gad and Reuben in this verse is original, whereas the mention of the half-tribe of Manasseh (and of the conquered territories) is secondary.

The Story of the Gadites and the Reubenites (Numbers 32)

621

The narrative (vv. 1–38), mainly a dialogue between Moses, on the one hand, and the Gadites and the Reubenites, on the other, takes place in several stages. There are various difficult features in the narrative: its excessive length, the presence of doublets, differing emphases and style, and different conceptions. I will mention some but not all of the narrativeʼs difficulties.5 First is the discrepancy between v. 17 and vv. 29–30 regarding the role of the two tribes in the conquest of Canaan. In v. 17, we find the Gadites and the Reubenites offering to lead the army ‫לפני בני ישראל‬. This conception is well reflected in the NJPS translation: “and we will hasten as shock-troops in the van of the Israelites.” Verses 29–30, however, make no mention of their leading the army. Their participation is ‫אתכם‬, “with you.” Another problem is the sudden reference to Godʼs involvement at the end of the narrative (v. 31); up to that point the negotiations have proceeded without divine involvement and without an appeal to God. But at the end of the narrative, in v. 31, the two tribes say: “whatever the Lord has spoken to your servants, so we will do.”6 An additional problem relates to the stage at which the Gadites and the Reubenites will receive their inheritance. The impression conveyed in the account is that they would receive the territory immediately, as explicitly stated in v. 33 (cf. vv. 16, 19, 24, 32, 34–38).7 From v. 29, however, we learn that they will receive their inheritance through Eleazar, Joshua, and the tribal leaders only after fulfilling their commitment to participate in the war against Canaan (cf. vv. 20–22). A further difficulty relates to the question of what would happen if they were to fail to fulfill their promise to participate in the war. In v. 23 Moses utters a vague threat: “But if you do not do this, you have sinned against the Lord; and be sure your sin will find you out.” In contrast, v. 30 explicitly states: “but if they will not cross over with you armed, they shall have possessions among you in the land of Canaan.” Finally, there is the problem of the response of the Gadites and the Reubenites as found in the concluding stage of the dialogue, namely, the necessity of 5  For a detailed list of problems in the narrative, see L oewenstamm , From Babylon to Canaan (see n. 4), 115–116. Marquis underlined three major problems; L. M. Marquis, “The Composition of Numbers 32: A New Proposal,” VT 63 (2013), 408–432, here 410–412. 6  Some scholars proposed that the original reading in this verse in place of Yahweh is “my lord [Moses],” as in vv. 25, 27; e.g., Baentsch, Exodus – Leviticus – Numeri (see n. 4), 668; J. Van Seters, The Life of Moses (CBET 10; Kampen: Kok Pharos, 1994), 439. There is no textual evidence, however, for this speculation. To the contrary, the similar idea in Josh 22:9, ‫אל ארץ אחזתם אשר נאחזו בה על פי יהוה ביד משה‬, testifies that such an idea (that the condition was made according to God) existed in a similar Priestly context. 7  It seems that the statements relating to the (re)building of the cities (Num 32:16, 24, 34–38) presume that the Gadites and the Reubenites will receive the area immediately; cf. Loewenstamm, From Babylon to Canaan (see n. 4), 116.

622

Itamar Kislev

vv. 31–32, which repeat the tribesʼ acceptance of the terms. In the course of the narrative the two tribes again and again express their acceptance of the condition they themselves had set (vv. 16–19, 25–27), which is repeated several times by Moses, too (vv. 20–24, 28–30). The detailed repetition at each stage seems superfluous. We can understand that, following their initial offer (vv. 16–19), Moses repeated their promise to make it official and obligatory (vv. 20–24). It is also understandable that this would be followed by the Gadites and the Reubenites voicing their acceptance, since this was now Mosesʼs proposal (vv. 25–27). We can also comprehend why Moses presents the agreement in the presence of Eleazar, Joshua, and the tribal leaders: in order to make them responsible for its execution after Mosesʼs death (vv. 28–30). Even if we do not understand why the author repeated all the details at each stage, we can find logic for each stage to this point. This is not the case, however, for the additional acceptance by the two tribes in vv. 31–32, after responsibility for the contract was transferred to the future leadership. This repetition is difficult to explain.8 It may be argued that the problems I have noted are not all on the same level and that some can easily be resolved. For example, the first question, whether the Gadites and the Reubenites will lead the army or only participate in the war, can be explained by understanding the phrase ‫ לפני בני ישראל‬in v. 17 as meaning that the two tribes will be part of the army that marches before the Israelites, not that they will lead the army themselves.9 This is only an example, and individual

8  Some have suggested that the two tribes made the earlier affirmation in a private conversation with Moses, and now they publicly affirmed the condition; see T. R. Ashley, The Book of Numbers (NICOT; Grand Rapids, MI: Eerdmans, 1993), 614–615; Marquis, “Composition of Numbers 32” (see n. 5), 418 n. 22. This solution, however, does not explain why the negotiation that begins publicly in v. 2 continues privately and ends publicly again. 9  Or, alternatively, the word ‫ אתכם‬does not necessarily mean in the middle of the military camp and not in front of it. The meaning may be that the two tribes will cross the Jordan together with all the other Israelites. According to this proposal, the word ‫ אתכם‬in vv. 29–30 does not relate to the location of the Gadites and the Reubenites in the military camp. Marquis has surmised that the phrase ‫ לפני יהוה‬in this context (v. 29) alludes to the order of march set forth in Num 2, according to which the Gadites and Reubenites marched in front of the tabernacle but behind Judahʼs standard; Marquis, “Composition of Numbers 32” (see n. 5), 412, and n. 10. This interpretation indeed creates a tension in the story but does not derive from the text itself. The order of march is not implied here, and it could be understood that the verse does not negate the idea that after the two tribes cross the Jordan together with all the people (‫)אתכם‬, they will lead the army as stated in v. 17; cf. J. Milgrom, Numbers (JPS Torah Commentary; Philadelphia: Jewish Publication Society of America, 1990), 271. It is noteworthy that the phrase ‫ לפני יהוה‬does not necessarily teach anything relating to the structure of the military camp, and it may mean only “in the presence of YHWH”; cf. BDB s.v. “[‫]פנה‬,” 817, § II 4 a (h). In Josh 4:12–13, which is concerned with the fulfillment of the promise of the two and a half tribes, the phrases ‫ לפני בני ישראל‬and ‫ לפני יהוה‬appear in proximity.

The Story of the Gadites and the Reubenites (Numbers 32)

623

scholars offer their own evaluations of each difficultyʼs force and each solutionʼs validity.10 The view of most scholars is rightly that the aggregation of problems, those mentioned here and others, makes the option of single authorship of this narrative (vv. 1–38) unlikely.11 The crucial question then is: Which approach best explicates the creation of the narrative and this unitʼs lack of uniformity?12 Some adherents of the Documentary Hypothesis divide the narrative into the usual sources;13 others think that, in this case, the manner in which the redactor combined the sources does not allow us to separate them into independent threads.14 For their part, the supplementarians argue that additions were made to the story in one or several stages.15 Theoretically, either approach is feasible here; nor can we rule out one or the other in advance. Unfortunately, even after profound study, we will not be able to state for certain how this story was composed. Clearly, the story was composed in one specific way. But that does not mean that scholars will be able to determine with certainty the correct way to describe its creation. We are not dealing with certainties. We have no unambiguous criteria for reaching a conclusive determination as to which approach correctly explains a unitʼs formation. All we can say is that one or another approach to an existing text appears more likely and that we need to determine which best explains its formation. We must ask not what the correct explanation is but rather what seems 10  Achenbach, for example, ignores the different results in the instance of nonfulfillment of the two tribesʼ commitment in vv. 23, 30 and attributed both to his Theokratische Bearbeitung; R. Achenbach, Die Vollendung der Tora: Studien zur Redaktionsgeschichte des Numeribuches im Kontext von Hexateuch und Pentateuch (BZABR 3; Wiesbaden: Harrassowitz, 2003), 386–387. 11  Only a few researchers believe that vv. 1–38 are a unified unit; e.g., S. M owinckel , Erwägungen zur Pentateuch Quellenfrage (Oslo: Universitetsforlaget, 1964), 102–104; M. O ttosson , Gilead: Tradition and History (trans. J. Gray; ConBOT 3; Lund: Gleerup, 1969), 74–78; M ilgrom , Numbers (see n. 9), 265–275, 492–494; A shley , Numbers (see n. 8), 605–616. 12  Many scholars, including some proponents of the Documentary Hypothesis, maintain that vv. 7–15 are a late addition; see, e.g., Marquis, “Composition of Numbers 32” (see n. 5), 429–431. 13  E.g., H. H olzinger , Numeri (KHC; Leipzig: Mohr Siebeck, 1903), 152–155; and recently Marquis, “Composition of Numbers 32” (see n. 5), 408–432. Joel Baden has indeed offered an analysis of the chapter and found in it two threads, E and P (J. S. Baden, J, E, and the Redaction of the Pentateuch [FAT 68; Tübingen: Mohr Siebeck, 2009], 142–152), but later he informed me in private communication that he prefers Marquisʼs analysis. 14  E.g., Gray, Numbers (see n. 3), 426: “it [is] more probable that the whole narrative has been recast than that it is the result of simple compilation from JE and P, such as is generally found elsewhere”; Levine, Numbers 21–36 (see n. 3), 478: “The source-critical analysis of Numbers 32 presents particular difficulties [. . .] and it would be of dubious value to attempt a specific breakdown into JE and P.” 15  E.g., Van S eters , Life of Moses (see n. 6), 436–450; and recently A rtus , “Numbers 32” (see n. 3), 369–371.

624

Itamar Kislev

more likely, more reasonable, and more plausible. Furthermore, in the absence of a convincing, agreed-upon criterion for determining the means of formation, we cannot completely rule out the possibility that the text in question is the work of a single author. Let me say, however, that this appears unreasonable. My aim is both to demonstrate a method for examination of the issue at hand and to underscore a criterion that must be taken into account. In my opinion, prior to analyzing this text, two points must be noted. The first is that the narrative displays no sharp, intolerable tensions. For example, throughout the story, with the exception of v. 33, only the Gadites and the Reubenites stand before Moses. This is significant because other biblical passages treating this issue include the half-tribe of Manasseh.16 Moreover, this narrative is one of very few examples in which Gad appears before Reuben (cf. Josh 18:7; 2 Kgs 10:33). With the exception of v. 1, this order is repeated throughout the story.17 An additional example of the absence of tension in the narrative concerns the two tribes taking the initiative regarding both the request and the solution.18 Another is that, throughout the story, there is one condition for their acquisition of the Transjordanian territory: their participation in the war against Canaan. Accordingly, documentarians may attribute the lack of sharp tensions here to the similarity of the narrative threads that form the story, whereas the supplementarians would simply say that the interpolators and the editors took into consideration the original, basic core of the narrative. The second point relates to the fact that a large number of the difficulties are concentrated in the final stage of the dialogue, in which Moses transfers respon16  The object of the addition of the half-tribe of Manasseh to v. 33 (see above, n. 3) is surely to harmonize Num 32 and the other biblical passages relating to this. In SP (and probably in 4QNumb as well) this tendency is more noticeable, and throughout the entire narrative the half-tribe of Manasseh is added to the Gadites and the Reubenites. 17  This is the case in the MT. In LXX and SP, however, Reuben usually precedes Gad (except in v. 2 in SP, and in v. 6 in LXX). The argument made by Loewenstamm that “the list of towns in vv. 34–38 clearly shows that in v. 33 the order Gad-Reuben is the original one” (Loewenstamm, From Babylon to Canaan [see n. 4], 113) demonstrates the textual superiority of MT to LXX and SP in this regard. Some scholars have argued that the order Reuben-Gad in v. 1 is characteristic of a source other than the source that used the order Gad-Reuben; see, e.g., B aentsch , Exodus – Leviticus – Numeri (see n. 4), 660; M arquis , “Composition of Numbers 32” (see n. 5), 429. Theoretically, however, there might be additional explanations for the unique order in v. 1. Loewenstamm, for example, suggests that “Reubenʼs preceding Gad in Num 32:1 probably arose from the slip of a scribeʼs pen; inadvertently he followed the order of the tribes he was used to in the other books of the Bible” (Loewenstamm, From Babylon to Canaan, 114). See also the discussion of U. Schorn, Ruben und das System der zwölf Stämme Israels (BZAW 248; Berlin: de Greuter, 1997), 144–145. 18  In Deut 3:12–20 it seems that the initiative is taken by Moses, since there is no request by the two and a half tribes. Indeed, according to Marquisʼs analysis, it is Moses who takes the initiative in the Priestly thread (Marquis, “Composition of Numbers 32” [see n. 5], 414), but in the current form of the narrative there is no tension in this regard.

The Story of the Gadites and the Reubenites (Numbers 32)

625

sibility to Eleazar, Joshua, and the tribal leaders, with the redundant acceptance of the contractual agreement by the two tribes (vv. 28–32). Only in this part of the dialogue is it explicitly stated that nonfulfillment of the condition will lead to the settlement of the Gadites and the Reubenites in Canaan (v. 30). Only here is the contract ascribed to the divine word (v. 31); and only here is it stated that the two tribes will participate in, not lead, the campaign (vv. 29–30). Should we not then focus our efforts to analyze the story and reveal the development process of its formation on this part? Based on the intolerable redundancy in vv. 31–32, which recount the additional acceptance of the contract by the Gadites and the Reubenites, I suggest that the final part of the dialogue, vv. 28–32, was added at a later stage. Removal of this part of the account does not impair the narrative; moreover, some serious problems disappear. Furthermore, this section adds some significant elements to the story, which perhaps explain the motivation for its addition. First, the turning over of responsibility to Eleazar, Joshua, and the tribal leaders accords with the account of the division of Canaan in the book of Joshua (Josh 14:1; 19:51; 21:1).19 Second, the superfluous stage that recounts the additional acceptance of the condition by the two tribes was probably added in order to underscore divine confirmation of the settlement in Transjordan: “whatever the Lord has spoken to your servants, so we will do.” A closer look at this part of the narrative (vv. 28–32) reveals the presence of many Priestly expressions: Eleazar the priest (v. 28), the family heads of the Israelite tribes (‫ ;ראשי אבות המטות‬v. 28),20 the term ‫( אחזה‬holding; vv. 29, 32), the verb ‫ אח״ז‬in the niphal form (v. 30), the land of Canaan (vv. 30, 32), and perhaps the verb ‫ כב״ש‬with the noun ‫( הארץ‬v. 29) as well.21 This suggests that 19  Cf. Schmidt, “Die Ansiedlung” (see n. 4), 503. Note should be taken of the order of the parts within the tripartite leadership consisting of Eleazar, Joshua, and the tribal leaders, and of the phrase ‫ראשי אבות המטות לבני ישראל‬, which is common to Num 32:28; Josh 14:1; 19:51 (the wording is slightly different); 21:1, all relating to the division of the land. 20  As noted in n. 19 above, the order of the members of the tripartite leadership is shared by some Priestly passages. This order appears also in other cases, in which the term ‫נשיאים‬ occurs instead of ‫ ;ראשי האבות‬see Josh 17:4 and cf. Num 34:16–29. In all these cases Eleazar the priest precedes Joshua, whereas in instances in which Moses is part of the tripartite (or more) leadership he appears first, followed by Eleazar (Num 27:2; 31:12–13, 26; 32:2; 36:1 [according to LXX, Syr, and 4QNumb, which seems preferable]). On the difference between these cases and the historical background of placing Eleazar first, see I. Kislev, “The Investiture of Joshua (Numbers 27:12–23) and the Dispute over the Form of Leadership in Yehud,” VT 59 (2009), 429–445, here 440–445. 21  The verb ‫ כב״ש‬with the noun ‫ הארץ‬appears in Priestly material in Gen 1:28; Num 32:22; Josh 18:1. In the latter two cases ‫ כב״ש‬is in the niphal form as in v. 29 here. The verb also appears in this form in 1 Chr 22:18 (17), in a formulation that is very similar to Num 32:22 and that was undoubtedly influenced by it. The verb has a territory as a subject (if in the niphal) or direct object (if in the qal) only in these verses. In other cases the verb appears with people

626

Itamar Kislev

we have here a Priestly addition22 to a non-Priestly document,23 which prompted us to seek Priestly additions in other parts of the narrative. Such elements appear in v. 2, which mentions Eleazar the priest and ‫נשיאי העדה‬,24 and in v. 5, with the word ‫לאחזה‬.25 In v. 22 we can identify two Priestly expressions: the condition ‫ונכבשה הארץ לפני יהוה‬, “and the land is subdued before the Lord,”26 and ‫והיתה‬ ‫הארץ הזאת לכם לאחוזה לפני יהוה‬, “and this land shall be your possession before the Lord,”27 and it is indeed possible that other Priestly elements were added to the non-Priestly text.28 The Priestly elements were added so that the story would as a subject (niphal) or direct object (qal), except for Zech 9:15, in which the direct object is the obscure ‫אבני קלע‬, but see BHS. 22  Most scholars do not deny the Priestly character of vv. 28–32. Van Seters, though he concedes that the phrases ‫ אלעזר הכהן‬and ‫ ראשי אבות המטות לבני ישראל‬are Priestly (as an addition), argued, ignoring the other characteristically Priestly elements, that the rest of the passage comes from J; Van Seters, Life of Moses (see n. 6), 439. 23  By using the term non-Priestly, I avoid attributing the basic narrative to either J or E. Some scholars use the siglum JE for identifying the non-Priestly thread in this case (e.g., W ellhausen , Composition [see n. 3], 113–115). Others divide the non-Priestly parts into two separate threads (e.g., A. Dillmann, Numeri, Deuteronomium, und Josua [2nd ed.; KeH; Leipzig: Hirzel, 1886], 193–194). Martin Noth does not decide which source to attribute the non-Priestly narrative to here (M. Noth, Numbers [trans. J. D. Martin; OTL; London: SCM Press, 1968], 135–136). Marquis identifies the non-Priestly thread as E (Marquis, “Composition of Numbers 32” [see n. 5], 427–428), but her evidence is inconclusive, since it is mainly based on late additions, namely, the references to Sihon and Og in v. 33 (see n. 4) as well as Num 21:33–35 or the fragment in vv. 39–42 (see above and n. 3). 24  The verse reads better without the words ‫ואל אלעזר הכהן ואל נשיאי העדה לאמר‬, since in vv. 4–5 it seems that they speak with Moses alone; cf. A rtus, “Numbers 32” (see n. 3), 368. This is not a decisive consideration, since the same phenomenon is evident in other Priestly passages (Num 27:2–4; 36:1–2). In the Priestly formulation, however, the verb used is ‫( קר״ב‬Num 27:1; 36:1) and not ‫ בו״א‬as in Num 32:2. It seems, therefore, that the original non-Priestly statement ‫ ויבאו בני גד ובני ראובן ויאמרו אל משה‬was expanded to its current form by adding the Priestly elements Eleazar the priest, the chieftains, and the word ‫ לאמר‬as a kind of a resumptive repetition. 25  This word, too, is unnecessary in this verse, and many consider it as secondary; e.g., Schmidt, “Die Ansiedlung” (see n. 4), 501. 26  On ‫ונכבשה הארץ‬, see above, n. 21. 27  Cf. Driverʼs statement on the Priestly parts of vv. 20–24: “the phrases suggestive of P might even be removed without injury to the narrative”; S. R. D river , An Introduction to the Literature of the Old Testament (Edinburgh: T&T Clark, 1913), 69. The phrase ‫והיתה‬ ‫ הארץ הזאת לכם לאחזה לפני יהוה‬implies that the inheritance is divinely sanctioned. This idea is expressed more explicitly in v. 31, in which the whole contract was divinely approved; cf. H. Seebass, “Erwägungen zu Numeri 32:1–38,” JBL 118 (1999), 33–48, here 36; nonetheless, he attributes 22b and 31 to different strata. 28  V. 4 contains the phrase ‫עדת ישראל‬. The term ‫ עדה‬is clearly Priestly, and therefore this verse, or part of it, may be considered Priestly. LXX, however, reflects the phrase ‫בני ישראל‬ instead, so we cannot be sure of the Priestly involvement in this verse. V. 18 contains the verb ‫ נח״ל‬in the hithpael, which is Priestly, and some scholars indeed consider it (and v. 19, as well)

The Story of the Gadites and the Reubenites (Numbers 32)

627

be similar to and compatible with other Priestly narratives and would emphasize divine involvement in the negotiation and the conquest.29 We can now discern a similarity between the Priestly addition in v. 29 and vv. 20–21, one that cannot be ignored. In both instances, two actions are required as part of the contractual condition: going to war and crossing the Jordan. Not only is the formulation similar in both cases, but there is also a chiastic relationship between them, as seen in the chart: Numbers 32:20–21

Numbers 32:29

‫אם תחלצו‬ ‫לפני יהוה‬ ‫למלחמה‬

. . . ‫אם יעברו‬ ‫את הירדן‬ ‫כל חלוץ‬

‫ועבר לכם‬ ‫כל חלוץ‬ ‫את הירדן‬ ‫לפני יהוה‬

‫למלחמה‬ ‫לפני יהוה‬

In vv. 20–21 the action of going to war comes first and the crossing of the Jordan at the end; in v. 29 the actions are in the opposite order. The chiastic relation between vv. 20–21 and 29 strengthens the connection between them and indicates that the Priestly condition in v. 29 was formulated according to the non-Priestly vv. 20–21. Furthermore, the chiasmus is not limited to the order of the actions but is also apparent in the relationship between the elements of the parallel parts, as shown in the chart: ‫ למלחמה לפני יהוה‬in v. 29; ‫לפני יהוה למלחמה‬ in v. 20; ‫ את הירדן כל חלוץ‬in v. 29; and ‫ כל חלוץ את הירדן‬in v. 21.30 This strong to be Priestly; e.g., Schorn, Ruben (see n. 17), 149. The entire verse seems to be a doublet of the phrase ‫ עד אשר אם הביאנם אל מקומם‬in v. 17; cf. S eebass , “Erwägungen zu Numeri 32:1–38” (see n. 26), 38. V. 18 seems to adjust the condition here to the account in Josh 22:1–6, in which the tribes completed their mission only after the settlement and not merely after the occupation. It is noteworthy that the Priestly addition in v. 22, ‫ונכבשה הארץ לפני יהוה‬, reflects a different conclusion of their mission, namely, after the occupation (so in v. 29, as well). The similarity of this verse and Deut 3:20, with their similar use of the verb ‫( שו״ב‬cf. Josh 1:15) and the distributive use of the noun ‫איש‬, probably indicates that this is an additional edition which drew upon not only Priestly material but also Deuteronomistic texts. 29  Thanks to the additions in vv. 2 and 5, the narrative resembles the account of Zelophehadʼs daughters in Num 27:1–11. Both are narratives of special cases of inheritance. The tripartite leadership that receives the request is the same after the addition (Num 27:2; 32:2), and the request now is formulated in a similar manner (Num 27:4; 32:5). As I noted, the words ‫ והיתה הארץ הזאת לכם לאחזה לפני יהוה‬in v. 22 imply that the condition has divine approval. The phrase ‫ ונכבשה הארץ לפני יהוה‬in v. 22 was probably added to clarify that there will also be an Israelite occupation, and not just a victory, as could be understood from v. 21b. 30  Admittedly, there is also a similar connection between the formulation in v. 27 and that in vv. 20–21 and 29. In regard to v. 27, however, the chiasmus is not as complete as it appears in the chart. Moreover, in both passages, vv. 20–21 and 29, the chiastic connection appears

628

Itamar Kislev

resemblance leads to the conclusion that v. 29, the Priestly verse, was formulated in accordance with vv. 20–21, which are non-Priestly.31 We can now state that the similarity between the two literary threads, according to the different analyses based on the Documentary Hypothesis, is not accidental. It is more likely, rather, that this similarity stems from the fact that this narrative is not composed of independent threads. I find it more plausible that the narrative contains an original core with Priestly additions.32 Thus, in within a two-part condition, whereas v. 27 lacks any condition. Without denying the literary connection of v. 27 to the others, the connection between vv. 20–21 and 29 is stronger. The different formulation of v. 27 indicates that the author of v. 29 tried to accurately imitate the condition in vv. 20–21. 31  Marquis claims that the conditionʼs elements in vv. 20–21, which parallel the condition in v. 29, are part of the Priestly thread. This argument is based on her previously unproven assumption that the two distinct perspectives she identifies must derive from two different independent accounts. Following this she examined Mosesʼs condition in vv. 20–23, where she found two parallel conditions; Marquis, “Composition of Numbers 32” (see n. 5), 413. Clearly, distinct perspectives do not necessarily derive from different independent threads but can be a result of additions to a basic text. Moreover, her Priestly condition in this passage (vv. 20–23) contains only a positive condition (vv. 20b, 21, 22aα, b), whereas the important negative part of the condition is missing. It is noteworthy that Mosesʼs recitation of this Priestly condition in vv. 29–30 before Eleazar, Joshua, and the tribal leaders does contain a negative element. It is more likely that the Priestly editor interpolated only some short additions to this condition in v. 22, as I suggested above, whereas in vv. 29–30, when he himself formulated the whole condition, he more closely modeled it on the non-Priestly two-part condition in vv. 20–23. Furthermore, it was only in his condition, in vv. 29–30, that he could add his result if the tribes were to fail to keep their promise, which is different from the result according to the non-Priestly account (v. 23). The view that there was no independent Priestly thread of this story, with the corresponding absence of an independent Priestly formulation of the condition made by Moses in the early stage of the negotiation, is the clear explanation of the overly late appearance of the Priestly consequences of failing to keep the commitment. Support for the view that v. 21 is not Priestly can be found in the seeming superfluity in the use of both ‫עד‬ ‫ הורישו את איביו מפניו‬at the end of v. 21 and the Priestly ‫ ונכבשה הארץ לפני יהוה‬at the beginning of v. 22. We can conclude from this that at least the end of v. 21 is not Priestly. This consideration is valid also against Seebassʼs claim that vv. 20b–22a should be considered a redactional passage, since it is superfluous (Seebass, “Erwägungen zu Numeri 32:1–38” [see n. 23], 38). Moreover, besides the inability of superfluity to be the sole consideration for the determination of secondariness, the detailed condition in these verses is not necessarily superfluous, since the tribesʼ suggestion now becomes an official one by Moses, as I noted above. 32  The addition of vv. 7–15 is different in nature from that of the Priestly editing that we found throughout the narrative, since the former draws upon Deuteronomistic texts as well as other Priestly and non-Priestly material, as was realized by many researchers; e.g., Marquis, “Composition of Numbers 32” (see n. 5), 429–431. To this editorial layer we may add v. 18 (see above, n. 28) and the additions in v. 33 (see above, n. 4). It is not clear how Schmidt connected the view of the conclusion of the two tribesʼ commitment in v. 18 with that in v. 29 (Schmidt, “Die Ansiedlung” [see n. 4], 504, 507), which differs from that in v. 18 and apparently is linked to v. 22.

The Story of the Gadites and the Reubenites (Numbers 32)

629

this instance, the approach of the Supplementary Hypothesis seems preferable.33 And now to some fundamental statements. Elsewhere I analyzed the text of Num 25 and proposed that, since the Priestly parts of that narrative differ radically from the non-Priestly ones, it is not reasonable to view all the Priestly parts as the work of a redactor; rather, we should postulate the existence of an independent Priestly account.34 Now I wish to build on that claim to create a more complete argument. Where the Priestly parts are very similar to the nonPriestly ones, as in Num 32, we must consider the possibility that the Priestly elements are additions to the non-Priestly verses.35 Where the differences are very distinctive, as in Num 25, we should give greater weight to the approach of the Documentary Hypothesis. That is, neither the Documentary Hypothesis nor the Supplementary Hypothesis is valid in all cases. Rather, for each pentateuchal unit, the specific data must be examined in order to determine which hypothesis best fits the data. Where there are strong differences between the literary parts, the assumption that they come from parallel sources is preferable; where there are strong similarities, however, we must consider giving greater weight to the option of a basic core with additions.

 Some documentarians, too, as I suggest here, think that in this case the Priestly elements are not part of the Priestly source but are rather a Priestly redaction; see, e.g., R. S mend , Die Erzählung des Hexateuch auf ihre Quellen untersucht (Berlin: Reimer, 1912), 236; W. R udolph , Der “Elohist” von Exodus bis Josua (BZAW 68; Berlin: Töpelmann, 1938), 131–134. 34  I. Kislev, “P, Source or Redaction: The Evidence of Numbers 25,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 387–399. 35  In other words, the independent Priestly document (P) did not describe the settlement in Transjordan. Noth, according to his general thesis that P was not at all interested in the conquest and the division of the land, thought that P did not deal with the settlement in Transjordan. He commented that the Priestly style in this chapter is “due to the fact that this is the generally predominant style in the book of Numbers” (Noth , Numbers [see n. 23], 236). It seems, however, that P had a different view of the settlement east of the Jordan. P did not justify it, probably thinking that the only valid settlement is in the land of Canaan, and therefore was silent on the settlement east of the Jordan in the time of Moses. This view is reflected in other Priestly passages, for example in Num 34:1–12. Priestly authors accepted the settlement in Transjordan only at a later stage and justified it by editing ancient texts; such are the Priestly additions in Num 32:1–38, as well as the addition in Num 34:13–15; cf. A. G. Auld, Joshua, Moses and the Land (Edinburgh: T&T Clark, 1980), 76–78, 82. To these we should add the late Priestly composition of Josh 22:9–34; cf. R. Goldstein, “Joshua 22,9–34: A Priestly Narrative from the Second Temple Period,” Shnaton 13 (2002), 43–81 (Hebrew). 33

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative Different Solutions in MT Deut 31:1–32:47 and (the Hebrew Vorlage of) LXX Deut 31:1–32:47 Karin Finsterbusch The Masoretic Text (MT) of Deut 31–321 and the Septuagint (LXX) of the same2 differ significantly in terms of plot. Since the Greek translation of Deuteronomy typically maintains a very close relationship to its Hebrew source text, the differences in the two chapters point to a different Hebrew Vorlage of the Greek text in comparison to the (Proto-)Masoretic one.3 In scholarship, the meaning of the different versions of Deut 31–32 has not been fully recognized yet. The reasons are manifold. The few studies using lower criticism concentrated on single variants and did not notice the implications for the plot.4 The plot of Deut 31–32 has received attention in scholarship on only a  In terms of the narrative structure of Deut 31–32, the Samaritan Pentateuch (Sam) agrees with MT. 2  I use the critical edition provided by J. W. Wevers, Deuteronomium (Göttingen: Vandenhoeck & Ruprecht, 1977). In the verses that are key to the plot, Deut 31–32 in Weverʼs edition is in accordance with the text of Codex B; cf. the Septuagint edition provided by A. E. Brooke and N. McLean, The Old Testament in Greek (Cambridge: Cambridge University Press, 1911), vols. 1 and 3. 3  See especially A. Aejmelaeus, “Die Septuaginta des Deuteronomiums,” in Das Deutero­ nomium und seine Querbeziehungen (ed. T. Veijola; Schriften der Finnischen Exegetischen Gesellschaft 62; Göttingen: Vandenhoeck & Ruprecht, 1996), 1–22; M. K. H. Peters, “Deuteronomion: To the reader,” in A New English Translation of the Septuagint (ed. A. Pietersma and B. G. Wright; New York: Oxford University Press, 2007), 141–146. Some fragments of the Deuteronomy manuscripts from Qumran point to the existence of a Hebrew Vorlage of LXX Deut 31–32 (different from the MT). With regard to the plot, what is especially important is 1QDeutb, frag. 13 ii, which preserves the beginning of Deut 31:1 ([‫)ויכל משה לדבר את כל הד‬, and 4QDeuth, frag. 10, which confirms a long version of Deut 31:9. For a quick overview of the textual evidence, see E. Ulrich, The Biblical Qumran Scrolls: Transcription and Textual Variants (VTSup 134; Leiden: Brill, 2010), 236–243. 4  Cf. L. Laberge, “Le Texte de Deutéronome 31 (Dt 31,1–29; 32,44–47),” in Pentateuchal and Deuteronomistic Studies: Papers Read at the XIIIth IOSOT Congress, Leuven, 1989 (ed. C. Brekelmans and J. Lust; BETL 94; Leuven: Leuven University Press, 1990), 143–160; F. Nwachukwu, “The Textual Differences between the MT and the LXX of Deuteronomy 31: 1

632

Karin Finsterbusch

few occasions; these studies focus mainly on the MT and its problems.5 The most studied text in Deut 31–32 is undoubtedly the so-called Song of Moses (32:1–43), which according to a broad scholarly consensus is of non-Deuteronomic provenance.6 However, in searching for its time of origin, the related narrative context, when it was considered at all, was considered only in the MT.7 In this study, I will not analyze all variants comprehensively but will focus on the plot. I will briefly analyze first the narrative sequence of MT Deut 31:1–32:47 and then that of (the Hebrew Vorlage of) LXX Deut 31:1–32:47. An appendix with a synoptic translation of the key verses provides a quick overview of the main textual differences between the versions. It is my thesis that both versions reflect different processes of redactional integration of texts into other texts. In the case of MT Deut 31:1–32:47, redactors inserted the Song of Moses into an independent edition of Deuteronomy in the late exilic or early postexilic period and reshaped the narrative. The emphasis is on A Response to Leo Laberge,” in Bundesdokument und Gesetz. Studien zum Deuteronomium (ed. G. Braulik; Herders Biblische Studien 4; Freiburg: Herder, 1995), 79–92. 5  Cf. E. Talstra , “Deuteronomy 31: Confusion or Conclusion? The Story of Mosesʼ Threefold Succession,” in Deuteronomy and Deuteronomic Literature: Festschrift C. H. W. Brekelmans (ed. M. Vervenne and J. Lust; BETL 133; Leuven: Leuven University Press, 1997), 87–110; J.-P. Sonnet, The Book within the Book: Writing in Deuteronomy (BibInt 14; Leiden: Brill, 1997), 117–182; B. Britt, “Deuteronomy 31–32 as a Textual Memorial,” BibInt 8 (2000), 358–374; D. Markl, Gottes Volk im Deuteronomium (BZABR 18; Wiesbaden: Harrassowitz, 2012), 133–230. An exception is N. L ohfink, Studien zum Deuteronomium und zur deuteronomistischen Literatur IV (SBAB 31; Stuttgart: Katholisches Bibelwerk, 2000), 219–245. Lohfinkʼs focus was all in all on the MT, but he pointed to some important plot differences between the versions of MT and LXX Deut 31–32 in order to show that plot questions mattered for ancient Israelite authors. With respect to the terminology, Lohfink referred to the difference between fabula and sujet made by the Russian Formalists and used the terms Fabel (chronological order of events) and Erzählfolge (order of narration or narrative order). 6  Scholars have dated the song anywhere from the Mosaic to the Maccabean period. For a survey of the different scholarly positions, see P. Sanders, The Provenance of Deuteronomy 32 (OTS 37; Leiden: Brill, 1996), 1–98, and E. Otto, “Moses Abschiedslied in Deuteronomium 32: Ein Zeugnis der Kanonsbildung in der Hebräischen Bibel,” in Die Tora: Studien zum Pentateuch; Gesammelte Aufsätze (Wiesbaden: Harrassowitz, 2009), 641–678, here 641–650. The song belongs to a group of psalms that appear “irregularly” in the narrative and prophetic literature of the Hebrew Bible; cf. J. W. Watts, “Biblical Psalms outside the Psalter,” in The Book of Psalms: Composition and Reception (ed. P. W. Flint and P. D. Miller; VTSup 99; Leiden: Brill, 2005), 288–309, here 288. 7  Cf. O. E issfeldt , Kleine Schriften (5 vols.; Tübingen: Mohr, 1966–1973), 3:322–334; J. W. Watts, Psalm and Story: Inset Hymns in Hebrew Narrative (JSOTSup 139; Sheffield: JSOT Press, 1992), 63–81; S. Weitzman, Song and Story in Biblical Narrative: The History of a Literary Convention in Ancient Israel (Bloomington: Indiana University Press, 1997), 32–58; M. L euchter , “Why Is the Song of Moses in the Book of Deuteronomy?,” VT 57 (2007), 295–317; Otto, “Moses Abschiedslied” (see n. 6).

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

633

motivating the addressees to live in their future according to the Deuteronomic commandments. In the case of the Hebrew Vorlage of LXX Deut 31:1–32:47, pentateuchal redactors rewrote and restructured the passage. The emphasis is on comprehensive Torah teaching and learning before Israel can start a new life in the promised land.

1 MT Deuteronomy 31:1–32:47 First, I want to present a rough overview of the main scenes and themes of MT Deut 31:1–32:47, outlined in the table below: I: Moses addresses Israel and Joshua about the inheritance of the land 31:1–6

Moses goes and speaks “these words” to all Israel, addressing them about their inheritance of the land.

31:7–8

Moses addresses Joshua in front of Israel, commanding him about his twofold future task.

II: Moses writes the Torah and commands regular public reading of the written Torah 31:9

Narrative statement: Moses writes down “this Torah” and gives the Torah to the Levitical priests and all the elders.

31:10–13

Moses addresses them, commanding them to conduct a public Torah reading regularly.

III: YHWH addresses Moses and Joshua in the tent and gives his revelation of the song 31:14a

YHWH addresses Moses, commanding him to appear together with Joshua in the tent for the purpose of addressing Joshua.

31:14b–15

Narrative statement: Moses obeys; YHWH appears in the tent in a pillar of cloud.

31:16–21

YHWH addresses Moses in the tent, speaking about the future behavior of the people, commanding Moses and Joshua to write and Moses to teach “this song.”

31:22

Narrative statement: Moses writes “this song” on that day and teaches it to the children of Israel.

31:23

He (i.e., YHWH) addresses Joshua (in the tent), commanding him about his first future task.

634

Karin Finsterbusch

IV: After he finishes writing the Torah, Moses speaks the song to the elite; Moses and Joshua speak the song to the people 31:24–29

After Moses finishes writing “the words of this Torah” on a scroll, he addresses the Levites, commanding them to deposit the Torah and to assemble the elite (i.e., all elders of the tribes and the officials), so that he may speak to them “these words” (i.e., the words of the song).

31:30

Narrative statement: Moses speaks “the words of this song” to the “whole assembly of Israel” (i.e., the assembled elite).

32:1–43

The words of the song.

32:44

Narrative statement: Moses comes and speaks “all the words of this song” to the people, “he and Joshua.”

V: Moses finishes speaking “all these words” and admonishes Israel for a final time 32:45

Narrative statement: Moses finishes speaking “all these words” to all Israel.

32:46–47

Moses addresses them, admonishing them for a final time to keep the Torah.

I will now turn to the single passages, briefly describing their narrative logic and pointing to the main inconsistencies in and between the different scenes. I: Moses addresses Israel and Joshua about the inheritance of the land. After Moses establishes the Moabite covenant (Deut 28:69–30:20), the events in the world of Deuteronomy move forward. In Deut 31 the scene changes, indicated by the use of the introductory verb ‫ הלך‬in v. 1 (literally: “he [i.e., Moses] went”). In his speech to the Israelites, Moses focuses on the future inheritance of the land. In his speech to Joshua, Moses appoints Joshua as his successor. Joshua will have to fulfill two tasks: to lead the people to the land and to cause the people to inherit the land.8 II: Moses writes the Torah and commands regular public reading of the written Torah. According to 31:9, Moses wrote the Torah and gave the Torah to the Levitical priests and all the elders of Israel (i.e., the religious and civil representatives of the people).9 What does “give” mean exactly? Closely connected with v. 9 is 8  Cf. G. Braulik, “Die deuteronomistische Landeroberungserzählung aus der Joschijazeit in Deuteronomium und Josua,” in Das deuteronomistische Geschichtswerk (ed. H.-J. Stipp; ÖBS 39; Frankfurt: Lang, 2011), 89–150, here 129; K. F insterbusch , “Deuteronomy and Joshua: Torah in the Book of Joshua in Light of Deuteronomy,” JAJ 3 (2012), 166–196, here 169. 9  Cf. Deut 27:1, 9: The levitical priests “represent” the divine side, the elders the side of Israel; cf. N. Lohfink, Studien zum Deuteronomium IV (see n. 5), 270–273; Sonnet, Book

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

635

v. 10: Moses commanded both groups to cooperate in reading the Torah in public at the Succoth feast in every year of remission. In my view, it is reasonable to read this command as an explanation of v. 9: the giving of the Torah is a kind of charge giving the Levitical priests and all the elders special responsibility for its public teaching. In any case, the emphasis in the passage is not on the scroll and its deposition.10 What exactly happened to the material Torah scroll remains open. III: YHWH addresses Moses and Joshua in the tent and gives his revelation of the song. There are several problems in Deut 31:14–23 in terms of plot: according to 31:14–15, YHWH wanted to address Joshua in the tent. Unexpectedly, YHWH first addressed Moses, in 31:16–21: YHWH left Moses with no illusions about the future behavior of the people, prophesying that they would break the covenant at a future date. Furthermore, YHWH commanded Moses and Joshua to write and Moses to teach the song. The function of the song is to be a witness against the people (see vv. 19b, 21a). According to 31:22, Moses wrote and taught the song the same day (meaning after the revelation in the tent). This narrative statement interrupts the scene in the tent, because in 31:23 the scene in the tent continues: the verse begins with ‫ויצו את יהושע‬. This clearly refers back to 31:14–15: he, that is YHWH, has now confirmed Joshuaʼs appointment regarding his first future task.11 IV: After he finishes writing the Torah, Moses speaks the song to the elite; Moses and Joshua speak the song to the people. This complex passage (31:24–32:44) requires several comments: (a) Moses, after he finished writing the words of the Torah on a scroll, addressed the Levites and commanded them to take and deposit the scroll (31:24–27). When compared to 31:9, the differences in the narrated details (in within the Book (see n. 5), 140; K. Finsterbusch, Deuteronomium: Eine Einführung (UTB 3626; Göttingen 2012), 159–162. 10  Nota bene: the term ‫ ספר‬is not mentioned. 11  According to Josh 1:2–5, YHWH commanded Joshua to fulfill the first task, which Joshua did in 1:10–12:24. According to Josh 1:6, YHWH confirmed the appointment of Joshua with regard to the second task. In Josh 13:1–7, God commanded Joshua to fulfill the second task, which Joshua did in Josh 14–21. On this “appointment-confirmation system” in Deut–Josh, see N. Lohfink, Studien zum Deuteronomium und zur deuteronomistischen Literatur I (SBAB 8; Stuttgart: Katholisches Bibelwerk, 1990), 83–97; Braulik, “Die deuteronomistische Landeroberungserzählung” (see n. 8), 125–132; Finsterbusch, “Deuteronomy and Joshua” (see n. 8), 168–170. This system was in all likelihood part of an old preexilic story (for a younger perspective on Joshuaʼs appointment, cf. Num 27:23); see Braulik, “Die deuteronomistische Landeroberungserzählung,” 133–139; Finsterbusch, “Deuteronomy and Joshua,” 194.

636

Karin Finsterbusch

MT) point to two different acts of writing:12 according to 31:9 Moses wrote down “this Torah” (‫ )התורה הזאת‬and entrusted it to the Levitical priests and the elders for reading. The emphasis in 31:24–27, however, is focused on the completion of writing “the words of this Torah” (‫ )דברי התורה הזאת‬and on the depositing of the Torah scroll next to the ark by the Levites. The last point shows that the purpose of entrusting the Torah is different. The first point, emphasizing completion of writing the Torah, inevitably leads the reader to the question: Which Torah text was missing beforehand? In context, there is a reasonable answer: it is the song.13 It should be noted that the revelation of the song as a divine Torah supplement in Moab is completely unexpected in the Deuteronomic plot.14 Moses received the divine commandments of the Deuteronomic Torah at Sinai (Deut 5), taught them in Moab, established a covenant, wrote the Torah down, and ensured its public dissemination for future generations. Until Deut 31:9–13, neither Israel within the world of Deuteronomy nor the reader would have guessed that the Torah was incomplete. (b) In his speech to the Levites, Moses commanded them not only to deposit the Torah scroll but also to assemble the elite, i.e., the elders of the tribes and the officials,15 in order to speak to them “these words” (31:28–29).16 As indicated by the reference to the words used by Moses in his speech to the Levites (‫ דבר‬+ 12  Cf. Talstra , “Deuteronomy 31” (see n. 5), 100; S onnet , Book within the Book (see n. 5), 125; Markl, Gottes Volk im Deuteronomium (see n. 5), 139–140. Differently Lohfink, Studien zum Deuteronomium IV (see n. 5), 231–32 (one writing act reported twice). 13  Cf. C. Schäfer-Lichtenberger, Josua und Salomo: Eine Studie zu Autorität und Legi­ ti­mität des Nachfolgers im Alten Testament (VTSup 58; Leiden: Brill, 1995), 182; Talstra, “Deuteronomy 31” (see n. 5), 100; Sonnet, Book within the Book (see n. 5), 158–159; Otto, “Moses Abschiedslied” (see n. 6), 653; Finsterbusch, Deuteronomium (see n. 9), 185–186. Because there is no explicit statement about the relationship between song and Torah in the section, L ohfink , Studien zum Deuteronomium IV (see n. 5), 235–236, considered an alternative interpretation: the song is not part of the Torah and was written on a separate scroll. However, this document would remain as a totes Motiv in the text. Therefore, this interpretation is in my view unlikely. 14  Cf. S onnet , Book within the Book (see n. 5), 148; M arkl , Gottes Volk im Deutero­ nomium (see n. 5), 139. 15  The expression “elders of the tribes” is unique. With regard to the ‫שטרים‬, Deut 31:28 implies they have high status, as in Num 11:16 and Prov 6:7. In other contexts the ‫ שטרים‬are lower assistants in the realm of jurisdiction and military, e.g., in Deut 16:18; cf. J. H. Tigay, Deuteronomy (JPS Torah Commentary; Philadelphia: The Jewish Publication Society of America, 1996), 12; J. C. G ertz , Die Gerichtsorganisation Israels im deuteronomischen Gesetz (FRLANT 165; Göttingen: Vandenhoeck & Ruprecht, 1994), 83–84. 16  The shifts in number in vv. 26–29 do not indicate shifts in the addressees, pace Sonnet, Book within the Book (see n. 5), 172: according to 31:25, Moses undoubtedly spoke all the following words to the Levites. However, the plural forms within his speech refer to the Levites as individuals, who have special tasks to fulfil, whereas the singular forms stress the whole group, i.e., the people, to which the Levites belong or which they represent here; cf. Tigay, Deuteronomy (see n. 15), 297.

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

637

‫ כל‬+ ‫ קהל‬+ ‫)דברים‬, 31:30 is to be interpreted as the executive statement: Moses did in fact teach the elite (‫ )כל קהל ישראל‬the words of the song.17 In 32:44, the bookʼs narrator refers a second time to the teaching of the song. However, this teaching is different from the one narrated in 31:30: not only is the change of scene marked by the use of the introductory verb ‫( בוא‬literally, “he [i.e., Moses] came”), but the teaching of the song is also addressed to a different group, ‫העם‬ (“the people”). Furthermore, the teaching staff is different: this time it is not Moses alone but Moses and Joshua.18 (c) I want to refer briefly to the complex content of the song as well. The speaker is Moses, in part quoting the words of YHWH. In the first part of the song (32:1–33),19 all emphasis is on Israel turning to foreign gods in the future and on YHWHʼs anger. The second part (32:34–43) describes YHWHʼs unexpected and merciful turning back to his people,20 punishing Israelʼs enemies for their boasting.21 (d) An interesting inconsistency remains concerning teaching and writing the song in MT Deut 31:1–32:47. According to 31:19, YHWH commanded Moses and Joshua to write and Moses to teach the song. According to 31:22, Moses alone wrote the song and taught it. According to 31:30, Moses alone taught the song to the elite. According to 32:44, Moses and Joshua together taught the song to the people. In double contradiction to YHWHʼs command, then, Joshua had no part in writing the song (or better: in writing the Torah) but only in its teaching.22 V: Moses finishes speaking “all these words” and admonishes Israel for a final time. In 32:45, Moses is named again as the sole speaker. He now finishes speaking “all these words” (‫)כל הדברים האלה‬. “These words” are not only the words of the song (31:30), the words of the Torah (31:24), or Mosesʼs words to Israel in 17  The expression ‫( כל קהל ישראל‬13× in the Hebrew Bible) can refer to all Israel or to only a part of Israel. In MT Josh 8:35 and 1 Chr 13:1, the expression undoubtedly refers to only a part of the people. In MT Deut 31:30 (in contrast to the Hebrew Vorlage of LXX Deut 31:30, see below), the expression most likely refers only to the elite; cf. Lohfink, Studien zum Deuteronomium IV (see n. 5), 230, 240–244; M arkl , Gottes Volk im Deuteronomium (see n. 5), 217, pace Sonnet, Book within the Book (see n. 5), 171–173. 18  Thus in MT Deut 31–32 the song is taught twice; cf. Lohfink, Studien zum Deuteronomium IV (see n. 5), 230–231. 19  On the structure of the song, see M arkl , Gottes Volk im Deuteronomium (see n. 5), 232–238. 20  See especially Deut 32:36. 21  See especially Deut 32:27. 22  Do these inconsistencies reflect different positions of late scribes/redactors with regard to the authority of “Moses” and “Joshua” (representing “Pentateuch” and “Hexateuch,” respectively)?

638

Karin Finsterbusch

31:1–6. The reference to the beginning of the book, indicated by the exclusive combination of several expressions,23 can hardly be overlooked:24 1:1a: These are the words that Moses spoke to all Israel. 32:45: And Moses finished speaking all25 these words to all Israel.

This reference indicates that in the world of Deuteronomy, Moses has now fulfilled his main task. The bookʼs narrator lets Moses conclude with a final admonition, focusing on the main point: Israel must keep “all the words of this Torah” in order to live (32:46–47). As we have seen, the rough structure of MT Deut 31:1–32:47 is fairly clear.26 However, brief examination of the narrative logic of the individual passages reveals that the song texts are not always well integrated into the plot: 31:16–21 and 22 interrupt the connection between 31:14–15 and 23. Furthermore, the revelation of the song causes some “plot-trouble”: Moses has to deal with the Torah a second time, and he is forced to add the song as a supplemental text on the scroll. This synchronic evidence points to the following diachronic solution: redactors inserted the song and the related narrative texts. The song texts form a fairly coherent entity: 31:16–21

YHWHʼs commandment to write and to teach the song.

31:22

Narrative (proleptic)27 statement: Moses wrote and taught the song “on that day.”

31:24–32:44

Detailed explanation of v. 22: the words of the song and its writing and teaching.

This coherence points to rather concentrated redactional activity.28 MT Deut 31:1–32:47 without the song texts must have then been reasonably compre23  As already observed by L. Perlitt, Deuteronomium, 1. Teilband: Deuteronomium 1–6* (BKAT 5/1; Neukirchen-Vluyn: Neukirchener Verlag, 2013), 7, 9. 24  G. Braulik, “ ‘Die Worte’ (haddebārîm) in Deuteronomium 1–11,” in “Gerechtigkeit und Recht zu üben” (Gen 18,19): Studien zur altorientalischen und biblischen Rechtsgeschichte, zur Religionsgeschichte Israels und zur Religionssoziologie; Festschrift für Eckart Otto zum 65. Geburtstag (ed. R. Achenbach and M. Arneth; BZABR 13; Wiesbaden: Harrassowitz, 2010), 200–216, here 202–203, is certainly right that Deut 1:1a is in particular a superscription of Mosesʼs first speech (1:6–4:40). In addition, Deut 1:1a has in my view a further function in the MT, framing the main speeches of Moses together with 32:45; cf. F insterbusch , Deuteronomium (see n. 9), 50–51. 25  Sam lacks ‫כל‬. ‫ כל‬was probably inserted by late protomasoretic redactors in order to underline that Moses has now definitively fulfilled his main task in Moab. 26  The narrative order and the order of the events in MT Deut 31:1–32:47 are, all in all, in accordance; cf. Sonnet, Book within the Book (see n. 5), 180–182, and Markl, Gottes Volk im Deuteronomium (see n. 5), 138–140, pace Lohfink, Studien zum Deuteronomium IV (see n. 5). 27  Cf. Sonnet, Book within the Book (see n. 5), 151. 28  Cf. Otto, “Moses Abschiedslied” (see n. 6), 652–656.

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

639

hensible. The cross-check shows that this is indeed the case: Mosesʼs address to Israel about the inheritance of the land, his appointment of Joshua and his twofold future task, the writing of the Torah, the assurance of its transmission to following generations, YHWHʼs confirmation of Joshua as successor of Moses with regard to his first future task, and some final words of Moses together form a rather satisfying plot.29 The question is what the rationale of the song-redactors (i.e., the redactors who inserted the song and the related narrative texts) was and when they acted. Most scholars assume the existence of an independent edition of Deuteronomy. This independent edition was in my view composed in the early exilic period (though the authors used and referred to preexilic sources like the Covenant Code,30 the Vassal Treaties of Esarhaddon, and other Neo-Assyrian treaty material). The exilic edition in all likelihood contained most of the material in Deut 1:1–31:13; 32:45–47, plus a narrative about the death of Moses.31 The aim of the authors was first and foremost to strengthen the identity of the exilic community, outlining the ideal Israelite society: if Israel acted according to the Deuteronomic commandments, catastrophes like 597 and 586 BCE would never happen again. In order to understand the rationale of the song-redactors, the threefold emphasis ‫“( ביום ההוא‬on that day”) in YHWHʼs speech in the tent (31:17 [2×], 18) is especially revealing. “That day” on which Israel will turn to foreign gods and break the covenant certainly refers to a specific time, namely, the beginning of the sixth century BCE: according to the song-redactors, YHWH knew that the events of 586 BCE would happen. God and his prophet Moses are not naïve. Most important, however, the song gives the reason why God let Israel survive the catastrophe: because of his mercy and his devotion to his people. This point remained open in the early exilic edition Deuteronomy.32 The song-redactors, 29  Some aspects of content and language underline the secondary character of the song texts. Thus, the idea of Mosesʼs teaching the elite is a rather “un-Deuteronomic” idea: in Deut 1:1–31:13, Moses constantly addressed the entire people. Furthermore, in comparison to other Deuteronomic texts, which call on Israel to choose between blessing and curse (e.g., Deut 4:25–31; 28:1–68), the song texts insert an unusually pessimistic prospect on Israelʼs behavior in the land: the Israelites will definitively break the covenant in the future (cf. 31:17). Finally, the song texts display some “un-Deuteronomic” expressions such as ‫( אל נכר‬cf. 31:18, 20; 32:12) or ‫( עד ב׳‬cf. 31:19, 26); and see Sonnet, Book within the Book (see n. 5), 174–175. 30  See the contribution of D. P. Wright, “The Covenant Code Appendix (Exodus 23:20–33), Neo-Assyrian Sources, and Implications for Pentateuchal Study,” in this volume, pp. 47–85. 31  Cf. Finsterbusch, Deuteronomium (see n. 9), 35–38. 32  For example, in Deut 4:25–31 Moses predicts destruction and exile if Israel does not keep the commandments. Moses also predicts that the surviving community in the exile can return to YHWH only because YHWH is faithful to the covenant that he made with the patriarchs. The question, however, why YHWH would not destroy the entire unfaithful people but let a remnant survive, is not answered. Only in Deut 30:3, part of the larger unit Deut 30:1–10, which is in all likelihood an early postexilic “Deuteronomistic” addition (cf. W. G ross ,

640

Karin Finsterbusch

in all likelihood, wanted to underline the message of the exilic edition:33 the addressees, who live only by Godʼs mercy after “that day,” should by all means behave differently than Israel did before “that day” and should learn and keep the commandments of the Deuteronomic Torah in the future. In other words, to explain the evidence of the song texts in MT Deut 31:1– 32:47, there is no need to assume the Pentateuchʼs larger narrative horizon.34 In contrast, however, the evidence of the Hebrew Vorlage of LXX Deut 31:1–32:47 cannot be explained without the larger pentateuchal horizon.

2 (The Hebrew Vorlage of) LXX Deuteronomy 31:1–32:47 I first want to present an overview of the main scenes and themes in LXX Deut 31:1–32:47 and its Hebrew Vorlage, respectively. In the table below, significant additions in comparison to the MT are set in italics, and important differences in terms of content are underlined. I: 31:1

Moses addresses Israel and Joshua about the inheritance of the land Narrative statement: Moses finishes speaking “all these words” to all children of Israel.

Zukunft für Israel: Alttestamentliche Bundeskonzepte und die aktuelle Debatte um den Neuen Bund [SBS 176; Stuttgart: Katholisches Bibelwerk], 40–44), is Godʼs mercy mentioned as the reason why he let his exiled people return to the land. 33  The “monotheistic” statement in Deut 32:39 is in my view a decisive argument against the preexilic origin of the song. The “polytheistic” statements in 32:8* and 32:43* (see on those verses especially A. Rofé , “The End of the Song of Moses [Deuteronomy 32:43],” in Liebe und Gebot: Studien zum Deuteronomium [ed. R. G. Kratz and H. Spieckermann; FRLANT 190; Göttingen: Vandenhoeck, 2000], 164–172, here 167) provide an argument against a dating later than the late Persian period. In terms of method it does not seem possible to date the song on the basis of allusions, as the following positions may show: according to O tto , “Moses Abschiedslied” (see n. 6), 672, the authors alluded to the “Texte der drei Kanonsteile”; however, according to G. Fischer, “Das Ende von Deuteronomium (Dtn 26–34) im Spiegel des Jeremiabuches,” in “Gerechtigkeit und Recht zu üben” (Gen 18,19): Studien zur altorientalischen und biblischen Rechtsgeschichte, zur Religionsgeschichte Israels und zur Religionssoziologie; Festschrift für Eckart Otto zum 65. Geburtstag (ed. R. Achenbach and M. Arneth; BZABR 13; Wiesbaden: Harrassowitz, 2010), 281–292, here 290–92, the author of Jeremiah alluded to the song. 34  Pace Otto, “Moses Abschiedslied” (see n. 6), 653–54: “Mit der Einfügung des Moseliedes und seines Rahmens sollte an die Pentateuchredaktion in Dtn 4,25–31 und Dtn 34,10–12 anknüpfend Mose als Erzprophet, der letzte, mit dem YHWH von Angesicht zu Angesicht wie in Dtn 31,16–22 gesprochen habe, zur Geltung gebracht werden.” However, MT Deut 31 stands against this interpretation. First, the last person God speaks to in the tent is Joshua, not Moses (cf. 31:23). Second, according to 31:14–15, God speaks in the tent to Moses and Joshua “in a pillar of cloud” and not “face to face.”

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

31:2–6

Moses addresses them, speaking about their inheritance of the land.

31:7–8

Moses addresses Joshua in front of Israel, commanding him about his twofold future task.

641

II: Moses writes the Torah, and he commands, on that day, regular public reading of the written Torah 31:9

Narrative statement: Moses writes down “the words of this Torah” on a scroll, and he gives (it) to the Levitical priests and the elders.

31:10–13

Moses addresses them on that day, commanding them to conduct a public Torah reading regularly.

III: YHWH addresses Moses and Joshua in the tent and gives his revelation of the song 31:14a

YHWH addresses Moses, commanding him to appear together with Joshua in the tent for the purpose of addressing Joshua.

31:14b–15

Narrative statement: Moses obeys; YHWH appears in the tent in a pillar of cloud.

31:16–21

YHWH addresses Moses in the tent, speaking about the future behavior of the people, commanding Moses and Joshua to write and Moses and Joshua to teach “the words of this song.”

31:22

Narrative statement: Moses writes “this song” on that day and teaches it to the children of Israel.

31:23

He (i.e., Moses) addresses Joshua, commanding him about his first future task.

IV: After he finishes writing the Torah, Moses speaks the Torah (including the song) to the elite 31:24–29

After Moses finishes writing “all the words of this Torah” on a scroll, he addresses the Levites, commanding them to deposit the Torah and to assemble the elite (i.e., the tribal leaders and the elders and the judges and the officials), so that he may speak to them “all these words.” (i.e., the words of the Torah, including the song).

V: Moses speaks the song to the people 31:30

Narrative statement: Moses speaks “the words of this song” to the “whole assembly of Israel” (i.e., the people).

32:1–43

The words of the song.

32:44(a)

Narrative statement: Moses writes “this song” on that day and teaches it to the children of Israel (= 31:22).

642

Karin Finsterbusch

VI: Moses and Joshua speak the Torah to the people 32:44(b)

Narrative statement: Moses comes and speaks “all the words of this Torah” to the people, “he and Joshua.”

VII: Moses finishes speaking and admonishes Israel for a final time 32:45

Narrative statement: Moses finishes speaking to all Israel.

32:46–47

Moses addresses them, admonishing them for a final time to keep the Torah.

I will now briefly comment on select aspects of the individual passages and will point in particular to the main differences from the MT. I: Moses addresses Israel and Joshua about the inheritance of the land. Deuteronomy 31:1 opens the final part of the book (Deut 31–34): Moses finished speaking “all these words.” In contrast to the MT of the verse, the expression “all these words” in LXX refers back, within Deuteronomy,35 at least to (the Hebrew Vorlage of) LXX Deut 29:1 (“these are the words of the covenant”) and to 1:1 (“these are the words that Moses spoke to all Israel on the other side of the Jordan”). Deuteronomy 31:1 is the point in the story at which Moses definitively delivered his main message to the people. II: Moses writes the Torah, and he commands, on that day, regular public reading of the written Torah. According to the first sentence in 31:9, Moses wrote “the words of this Torah” on a scroll (assumed Hebrew Vorlage, ‫)על ספר‬.36 Strikingly, the following sentence does not contain the direct object: the text does not report what exactly Moses gave to the Levitical priests and the elders (it could be the “scroll” or the “Torah”). As we will see below, there is a reason why the object is left out. The narrated acts in v. 9 are rather isolated in context. In contrast to the MT, 31:10–13 is chronologically unconnected to v. 9: Moses “on that day” (assumed Hebrew Vorlage, ‫ )ביום ההוא‬commanded the Levitical priests and the elders. This means that Moses commanded them the same day (on which he wrote down the Torah), but that could be significantly later than his giving the Torah/ the scroll to the Levitical priests and the elders.

 Within the Pentateuch, “these words” can be understood as all words that Moses had to deliver to “the children of Israel” on Godʼs behalf. 36  Cf. 4QDeuth, frag. 10; see Ulrich, Biblical Qumran Scrolls (see n. 3), 237. 35

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

643

III: YHWH addresses Moses and Joshua in the tent and gives his revelation of the song. After the tent scene (31:14–21), Moses wrote and taught the song “on that day” (v. 22) and addressed Joshua (v. 23). In terms of plot, 31:14–23 is puzzling: according to 31:14, YHWH wanted to address Joshua in the tent. However, an executive statement of such an address is missing. Rather, in 31:23 Moses addressed Joshua about his first future task outside the tent (and not YHWH inside the tent, as in MT). This is especially odd because the narrator repeats words that Moses had already spoken to Joshua (see v. 7). IV: After he finishes writing the Torah, Moses speaks the Torah (including the song) to the elite. The fourth passage, 31:24–29, requires more detailed reflection. (a) Deuteronomy 31:24 contains the second reference to Mosesʼs writing the Torah in our section. In contrast to the MT, vv. 9 and 24 most likely refer to one and the same act. A strong indication is the use of common language: the expression “words of the Torah” and the key motif of the scroll are used in both verses, underlining the identity of the reported act of Moses. Thus, the narrator refers to Mosesʼs writing of the Torah for the first time before the tent scene (v. 9) and for the second time after it (v. 24). This leads to the question, when did the Torah writing in fact happen? Since Deut 31:24–29 contains clear references to YHWHʼs words in the tent (taking the view that Israel in the land will by all means do “evil” and will be punished with “evil” by the angry God, and using the specific expression ‫עד ב׳‬, “witness against”), the answer must be that Moses wrote the whole Torah from the beginning to its very end after the tent scene. (b) Whereas 31:9 should be considered a kind of general proleptic statement that Moses wrote the Torah on a scroll, 31:24–29 reports the details. This report concentrates on the very moment when Moses finished writing the Torah. (c) The Torah most likely includes the song. In v. 24, emphasis is put on “all”: Moses wrote down “all the words of this Torah.” This “all” (assumed Hebrew Vorlage, ‫)כל‬, which is not present in the MT, points to an inclusive meaning for tôrâ here. Furthermore, only such an inclusive sense can illustrate the meaning of the statement in v. 22 that Moses wrote the song, which would otherwise remain as a totes Motiv in the section: when Moses wrote the whole Torah, the very last text he wrote was the song. (d) Immediately after Moses finished writing, he commanded the Levites to deposit the Torah scroll next to the ark and to assemble four groups of the elite of Israel (the tribal leaders, the elders, the judges, and the officials)37 in  Moses did not want to assemble the entire people. For example, women, children, old people, and gerim are missing from the list; cf. Deut 29:9–10 and 31:12, pace L ohfink , Studien zum Deuteronomium IV (see n. 5), 224: “In 31,28 LXX läßt Mose die Leviten nicht nur 37

644

Karin Finsterbusch

order to teach them “all these words” (v. 28). What is the meaning of “all these words”? Since the expression clearly echoes “all the words of this Torah” in v. 24, these words, in contrast to the MT, cannot be limited to the song. Thus, we can conclude that Moses intended to teach the elite “all the words of this Torah.” There is no executive statement. In 31:30, a new scene begins (see below). (e) In light of 31:24–29, the reason for the missing object (Torah or scroll) in the second sentence in v. 9 becomes clear: the material Torah scroll is given (to be deposited in the ark) only to those Levites who carry the ark (cf. 31:24). Responsibility concerning the written Torah text, namely, its reading in public at the Succoth feast in Jerusalem in every year of remission, is given to the elders in cooperation with the Levites. In other words, the missing object in 31:9aβ indicates that Moses “gave” the one and only Torah to the groups mentioned, but in different ways. V: Moses speaks the song to the people. In 31:30–32:44(a), the theme changes, and the focus shifts to Mosesʼs teaching the song.38 The narrator has already reported in a general proleptic statement in 31:22 that this act happened “on that day,” but now he relates the details (it is especially noteworthy that the narrator in the nonmasoretic version uses the same narrative pattern as he did for Mosesʼs Torah writing: first the proleptic statement, then a more detailed report later in the text). According to 31:30, the addressees of Mosesʼs teaching are “the whole assembly of Israel” (assumed Hebrew Vorlage, ‫)כל קהל ישראל‬. In light of the last sentence in the passage, 32:44(a), the expression means the whole people, not only the elite, in contrast to the MT.39 This last sentence, which is lacking in the MT, is an exact repetition of 31:22. In terms of plot, it makes no sense to understand 32:44(a) as a separate act, i.e., as a second writing and a second teaching of the song. Why should Moses write the song twice? Thus, the function of the statement 32:44(a) should be seen as a kind of affirmation: both acts, Mosesʼs writing the song and his teaching the people, are now indeed completed in the narrated world.

die Stammesältesten und Listenführer einberufen, sondern alle jene Amtsträger, die 29,9 LXX aufzählt. Dort werden zwar noch andere Gruppen hinzugefügt [. . .], doch selbst ohne diese Zusatzliste weckt die erweiterte Amtsträgerliste in 31,28 beim Leser schon die Vermutung, daß wie in 29,1 auch hier jetzt die Vollversammlung aller Israeliten einberufen werden soll.” It may be noteworthy that in Codex B (see n. 2) the elders are missing from 31:28. 38  Interestingly, the Greek translator rendered the ‫ דברים‬of the song in 31:30 with ῥήματα (cf. LXX Deut 31:21), whereas the ‫ דברים‬of the Torah in 31:24, 28 were rendered with λόγους. It seems likely that the translation reflects the different meaning of the “words” in the passages Deut 31:24–29 and 31:30–32:44(a). 39  Cf. Lohfink, Studien zum Deuteronomium IV (see n. 5), 224.

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

645

VI: Moses and Joshua speak the Torah to the people. In 32:44(b) the scene changes, as is indicated by the use of the introductory verb ‫ בוא‬in the assumed Hebrew Vorlage. The change of scene goes together with the change of theme: the people (including the elite40) must now learn “all the words of this Torah.” Interestingly, the teachers are Moses and Joshua. The Torah teaching of Joshua comes as a surprise. However, according to 31:19 in the version represented by the LXX, YHWH commanded Moses and Joshua together in the tent to write and to teach the song.41 In light of this divine command, 32:44(b) could be interpreted in the sense that Joshua shared Mosesʼs Torah teaching with respect to its last part, the song. One inconsistency concerning Joshua, however, remains in the nonmasoretic version of 31:1–32:47, too: in contradiction to YHWHʼs command in the tent, Joshua had no part in writing the Song. The writing of the Torah texts is Mosesʼs exclusive task. VII: Moses finishes speaking and admonishes Israel for a final time. In 32:45, the narrator repeats the first three words of the section, i.e., Deut 31:1: “and Moses finished speaking” (assumed Hebrew Vorlage, ‫)ויכל משה לדבר‬.42 Moses has now definitively fulfilled his task of transmission. This analysis has shown that the narrative order and the order of events in LXX Deut 31:1–32:47 and its Hebrew Vorlage, respectively, in contrast to the MT, are not in accordance at least in part.43 The following table presents an overview of the order of the events “on that day” in the narrated world of the text: Moses finishes speaking “all these words.” Moses addresses Israel and addresses Joshua about his twofold future task. YHWH reveals the song in the tent, commanding Moses and Joshua to write and to teach it. Moses writes down the Torah (including the song), commanding the Levites to deposit the scroll and to assemble the elite. Moses teaches the elite the Torah (including the song). Moses teaches the people the song. Moses teaches the people the Torah (including the song), together with Joshua.

 Interestingly, according to the Hebrew Vorlage of LXX Deut 31–32, the elite must learn the Torah twice. Like the king, who should according to Deut 17:19 read from his personal copy of the Torah daily, the elite must know the Torah better than the people. 41  Cf. 4QDeutc, frags. 54–55; see Ulrich, Biblical Qumran Scrolls (see n. 3), 238. 42  Cf. 1QDeutb, frag. 13 ii; see Ulrich, Biblical Qumran Scrolls (see n. 3), 236. 43  This was Lohfinkʼs idea concerning MT Deut 31–32, see above notes 12 and 26. 40

646

Karin Finsterbusch

(?) The same day: Moses commands the Levitical priests and the elders to conduct regular public Torah reading. (?) Moses addresses Joshua about his first future task. Moses finishes speaking. In terms of content, structure, and narrative dynamic, the nonmasoretic version of Deut 31:1–32:47 is extremely complex. Some inconsistencies and oddities remain. The song, however, is rather well integrated into the plot: in contrast to the MT, the song is not a supplement to an already-written text. After the tent scene, Moses writes down the Torah all in one go from the beginning to its very end; the song is the final Torah text on the scroll. This, however, reveals where the real focus of the plot is: it is not on the song but on the Torah. The main difference between the two versions of Deut 31–32 is undoubtedly that according to the nonmasoretic version, Israel in Moab must learn the Torah (32:44[b]). Within the narrated world of the independent edition of Deuteronomy, such comprehensive Torah learning at the very end of the narrated period (i.e., of the one day, at the end of which Moses dies) would not make any sense: Moses has already taught and Israel has already learned the Deuteronomic commandments (cf. Deut 5:1, 31; 6–26).44 Therefore, Torah in the nonmasoretic version of Deut 31–32 is not the Deuteronomic Torah and the redactors are not “Deuteronomistic” redactors. Rather, the motif of Torah teaching and learning must have been integrated by the pentateuchal redactors, who compiled and redacted separate sources into a single literary composition, the (Proto-)Pentateuch, making the independent edition of Deuteronomy “the fifth book of Moses.” What could have been their rationale for reshaping Deut 31–32?45 Torah was undoubtedly an important topic for the Pentateuchʼs redactors. In all likelihood they inserted Exod 24:12 (the only verse in Gen – Num in which the term ‫ תורה‬refers to a comprehensive collection of laws):46 God gave 44  Cf. K. Finsterbusch, Weisung für Israel: Studien zu religiösem Lehren und Lernen im Deuteronomium und in seinem Umfeld (FAT 44; Tübingen: Mohr Siebeck, 2005), 306–314. 45  Compared to MT Deut 31:1–32:47, the narrative structure of the Hebrew Vorlage of LXX Deut 31–32 should be considered the younger one; cf. Lohfink, Studien zum Deuteronomium IV (see n. 5), 222. However, some single variants of the Hebrew Vorlage of LXX Deut 31–32 (which do not affect the plot structure) are, in comparison to the MT, in all likelihood older; see, e.g., 31:5 (assumed Hebrew Vorlage of the LXX: ‫ ;כאשר‬MT/Sam: ‫)ככל המצוה אשר‬, and cf. N wachukwu , “Textual Differences” (see n. 4). Both versions reflect ongoing scribal activity in the Second Temple period. The song may have had a textual history of its own as significant variants indicate; cf., e.g., M. B ogaert , “Les trois rédactions conservées et la forme originale de lʼenvoi du cantique de Moïse (Dt 32,43),” in Das Deuteronomium: Entstehung, Gestalt, Botschaft (ed. N. Lohfink; BETL 68; Leuven: Leuven University Press, 1985), 329–340; Rofé, “End of the Song of Moses” (see n. 33). 46  Cf. R. A chenbach , “Die Tora und die Propheten im 5. und 4. Jh. v.Chr.,” in Tora in der Hebräischen Bibel: Studien zur Redaktionsgeschichte und synchronen Logik diachroner

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

647

the Torah to Moses on Mount Sinai. Based on certain key terms like ‫ ירה‬and ‫תורה‬, the Pentateuchʼs readership would understand this Sinaitic Torah as referring at least to Exod 25:1–31:17; the laws of Leviticus; Deut 5–26; the blessings and curses in Deut 28; and the song in Deut 32. In other words, in the world of the Pentateuch, Moses reveals the Sinaitic Torah in process. The problem is, the Sinaitic generation died in the desert; Israel in Moab is the second generation, who never heard the Sinaitic Torah in one piece. However, before starting life in the promised land, Israel must of course know the whole Torah.47 This was most likely the reason, first, for optimizing the timetable of the Torahʼs writing (Moses wrote the whole Torah only after the last Torah text – the song – was revealed) and, second, for inserting a subsequent final, comprehensive, and intensive Torah teaching (to the elite and the people) in Deut 31–32.48

3 Concluding Remarks The findings may be concluded in three parts: 1. The narrative structure of Deut 31–32 was changed fundamentally in the course of the textʼs transmission at least twice. Both changes resulted from secondary insertions: the insertion of the song and related narratives into the independent edition of Deuteronomy and the insertion of the independent edition of Deuteronomy into the Pentateuch. What is generally remarkable here is the Transformationen (ed. R. Achenbach et al.; BZABR 7; Wiesbaden: Harrassowitz, 2008), 26–71, here 40; G. Braulik , “ ‘Die Weisung und das Gebot’ im Enneateuch,” in Studien zu den Methoden der Deuteronomiumsexegese (SBAB 42; Stuttgart: Katholisches Bibelwerk, 2006), 111–135, here 131; K. F insterbusch , “Aufsummierte Tora: Zur Bedeutung von ‫ תורה‬als Bezeichnung für eine Gesetzessammlung im Pentateuch,” JAJ 2 (2011), 1–28, here 19–20. 47  The pentateuchal redactors, who reshaped Deut 31–32, most likely did not yet understand Torah as “Pentateuch”: Moses finished speaking the words of the Torah (see 32:44, 45), but this is not the end of the book or of the Pentateuch. 48  Based on the analysis of the Hebrew Vorlage of LXX/MT Deut 31–32 as presented in this study, much can be learned about the ways in which the pentateuchal redactors reworked the texts. The following lists only a few important points: (a) insertion of a chronological marker in order to disconnect acts (insertion of ‫ ביום ההוא‬in 31:10; thus, v. 9 can be read as an independent proleptic statement); (b) adjustment of language in order to create references between verses (use of the expression “words of the Torah” and the motif of the scroll in 31:9 and 31:24, thus indicating that Mosesʼs acts of writing are one and the same; insertion of ‫כל‬ in 31:24 and 31:29, thus indicating that “all the words” in both verses are identical, i.e., all the words of the Torah); (c) repetition of a textual unit in order to clarify the plot (repetition of 31:22 after the song = 32:44[a], thus clarifying the identity of Mosesʼs audience in 31:30, i.e., the people of Israel); (d) replacement of a word/theme in order to change the plot (change from “song” to “Torah” in 32:44[b], thus creating a new scene).

648

Karin Finsterbusch

freedom of the redactors to reshape the plot. In particular, the version represented by the LXX indicates that the pentateuchal redactors thought carefully about the narrative logic of the artificial literary composition of the (Proto-)Pentateuch. It is evident how much questions of plot mattered to them. 2. The interesting point is, of course, why the (Proto-)Masoretic Text of Deut 31–32 became canonical and not the other version. After all, the nonmasoretic version contains a narrative about Mosesʼs final Torah teaching! I can imagine two reasons. First, the plot structure of the nonmasoretic version was too complicated for the readership to grasp. Second, the plot was chronologically too unrealistic: Moses could not possibly teach the Deuteronomic law, establish a covenant, and teach the song and the whole Torah twice (to the elite and to the people) on the very last day of his life. 3. Deuteronomy 31–32 belongs to the very last part of the Pentateuch. This final part undoubtedly has special importance (as has every final section of an authoritative literary work). The different versions of Deut 31–32 are an exemplary demonstration of the fact that in the Hellenistic period (the time in which the LXX translators acted) the final shape of the Pentateuch still did not exist (otherwise they would of course have had the [Proto-]Masoretic version of Deut 31–32 as Vorlage). In light of this evidence, the work of the Jewish groups who were responsible for the standardization and the canonization of the Pentateuch in Judah can hardly be overestimated:49 they had a huge task, choosing from several possibilities and deciding what shape the Pentateuch of the Jewish community should finally take.

49  An idea of how textual standardization toward (Proto-)MT in Judah was achieved might be preserved in a baraita in the Jerusalem Talmud, Taʼanit 4:68a: “they confirmed the reading found in the two (scrolls) and abrogated the other [. . .].” In Ant. 3:38; 5:61; and 10:57, Josephus mentions that books/holy books of Moses were laid in the Jerusalem temple. These remarks could refer to master copies in the temple. With regard to the overwhelming evidence for textual plurality and the lack of any evidence for textual standardization in Judah up to the second half of the first century BCE, it seems highly unlikely that master copies of any text existed in the Jerusalem temple significantly earlier that the first century CE; cf. A. Lange , “ ‘Nobody Dared to Add to Them, to Take from Them, or to Make Changesʼ (Josephus, Ag. Ap. 1.42): The Textual Standardization of Jewish Scriptures in Light of the Dead Sea Scrolls,” in Flores Florentino: Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez (ed. A. Hilhorst et al., JSJSup 122, Leiden: Brill, 2007), 105–126, here 116–17. In ancient Judaism, the earliest evidence for textual standardization is the Letter of Aristeas. This locates the beginnings of the textual standardization of Jewish Scriptures in Egyptian if not Alexandrian Judaism. Hence, it seems feasible that the beginnings of Jewish textual standardization go back to Greek influence; cf. Lange, “Nobody Dared,” 118–122.

Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative

649

Appendix: Translation of key verses of MT-Deuteronomy 31–32 and LXX-Deuteronomy 31–32 Significant additions are set in italics; important differences in terms of content are underlined. Deuteronomy 31–32 MT

Deuteronomy 31–32 LXX

31:1 And Moses went and spoke these words to all Israel. 2 And he said to them: . . .

And Moses finished speaking all these words to all children of Israel. And he said to them: . . .

9 And Moses wrote this torah and gave it to the priests, the sons of Levi, who carried the ark of the covenant of YHWH, and to all the elders of Israel. 10 And Moses commanded them, saying: . . .

And Moses wrote the words of this law on a scroll and gave to the priests, the sons of Levi, who carried the ark of the covenant of the Lord, and to all the elders of the children of Israel. And Moses commanded them on that day, saying: . . .

14 And YHWH said to Moses: Behold, your days to die approach. Call Joshua and position yourselves in the tent of meeting, and I will command him. And Moses went and Joshua, – and they positioned themselves in the tent of meeting . . .

And the Lord said to Moses: Behold, the days of your death approach. Call Yesous and stand by the entrance of the tent of testimony, and I will command him. And Moses went and Yesous into the tent of testimony, and they stood by the entrance of the tent of testimony . . .

19 And now, write for you (pl.) this song and teach (sg.) it to the children of Israel, put (sg.) it in their mouths . . . 22 And Moses wrote this song on that day and taught it to the children of Israel. 23 And he (i.e., YHWH) commanded Joshua the son of Nun and said: Be strong and courageous, for you shall bring the children of Israel into the land which I swore unto them, and I, I will be with you.

And now, write (pl.) the words of this song and teach (pl.) it to the children of Israel, and put (pl.) it in their mouths . . . And Moses wrote this song on that day and taught it to the children of Israel. And he (i.e., Moses) commanded Yesous and said: Be strong and courageous, for you shall bring the children of Israel into the land which the Lord swore unto them, and he, he will be with you.

650

Karin Finsterbusch

24 And it happened when Moses had finished writing the words of this torah on a scroll to the very end, 25 then Moses commanded the Levites, who carried the ark of the covenant of YHWH, saying: 26 Take this scroll of the torah and put it beside the ark . . . 28 Assemble to me all elders of your tribes and your officials, and I will speak in their ears these words (i.e., of the song). . . .

And when Moses had finished writing all the words of this law on a scroll to the very end, then he commanded the Levites, who carried the ark of the covenant of the Lord, saying: Having taken the scroll of this law, put it beside the ark . . . Assemble to me your tribal leaders and your elders and your judges and your officials, so that I may speak in their ears all these words (i.e., of the law). . . .

30 And Moses spoke in the ears of the whole assembly of Israel the words of this song to the very end. . . .

And Moses spoke in the ears of the whole assembly of Israel the words of this song to the very end. . . .

32:44(a) – – 44(b) And Moses came and spoke all the words of this song in the ears of the people, he and Hoshea the son of Nun.

And Moses wrote this song on that day and taught it to the children of Israel. And Moses came and spoke all the words of this law in the ears of the people, he and Yesous the son of Naue.

45 And Moses finished speaking all these words to all Israel. 46 And he said to them: . . .

And Moses finished speaking to all Israel. And he said to them: . . .

Source Dependence and the Development of the Pentateuch The Case of Leviticus 24 David P. Wright Determining textual relationships is an indispensable tool for reconstructing the Pentateuchʼs historical development. Scholarship has long examined the pedigree of texts, particularly legal texts, to develop a framework for ascertaining the date and relationship not only of the law corpora but also of associable narrative materials. One of the problems in this undertaking is ascertaining if, in fact, one text has interacted with another. Close textual similarities are the best evidence that texts are genetically related, as found for example between the debt-slave legislation of Deut 15:12–18 and Exod 21:2–11. Various studies have been devoted to determining what constitute such close similarities and have set out criteria that may be considered probative for determining the relationship between texts. These seek to provide a statistical gauge to assure or support a conclusion of textual dependence.1 Such criteria must certainly guide analyses, but perspectives and considerations from the area of innerbiblical exegesis complicate the straightforward application of rules.2 These have shown that a target text may quite creatively transform a source. This feature, in fact, may be more the rule than the exception, given that the creation of new texts is driven by ideological goals that seek to advance new perspectives. Nevertheless, the differences arising from textual transformation, combined with similarities that otherwise point to a genetic relationship, may actually aid in the determination of textual dependence when cogent and compelling explanations can be provided for the hermeneutical innovations.

1  See the discussion and literature cited in D. P. Wright, Inventing Godʼs Law: How the Covenant Code of the Bible Used and Revised the Laws of Hammurabi (Oxford: Oxford University Press, 2009), 24–28, 375, n. 115; W. S. Morrow, “Legal Interactions: The mišpāṭîm and the Laws of Hammurabi,” BO 70 (2013), 309–331. 2  See, for example, B. M. L evinson , Legal Revision and Religious Renewal in Ancient Israel (Cambridge: Cambridge University Press, 2008), especially the descriptive bibliography on 95–181.

652

David P. Wright

Another complexity in comparative analysis is the problem of the direction of textual dependence. In many cases other evidence exists for determining dependence, such that this is not a difficulty. To refer again to the example of the laws on debt-slavery, most scholars think, for reasons other than the nexus in debt-slavery legislation, that Deuteronomy (D) is later than the Covenant Code (CC). But in questionable cases, the compositional logic involved in the hermeneutical transformation of texts can again help resolve questions of directionality. Including the possible effects of hermeneutical innovation in an analysis of textual dependence adds a nonstatistical and, indeed, an interpretive element to the assessment of textual relationships. This opens the door for the criticism that a particular analysis has gone too far. At the same time, there is always the risk of not going far enough and ending with an incomplete or skewed reconstruction of textual history due to a limiting methodology or stance. Considering the creativity of hermeneutical innovation may actually put one in a better position to envision a more compelling reconstruction and make sense of a broader set of data. In this framework, this essay explores the details of how hermeneutical transformation plays a role in the composition of the legal novella about blasphemy and talion from the Holiness school (H) in Lev 24:10–23, which is part of the larger Priestly-Holiness (PH) work.3 Several recent studies, especially those by Nihan, have shown that this pericope used and transformed legislation from CC.4  Since H uses the Priestly text (P) as a base but expands this significantly, I call the whole corpus PH. For my recent thinking about the relationship of the two parts, see D. P. Wright, “Law and Creation in the Priestly-Holiness Writings of the Pentateuch” in Laws of Heaven, Laws of Nature: Legal Interpretations of Cosmic Phenomena in the Ancient World (ed. K. Schmid and C. Uehlinger; OBO 276; Fribourg: Academic Press; Göttingen: Vandenhoeck & Ruprecht, 2016), 73–101. 4  For those who see dependence on CC in some degree, see mainly C. Nihan, “Révisions scribales et transformations du droit dans lʼIsraël ancien: Le cas du talion (jus talionis),” in Loi et Justice dans la Littérature du Proche-Orient ancien (ed. O. Artus; BZABR 20; Wiesbaden: Harrassowitz, 2013), 123–158; idem, “Murder, Blasphemy and Sacral Law: Another Look at Lev 24,10–24,” ZABR 17 (2011), 211–240; idem, From Priestly Torah to Pentateuch (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 512–520. See also S. Chavel, Oracular Law and Priestly Historiography in the Torah (FAT 2/71; Tübingen: Mohr Siebeck, 2014), 23–92; idem, “Lev 24:10–23: Apostasy in H,” in J. Milgrom, Leviticus 23–27 (AB 3B; New York: Doubleday, 2001), 2143; M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon Press, 1985), 101; J. B. Gabel and C. B. Wheeler, “The Redactorʼs Hand in the Blasphemy Pericope of Leviticus XXIV,” VT 30 (1980), 227–229, here 228; B. S. Jackson, Wisdom-Laws: A Study of the Mishpatim of Exodus 21:1–22:16 (Oxford: Oxford University Press, 2006), 447–449; idem , Studies in the Semiotics of Biblical Law (JSOTSup 314; Sheffield: Sheffield Academic Press, 2000), 292–293; M. Leuchter, “The Ambiguous Details in the Blasphemer Narrative: Sources and Redaction in Leviticus 24:10–23,” JBL 130 (2011), 431–450; Milgrom, Leviticus 23–27, 2106; J. Vroom, “Recasting Mišpāṭîm: Legal Innovation in Leviticus 24:10–23,” JBL 3

Source Dependence and the Development of the Pentateuch

653

This essay highlights additional significant dimensions of this creative compositional engagement with CC and also shows that D was a considerable catalyst in this process. This will cast light on the compositional procedure by which the passage came to be, on the passageʼs inherent ideology, on its significance for the history of ideas about law and ritual, and on the development of the Pentateuch.

The Talion Laws The talion laws in Lev 24:17–22 provide a starting place because they reflect rather clearly the use of CC as a source and provide a foundation for broader observations about the blasphemy narrative frame. It is generally recognized that this passage is dependent on CCʼs talion law (Exod 21:23b–24):5 Leviticus 24:17–20

Exodus 21:23–25

‫׃‬ ‫יוּמת‬ ָ ‫אָדם מֹות‬ ָ ‫וְ ִאם־אָסֹון יִ ְהיֶ ה וְ נָ ַת ָתּה וְ ִאישׁ ִכּי יַ ֶכּה ָכּל־נֶ ֶפשׁ‬ ‫שׁ־בּ ֵה ָמה יְ ַשׁ ְלּ ֶמנָּ ה‬ ְ ‫וּמ ֵכּה נֶ ֶפ‬ ַ ‫׃‬ ‫נֶ ֶפשׁ ַתּ ַחת נָ ֶפשׁ‬ ‫׃‬ ‫נֶ ֶפשׁ ַתּ ַחת נָ ֶפשׁ‬ ‫וְ ִאישׁ ִכּי־יִ ֵתּן מוּם ַבּ ֲע ִמיתֹו‬ ‫׃‬ ‫ַכּ ֲא ֶשׁר ָע ָשׂה ֵכּן יֵ ָע ֶשׂה לֹּו‬ ‫ֶשׁ ֶבר ַתּ ַחת ֶשׁ ֶבר‬ ‫ַעיִ ן ַתּ ַחת ַעיִ ן‬ ‫ַעיִ ן ַתּ ַחת ַעיִ ן‬ ‫ֵשׁן ַתּ ַחת ֵשׁן‬ ‫ֵשׁן ַתּ ַחת ֵשׁן‬ ‫יָ ד ַתּ ַחת יָ ד‬ ‫׃‬ ‫ֶרגֶ ל ַתּ ַחת ָרגֶ ל‬ ‫ְכּוִ יָּ ה ַתּ ַחת ְכּוִ יָּ ה‬ ‫ֶפּ ַצע ַתּ ַחת ָפּ ַצע‬ ‫׃‬ ‫בּוּרה‬ ָ ‫בּוּרה ַתּ ַחת ַח‬ ָ ‫ַח‬ ‫׃‬ ‫אָדם ֵכּן יִ נָּ ֶתן בֹּו‬ ָ ‫ַכּ ֲא ֶשׁר יִ ֵתּן מוּם ָבּ‬

Leviticus 24 has three talionic equations in common with CC: ‫נֶ ֶפשׁ ַתּ ַחת נָ ֶפשׁ‬, ‫עיִ ן ַתּ ַחת ַעיִ ן‬, ַ and ‫שׁן ַתּ ַחת ֵשׁן‬. ֵ The phrase ‫שׁ ֶבר ַתּ ַחת ֶשׁ ֶבר‬, ֶ though not in CC, corresponds conceptually to CCʼs ‫יָ ד ַתּ ַחת יָ ד ֶרגֶ ל ַתּ ַחת ָרגֶ ל‬. Leviticus 24ʼs equation referring to breakage is just a summary of CCʼs arm and leg. (More will be said later on how this is reflected more broadly in the H legislation.) The differences in the talion law of Lev 24 come from responses to difficulties in CCʼs law and the recasting of material for a new purpose. One of the 131 (2012), 27–44; J. Weingreen, “The Case of the Blasphemer (Leviticus XXIV 10 ff.),” VT 22 (1972), 118–123, here 120–121. Most of my observations developed independently through my work on the Covenant Code (e.g., Wright, Inventing [see n. 1], 507, n. 22, 506, n. 19, and 436–437, nn. 120 and 127, in connection with pp. 154–191). 5  E.g., N ihan , “Murder” (see n. 4), 228–229; idem , “Révisions” (see n. 4), 126–140; Vroom, “Recasting” (see n. 4), 31, 34.

654

David P. Wright

difficulties in CCʼs formulation is the attachment of the talion law to its second miscarriage law, as its apodosis: ‫וְ ִאם־אָסֹון יִ ְהיֶ ה וְ נָ ַת ָתּה נֶ ֶפשׁ ַתּ ַחת נָ ֶפשׁ‬, followed by the rest of the talion equations (Exod 21:23–25).6 Leviticus 24 separated the talion law off and formulated laws of a general nature, i.e., about killing a person or animal (Lev 24:17–18) and injuring a person (vv. 19–20), to govern and illustrate the principles expressed by the talion equations. In doing this, Lev 24 detached the ‫ נֶ ֶפשׁ ַתּ ַחת נָ ֶפשׁ‬equation from the others in the talion list (v. 18b). The ambiguity of this phrase allowed application of the equation to killing both a human and an animal. In the case of humans, “life for life” refers to capital punishment; in the case of animals, it refers to replacement of the animal. Leviticus 24 derived its two laws that reflect the “life for life” equation from CC. They summarize or epitomize, respectively, laws from the first and second halves of CCʼs casuistic laws.7 The first half of CC – if we bracket off the introductory group of debt-slave laws in 21:2–11 (whose influence is manifested in Lev 25) – is mainly concerned with cases of homicide, assault, and relatable capital crimes. The topics include homicide (21:12–14), child rebellion and kidnapping (vv. 15–17), personal injury (vv. 18–19), death of a slave (vv. 20–21), miscarriage and talion (vv. 22–25), slave injury (vv. 26–27), and a lethally goring ox (vv. 28–32). The homicide law in Lev 24 echoes the language of the initial homicide law of CC that heads this section. The Covenant Codeʼs law reads ‫יוּמת‬ ָ ‫( ַמ ֵכּה ִאישׁ וָ ֵמת מֹות‬Exod 21:12). Leviticus 24:17 converted the participle to a finite verb in the protasis but kept the same infinitive-finite verbal formulation for its apodosis: ‫יוּמת‬ ָ ‫אָדם מֹות‬ ָ ‫וְ ִאישׁ ִכּי יַ ֶכּה ָכּל־נֶ ֶפשׁ‬. The description of the victim as a ‫אָדם‬ ָ ‫ נֶ ֶפשׁ‬along with the verb for striking obviated the need to replicate CCʼs result clause, ‫וָ ֵמת‬.8 Striking a ‫ נֶ ֶפשׁ‬means “to kill.” Leviticus 24 uses ‫ נֶ ֶפשׁ‬for a human victim in order to echo the principal term in CCʼs talion equation, ‫נֶ ֶפשׁ ַתּ ַחת נֶ ֶפשׁ‬. That this was the intent is signaled by the use of the parallel term ‫שׁ־בּ ֵה ָמה‬ ְ ‫ נֶ ֶפ‬for the animal victim in the next verse (v. 18). This particular term has no exact parallel elsewhere in PH or in the Bible and appears to be generated by the ‫ נֶ ֶפשׁ ַתּ ַחת נֶ ֶפשׁ‬formula.9 That both these terms, for human and animal victims, are associated with this formula is clear from

6  For detailed discussion of the meaning of CCʼs miscarriage laws, see Nihan, “Révisions” (see n. 4), 126–140; Wright, Inventing (see n. 1), 154–191. 7  Vroom, “Recasting” (see n. 4), 39, observes similarly “that the excerpt of laws from Lev 24:17–21 was meant to evoke the whole mišpāṭîm [i.e., CCʼs casuistic law] corpus” (see his pp. 36–40). 8  The result clause ‫וָ ֵמת‬, “so that he dies,” appears in the Greek but not Lev 24:17 MT. Nihan, “Murder” (see n. 4), 223, argues that the Greekʼs reading is secondary. 9  The parallel term ‫)חיָּ ה‬ ַ ‫(ה‬ ַ ‫ נֶ ֶפשׁ‬occurs in Gen 1:20, 21, 24, 30; 9:10, 12, 15, 16 and ‫נֶ ֶפשׁ‬ ‫ל־בּ ָשׂר‬ ָ ‫ ָכּ‬in Lev 17:14, cf. v. 11. ‫ל־בּ ָשׂר‬ ָ ‫ ָכּ‬is a key term in Pʼs flood story (Gen 6:12, 13, 17, 19; 7:15, 16, 21; 8:17; 9:11, 15, 16. 17).

Source Dependence and the Development of the Pentateuch

655

appending the ‫ נֶ ֶפשׁ ַתּ ַחת נֶ ֶפשׁ‬rationale immediately after the laws in which they appear (v. 18b after vv. 17, 18a).10 The formula ‫יוּמת‬ ָ ‫ מֹות‬is a motif in CCʼs capital and assault laws broadly. It appears mainly in the laws on rebellion against parents and kidnapping that follow CCʼs homicide laws (Exod 21:15–17; also 22:17–19). The formulation is echoed in the less specific ‫ נָ קֹם יִ נָּ ֵקם‬in the slave homicide law of v. 20. Other laws in the first half of CC describe the death penalty with a simple verb, either ‫יוּמת‬ ָ in v. 29 or ‫ ָלמוּת‬in v. 14. Epitomizing the first half of CCʼs casuistic laws is also visible, as discussed earlier, in the detachment of the “life for life” equation from the miscarriage law in Exod 21:23b and its attachment to the homicide law that in CC appears several verses earlier (Exod 21:12–14). The law about killing an animal in Lev 24:18a exemplifies laws appearing in the second half of CCʼs casuistic laws that deal with animal and property loss or theft. The cases include negligent loss of an animal (Exod 21:33–34), an ox goring an ox (vv. 35–36), animal theft (21:36+22:2b–3), animals or fire destroying crops (22:4–5), deposit and property loss (22:6–8), loss or theft of 10  Nihan, “Murder” (see n. 4), 233, says the term ‫אָדם‬ ָ ‫ נֶ ֶפשׁ‬derives from Gen 9:5 (part of the P flood story) in order to argue that Lev 24 implicitly views homicide and personal injury as affronts against the deity, like blasphemy, because humans are created in the image of the deity (Gen 9:5–6). The relationship of Lev 24 to Gen 9 is more complex. No doubt Lev 24, as part of PH, knew the P flood story, but Gen 9 already appears to have been interacting with CC. Gen 9 reflects various motifs from CC. The concept that animals have bloodguilt for killing humans in Gen 9:5 may be an interpretation of the requirement for stoning an ox that lethally gores a human in Exod 21:28, 29, and 32. The next verse (Gen 9:6) adds its rationale of creation in the divine image to justify this liability. CC did not necessarily have this rationale in mind in formulating its law (see Wright, Inventing [see n. 1], 222–229, 455, n. 81). P in Gen 9 may well have been creatively resolving a legislative conundrum in CC in accord with its view of creation. In addition, the law against homicide in Gen 9:6 has an inverted or chiastic form: ‫שׁ ֵֹפְך‬ ‫אָדם ָדּמֹו יִ ָשּׁ ֵפְך‬ ָ ‫אָדם ָבּ‬ ָ ‫דּם ָה‬,ַ “one who sheds the blood of a human, by a human his blood shall be shed” (cf. Chavel, “Lev 24:10–23” [see n. 4], 2143; cf. Lev 20:9). This maps the pattern of the talion formula ‫נֶ ֶפשׁ ַתּ ַחת נֶ ֶפשׁ‬, “life for life,” and therefore looks like an interpretation of the theory of retribution encoded in that formula. Gen 9 is also otherwise preoccupied with ‫נֶ ֶפשׁ‬, “life,” terminology (‫אְַך־בּ ָשׂר ְבּנַ ְפשֹׁו ָדמֹו‬, ָ v. 4; ‫אָחיו‬ ִ ‫] ִמיַּ ד ִאישׁ‬. . .[ ‫יכם ֶא ְדר ֹשׁ‬ ֶ ‫ת־דּ ְמ ֶכם ְלנַ ְפשׁ ֵֹת‬ ִ ‫וְ אְַך ֶא‬ ‫אָדם‬ ָ ‫א ְדר ֹשׁ ֶאת־נֶ ֶפשׁ ָה‬, ֶ v. 5). This can be attributed to a preoccupation with CCʼs representation of talion. That Gen 9 may have been using CC as a source is also indirectly indicated, as I have argued, by the use of the non-P flood story as a model; see D. P. Wright, “Profane versus Sacrificial Slaughter: The Priestly Recasting of the Yahwist Flood Story,” in Current Issues in Priestly and Related Literature: The Legacy of Jacob Milgrom and Beyond (ed. R. Gane and A. Taggar-Cohen; SBLRBS 82; Atlanta: Society of Biblical Literature, 2015), 125–154. For Pʼs dependence on non-P, see also T. Abusch, “Biblical Accounts of Prehistory: Their Meaning and Formation,” in Bringing the Hidden to Light: Studies in Honor of Stephen A. Geller (ed. K. F. Kravitz and D. M. Sharon; New York: Jewish Theological Seminary; Winona Lake, IN: Eisenbrauns, 2007), 1–17. In view of this data, the homicide legislation in Lev 24 appears to be part of a continuing engagement of the PH school with CC and non-P sources broadly. The common term ‫אָדם‬ ָ ‫ נֶ ֶפשׁ‬may grow out of a background matrix of this scribal engagement.

656

David P. Wright

a deposited animal (22:9–12), and death of a rented animal (22:13–14). The Covenant Code regularly prescribes the requirement of replacement in these laws with an infinitive-finite verbal construction, ‫( ַשׁ ֵלּם יְ ַשׁ ֵלּם‬21:36; 22:2, 5, 6, 13), similar to the ‫יוּמת‬ ָ ‫ מֹות‬formula in the first half of CC. The requirement of repayment in the second half of CCʼs casuistic laws is also expressed with the simple verb ‫( יְ ַשׁ ֵלּם‬21:37; 22:3, 4, 8, 10, 11, 12, 14). Leviticus 24:18 used the simple verb for its legal penalty: ‫יְ ַשׁ ְלּ ֶמנָּ ה‬. In contrast to its homicide law, Lev 24 maintained the participial formulation from CCʼs homicide law in the protasis of its animal law. Compare ‫שׁ־בּ ֵה ָמה יְ ַשׁ ְלּ ֶמנָּ ה‬ ְ ‫וּמ ֵכּה נֶ ֶפ‬ ַ (Lev 24:18) with ‫ַמ ֵכּה ִאישׁ וָ ֵמת‬ ‫יוּמת‬ ָ ‫( מֹות‬Exod 21:12). That the animal law in Lev 24 summarizes the second half of CCʼs casuistic laws should not be ignored in an assessment of the legal ideology and purpose of the passage, even though the connection of the talion passage to blasphemy is primarily by way of the more serious case of homicide. Leviticus 24 derived the notion of applying the language and concept of talion to animals from the first few laws about animal loss in CC. These use the talion formula “X for (‫)תּ ַחת‬ ַ X” to describe the compensation due for animals. For example, the owner of an ox that kills another ox repays ox for ox (‫ַשׁ ֵלּם יְ ַשׁ ֵלּם‬ ‫שֹׁור ַתּ ַחת ַהשֹּׁור‬, Exod 21:36). One who steals an ox or flock animal is to repay multiple animals (‫אַר ַבּע־צֹאן ַתּ ַחת ַה ֶשּׂה‬ ְ ְ‫ח ִמ ָשּׁה ָב ָקר יְ ַשׁ ֵלּם ַתּ ַחת ַהשֹּׁור ו‬, ֲ v. 37). Leviticus 24ʼs detachment of the talion law from the context of miscarriage in Exod 21:22–25 arose from a correct understanding that CCʼs talion law, despite its being linked to miscarriage, is in fact a law of general application.11 This is indicated by CCʼs unusual second person (‫וְ נָ ַת ָתּה‬, “you shall give,” Exod 21:23) that heads the talion list, where one expects a third-person verb consistent with casuistic legal formulation broadly. The exceptional second person broadens the audience and, in this way, metonymically entails other cases besides miscarriage. The universality of the law is also indicated by the long list of talion equations, overkill in the rather sparse wording of CCʼs legislation if specifically limited to miscarriage alone (vv. 24–25). The broad application of the law is also seen in how it complements the homicide law in Exod 21:12–14. These earlier verses prescribe capital punishment for the intentional killer but, strangely, say nothing about any obligation an unintentional killer has. This is left hanging until the talion law, which prescribes payment of damages due for unintentional homicide. That the case is one of  For the details summarized in these paragraphs about the meaning of the miscarriage and talion laws in CC, see Wright, Inventing (see n. 1), 176–186; idem, “The Origin, Development, and Context of the Covenant Code (Exodus 20:23–23:19),” in The Book of Exodus: Composition, Reception, and Interpretation (ed. T. B. Dozeman et al.; FIOTL; VTSup 164; Leiden: Brill, 2014), 220–244; idem, “Miscarriage, Talion, and Slave Injury in the Covenant Code and Hammurabiʼs Laws,” in Gazing on the Deep: Ancient Near Eastern and Other Studies in Honor of Tzvi Abusch (ed. J. Stackert et al.; Bethesda, MD: CDL Press, 2010), 539–564. 11

Source Dependence and the Development of the Pentateuch

657

inadvertence is evidenced by its being set in the context of men fighting (v. 22).12 They only happen to knock the woman. Thus “giving” (*‫ )נתן‬life for life and so forth refers to payment. This verb is used elsewhere in CC to describe payment (Exod 21:19, 32) and is even used thus in the first miscarriage law (v. 22; compare the similar requirement of payment, also using this verb, in the goring ox law of v. 30). Thus the talion equation “life for life” does not decisively refer to exacting capital punishment. Even Lev 24:18 recognizes that this equation is equivocal and can refer to payment in the case of animals. The Covenant Codeʼs reliance on the Laws of Hammurabi (LH) as a source, for which I have argued in detail elsewhere, indicates further that the talion law is a general law and prescribes indemnification in other cases of inadvertent homicide and injury.13 The Covenant Codeʼs main homicide and injury laws in 21:12–14, 18–21 correlate with the series of laws on inadvertent homicide and injury in LH 206–208. The latter require payment of the physician in cases of injury and indemnification for homicide. The Covenant Codeʼs corresponding laws echo the requirement of paying the physician (v. 19) but say nothing about indemnification for inadvertent homicide. The Covenant Code uses the next laws on miscarriage (vv. 22–25) to fill this gap. These correlate with Hammurabiʼs two laws on miscarriage (LH 209–210), which similarly follow the aforementioned inadvertent homicide and injury laws in that context. The major difference in CCʼs miscarriage laws is the apodosis of the second law, where the woman dies. Hammurabiʼs text calls for vicarious punishment – “they shall kill his daughter” (mārassu idukkū) – because the victim is portrayed as a “daughter of a man” (mārat awīlim). The Covenant Code rejected the requirement of vicarious punishment (as it does explicitly in 21:3114) and replaced it with its talion law. The Covenant Code formulated this creatively by summarizing in list form Hammurabiʼs talion laws that appeared a bit earlier in LH 196–201. It used this as the occasion for encoding the rule about indemnification for inadvertent homicide together with injury. To do this, it converted the modality of intentional injury and homicide in Hammurabiʼs miscarriage laws to one of inadvertence by bringing in the context of a fight from Hammurabiʼs assault law in LH 206. (The Covenant Code also maintained the context of a fight in 21:18, which more directly parallels LH 206.) Thus CCʼs miscarriage law is compositionally 12  That the case involves inadvertence, see W right , Inventing (see n. 1), 178–179 and 437–438, n. 128; N ihan , “Révisions” (see n. 4), 136–140, 158; idem , “Murder” (see n. 4), 225–226; L. Schwienhorst-Schönberger, “Auge um Auge, Zahn um Zahn,” in Bibel und Liturgie 63 (1990), 166–167; idem, Das Bundesbuch (BZAW 188; Berlin: de Gruyter, 1990), 75, 101. 13  For the argument of dependence, see Wright, Inventing (see n. 1); for the homicide and injury laws discussed in this paragraph, see pp. 154–191 therein. 14  This appears to react to the description of the victim in Hammurabiʼs second goring-ox law (LH 251) as a mār awīlim, literally, “son of a man.” See Wright, Inventing (see n. 1), 211–212.

658

David P. Wright

intertwined with the general laws on homicide and injury in 21:12–14, 18–21. Together they constitute an integrated and systematic legal statement. Separating the talion law off from the particular legal context of miscarriage allowed Lev 24 to recast the operative modality. The cases of homicide, animal killing, and injury of a fellow Israelite in this passage are now intentional acts. Such rewriting can be considered a natural hermeneutical development from thinking about law, where introducing variables in the setup of a case produces different outcomes. It is an exploration within the range of maximal variation, as described by Eichler.15 Thus the differences in Lev 24 should be taken not as a rejection of CCʼs laws but as following and laying out an alternative legal algorithm. The main reason for this change was to provide an analogy for the case of blasphemy, an intentional act and capital crime. Characterizing these acts as intentional required Lev 24 to impose more austere penalties. In addition to capital punishment for homicide, Lev 24 used the talion equations for injuries to describe or justify the severe requirement of exacting the same injury on the assailant.16 Leviticus 24 states this twice (vv. 19 and 20), so there is no question. This forms an envelope around the talion equations referring to body parts and constitutes the center of a quasi-chiastic structure in the larger passage. More will be said on this structure in the section on Dʼs influence, below. A broader systematic stimulus for the change of the modality of intention was the association of the talion injuries with bodily defects of priests. The H legislation treats these a few chapters earlier in Lev 21 (Lev 21:17–23), with similar bodily defects of animals treated in chapter 22 (22:20–25). Leviticus 24, different from its CC source, describes bodily injuries in the context of talion legislation as a ‫מוּם‬, the same term used for defects in Lev 21–22. To tell from its use elsewhere in the Bible, this is not a synonym for injury. It refers rather to a physical defect and, by extension, to moral fault.17 The use of this term indicates that Lev 24 is considering the talion injuries alongside the human defects described in Lev 21. This connection is further indicated by the summary of CCʼs arm and leg under the general term “break,” ‫שׁ ֶבר‬. ֶ Leviticus 21 displays the itemization of the talion injuries in CC when it describes priestly blemishes as  B. L. Eichler, “Literary Structures in the Laws of Eshnunna,” in Language, Literature, and History: Philological and Historical Studies Presented to Erica Reiner (ed. F. Rochberg-Halton; New Haven, CT: American Oriental Society, 1987), 71–84. For introducing variables in rewriting law, see Wright, Inventing (see n. 1), 38, 92, 139, 143. For the related feature of conceptual inversion, see pp. 27, 33, 42, 64, 73, 95, 134, 138, 159, 214, 234, 296, 301, 332, 350. 16  Nihan, “Révision” (see n. 4), 149–150, is right that Lev 24 is not prescribing financial compensation. 17  For a physical defect, see Lev 21:17–18, 21–23; 22:20–21, 25; Num 19:2; Deut 15:21; 17:1; 2 Sam 14:25; Song 4:7; Dan 1:4; for a moral fault, see Prov 9:7; Job 11:15; 31:7; cf. Deut 32:5. 15

Source Dependence and the Development of the Pentateuch

659

including ‫( ֶשׁ ֶבר ָרגֶ ל אֹו ֶשׁ ֶבר יָ ד‬v. 19). The ‫ יָ ד‬and ‫ ֶרגֶ ל‬are precisely the body parts found in CCʼs talion list but missing in Lev 24 and subsumed under the general term ‫שׁ ֶבר‬. ֶ This overlap of terminology suggests an interconnected relationship in the formulation of these texts.18 Leviticus 21 took up the arm and leg from CC and made explicit the nature of the injury with the addition of the noun ‫שׁ ֶבר‬. ֶ The animal blemish list, apparently also developed as part of this process, summarized this with the adjective ‫( ָשׁבוּר‬22:22). Leviticus 24 let the general term ‫ ֶשׁ ֶבר‬suffice in view of the more extensive list of blemishes in Lev 21.19  For example, E. O tto , Kontinuum und Proprium: Studien zur Sozial- und Rechts­ge­ schichte des Alten Orients und des Alten Testaments (OBC 8; Wiesbaden: Harrassowitz, 1996), 241–242, observes the connection between the passages in the terms ‫ מוּם‬and ‫שׁ ֶבר‬. ֶ But he attributes these in Lev 24 to a Priestly redactor who is providing an etiological narrative for pre-Priestly cultic legal tradition (Lev 24:16aα, 17–21). He does not see the connection of ‫ ֶשׁ ֶבר‬to the context of CC nor is he aware of the use of sources and what this means for the compositional unity of the passage. See the conclusion and n. 59, below. N ihan , “Murder” (see n. 4), 224–225, n. 46, 234, and idem, “Révisions” (see n. 4), 148, 155, notes the unique connection between the terms in the two chapters but does not draw the connection of the language in Lev 21 to CCʼs talion list. In any case, I would agree with Nihan that, though Lev 24 has a terminological and conceptual connection with Lev 21, it is not thinking about injuries that prevent cultic access nor about priests as victims. For example, nowhere does PH speak of injuries or bodily defects preventing cultic access for the lay Israelite, and priests with such defects can eat even the most holy offerings (Lev 21:22), which must be eaten in the sanctuary court (see D. P. Wright, The Disposal of Impurity [SBLDS 101; Atlanta: Society of Biblical Literature, 1987], 232–235). A blemished priest is only restricted from working at the altar or inside the tabernacle tent (21:21–23). 19  Hʼs passages on priestly and animal blemishes are apparently influenced by several D passages (Deut 15:19–23; 17:1; 23:2–3) in addition to CCʼs talion list. That H depends on D is evident in the correlation of language and motifs, including Dʼs general description of an animal having a blemish (‫וְ ִכי־יִ ְהיֶ ה בֹו מוּם‬, Deut 15:21; ‫א ֶשׁר יִ ְהיֶ ה בֹו מוּם‬, ֲ 17:1) and Hʼs similar description of animal and priestly blemishes (‫כֹּל ֲא ֶשׁר־בֹּו מוּם‬, Lev 22:20; ‫כּל־מוּם ל ֹא יִ ְהיֶ ה־בֹּו‬, ָ ִ ‫כּ‬, ָ v. 21; ‫ל־אישׁ ֲא ֶשׁר־בֹּו מוּם‬ ִ ‫כּי ָכ‬, ִ 21:18; ‫אישׁ ֲא ֶשׁר־יִ ְהיֶ ה בֹו‬, ִ v. 19; ]. . .[ ‫] מוּם בֹּו‬. . .[ ‫ל־אישׁ ֲא ֶשׁר־בֹּו מוּם‬ v. 21; ‫כּי־מוּם בֹּו‬, ִ v. 23); Dʼs pair of conditions “lame and blind” (‫פּ ֵסּ ַח אֹו ִעוֵּ ר‬, ִ Deut 15:21), found for Hʼs priests (‫עוֵּ ר אֹו ִפ ֵסּ ַח‬, ִ Lev 21:18) and more loosely for animals (‫ ַעוֶּ ֶרת‬followed by circumlocution of lameness, 22:22); and Dʼs prohibitions not to sacrifice (‫לֹא ִתזְ ָבּ ֶחנּוּ ַליהוָ ה‬ ‫ֹלהיָך‬ ֶ ‫א‬, ֱ Deut 15:21; ‫ֹלהיָך‬ ֶ ‫א־תזְ ַבּח ַליהוָ ה ֱא‬ ִ ֹ ‫ל‬, 17:1) and similar prohibitions in H, which use PHʼs preferred verb for sacrifice (‫ל ֹא ַת ְק ִריבוּ‬, Lev 22:20; ‫ֹא־ת ְק ִריבוּ ֵא ֶלּה ַליהוָ ה‬ ַ ‫ל‬, v. 22; ‫ל ֹא ַת ְק ִריבוּ‬ ‫ליהוָ ה‬,ַ v. 24). Hʼs prohibitions against priests with bodily defects working in the cult are an extension of this language, using verbs for approaching complemented by verbs for offering or officiating (]. . .[ ‫לֹא יִ ְק ָרב ִאישׁ ִעוֵּ ר‬, Lev 21:18; ‫ֹלהיו‬ ָ ‫ת־א ֵשּׁי יְ הוָ ה מוּם בֹּו ֵאת ֶל ֶחם ֱא‬ ִ ‫לֹא יִ גַּ שׁ ְל ַה ְק ִריב ֶא‬ ‫לֹא יִ גַּ שׁ ְל ַה ְק ִריב‬, v. 21; more broadly ‫ל־ה ִמּזְ ֵבּ ַח לֹא יִ גַּ שׁ‬ ַ ‫ל־ה ָפּר ֶֹכת לֹא יָבֹא וְ ֶא‬ ַ ‫אְַך ֶא‬, v. 23). Note also the broad conceptual similarity of Lev 22:20 and Deut 17:1, where unacceptability in H correlates with Dʼs motif of abomination in a rationale clause. A factor leading H to extend animal blemishes to humans was presumably Dʼs listing of certain defects (genital and genealogical) that disqualify human access or inclusion in the ‫( ְק ַהל יְ הוָ ה‬Deut 23:2–3). The motif of testicular defect appears in Deut 23:2 and in Hʼs Lev 21:20b and 22:24. The latter uses the term ‫ּכרוּת‬, ָ found in D. 18

660

David P. Wright

Lastly for the analysis of the talion laws, my study of CC indicated to me that CCʼs trio of injuries ‫כּוִ יָּ ה‬, ְ ‫פּ ַצע‬, ֶ and ‫בּוּרה‬ ָ ‫ח‬, ַ which appears at the end of its talion list (Exod 21:25), was stimulated by LH and would presumably have been in CC before the formulation of Lev 24. But this trio does not appear in Lev 24.20 The connection with priestly blemishes suggests a reason why it is absent. Leviticus 21 is interested in permanent and serious defects, and Lev 24 may have this same interest. The final trio of injuries in CC may have been viewed as less serious and not necessarily permanent. Therefore Lev 24 omitted them. In Lev 21, if the trio is reflected at all, it was replaced with other conditions.21 In contrast to this The basic outline of the priestly blemishes may be based on a serial use and expansion of sources: the initial pair of conditions ‫( ִעוֵּ ר אֹו ִפ ֵסּ ַח‬Lev 21:18) comes from Deut 15:21; the pair of leg and arm breakage in Lev 21:19, from CC (Exod 21:24), as described in the body of this paper; and testicular defect in 21:20b from Deut 23:2. Hʼs animal blemish listing follows this same pattern, starting with ‫ ַעוֶּ ֶרת‬in 22:22 reflecting Deut 15:21; ‫שׁבוּר‬, ָ next in 22:20, reflecting CC; and genital defect at the end, from Deut 23:2. ‫ ַעוֶּ ֶרת‬and ‫ ָשׁבוּר‬in Lev 22:22 appear to be a secondary realization of these motifs from the correlates in the list of human defects in Lev 21:18–19, which for its part relies directly on D and CC (see below). To create its detailed list of conditions in Lev 21 and 22, H may have augmented the basic terms or motifs derived from its different sources with related conditions, though this is complicated by obscurity of several of the conditions. ‫ ָח ֻרם‬and ‫רוּע‬ ַ ‫( ָשׂ‬Lev 21:18) augment ‫ִּפ ֵּס ַח‬ and ‫ ִעוֵּ ר‬as basic conditions, perhaps related to the legs (21:18; see n. 21). Lev 22 delayed ‫רוּע‬ ַ ‫ָשׂ‬ and ‫קלוּט‬,ָ presumably animal correlates of human ‫ ָח ֻרם‬and ‫רוּע‬ ַ ‫שׂ‬, ָ until 22:23, as an exception in cases of some sacrifices. Human ‫ ּגִ ֵבּן‬and ‫ ַּדק‬in Lev 21:20, if they are bodily deformities (back deformity and problems of body form/musculature), augment leg and hand breakage as skeletal or structural defects. The conditions ‫תּ ַב ֻלּל ְבּ ֵעינֹו‬, ְ ‫ ּגָ ָרב‬and ‫ יַ ֶלּ ֶפת‬in Lev 21:20 – some defect of the eye followed by two skin irregularities – may refract CCʼs reference to the eye and the trio of superficial (skin) injuries (‫כּוִ יָּ ה‬, ְ ‫פּ ַצע‬, ֶ and ‫בּוּרה‬ ָ ‫)ח‬ ַ in CCʼs talion list (Exod 21:24–25). Lev 21:20 mentions this other eye condition because blindness was already stated in Lev 21:18, derived from Deut 15:21. The two conditions ‫ ּגָ ָרב‬and ‫ יַ ֶלּ ֶפת‬may (also) reflect concern about conditions similar to ‫צ ָר ַעת‬, ָ a priestly concern otherwise (Lev 13–14). The series ‫יַבּ ֶלת‬, ֶ ‫ּגָ ָרב‬, and ‫ יַ ֶלּ ֶפת‬in Lev 22:22 is the animal correlate to human ‫תּ ַב ֻלּל ְבּ ֵעינֹו‬, ְ ‫ּגָ ָרב‬, and ‫ יַ ֶלּ ֶפת‬in 21:20. Direct terminological correspondences with sources in the list of priestly blemishes (‫עוֵּ ר‬, ִ ִ ‫רגֶ ל‬,ֶ ‫ )יָ ד‬show that that list has some compositional priority over the animal blemish ‫ּפ ֵּס ַח‬, passage in Lev 22. But the topic of testicular defect in the animal list (Lev 22:24) has a direct correspondence with Deut 23:2 (‫)ּכרוּת‬, ָ which indicates it is not simply derivative from Lev 21:20b but directly based on a D source. Both lists thus appear to have been created in tandem, concomitantly using sources yet one influencing the other in the process of formulation. 20  It is a trifurcation of the simmum, “wound,” of LH 206–208, just as CCʼs “arm” and “leg” are a bifurcation of the broken bone from LH 197, 199. See Wright, “Origin” (see n. 11), 230; idem , Inventing (see n. 1), 181–182. 21  Part of the difficulty in this assessment is the unclarity of some of the terms in Lev 21. For example, it is not clear if ‫ ָח ֻרם‬and ‫רוּע‬ ַ ‫( ָשׂ‬21:18) refer to defects in the legs (respectively, stunted or elongated) or defects (e.g., splitting) of the nose and ear (for discussion, see J. Milgrom, Leviticus 17–22 [AB 3B; New York: Doubleday, 2000], 1827–1828). If the latter, they may have been caused by an injury and be precisely what might be included under CCʼs ‫ ֶפּ ַצע‬or ‫בּוּרה‬ ָ ‫ח‬. ַ But the list of animal blemishes permits an animal that is ‫רוּע‬ ַ ‫ ָשׂ‬in the case of a freewill

Source Dependence and the Development of the Pentateuch

661

omission, Lev 24 included one permanent condition from CC missing from the priestly blemishes: the loss of a tooth. Leviticus 21 likely omitted this because edentulism was common and would have debarred most priests.

The Blasphemy Narrative and Law The blasphemy narrative and legislation that surround Lev 24ʼs talion law are similarly indebted to CC and, like the talion laws, systematize laws from diverse points in CC. There are actually two blasphemy laws in Lev 24: Leviticus 24:15–16

‫יוּמת‬ ָ ‫ֹלהיו וְ נָ ָשׂא ֶח ְטאֹו וְ נ ֵֹקב ֵשׁם־יְ הוָ ה מֹות‬ ָ ‫ִאישׁ ִאישׁ ִכּי־יְ ַק ֵלּל ֱא‬

The attached penalties indicate that, conceptually, these are prohibitions against the behaviors they describe. They correlate with two rules directly prohibiting speech about deities in CC in Exod 22:27 and 23:13: Exodus 22:27 Exodus 23:13

‫ֹלהים לֹא ְת ַק ֵלּל וְ נָ ִשׂיא ְב ַע ְמָּך לֹא ָתאֹר‬ ִ ‫ֱא‬ ‫ל־פּיָך‬ ִ ‫ֹלהים ֲא ֵח ִרים לֹא ַתזְ ִכּירוּ לֹא יִ ָשּׁ ַמע ַע‬ ִ ‫וְ ֵשׁם ֱא‬

A number of scholars have already recognized that Lev 24ʼs first law echoes CCʼs cursing law in Exod 22:27.22 It reflects the same basic elements: the verb *‫( קלל‬piel)23 with the object ‫ֹלהים‬ ִ ‫א‬. ֱ These two laws also use *‫נשא‬, though with or vow offering (22:23). This seems to be a binary contrast to the companion word ‫קלוּט‬,ָ also obscure but whose root points to a meaning of “shortness.” Both terms together also seem to refer to naturally occurring (i.e., congenital) conditions. They therefore seem to be leg defects. Priestly ‫רוּע‬ ַ ‫ ָשׂ‬in 21:18 and its companion term ‫ ָח ֻרם‬would thus be similar conditions. A hint that ‫ ָח ֻרם‬and ‫רוּע‬ ַ ‫ ָשׂ‬refer to leg defects is Hʼs reversal of CCʼs order of limbs to leg and hand in ַ ‫ ָשׂ‬that suggests Lev 21:19. This with ‫ ִּפ ֵסּ ַח‬provides a contextual envelope around ‫ ָח ֻרם‬and ‫רוּע‬ they also have to do with the legs. See n. 19 for the suggestion that the ‫ ּגָ ָרב‬and ‫ יַ ֶלּ ֶפת‬in Lev 21:20 may reflect in some way the trio of injuries in Exod 21:25. 22  See N ihan , “Murder” (see n. 4), 219–220; idem , “Révisions” (see n. 4), 144–145; Milgrom, Leviticus 23–27 (see n. 4), 2115; Chavel, Oracular Law (see n. 4), 55; idem, “Lev 24:10–23” (see n. 4), 2143. 23  The root *‫ קלל‬has a specific meaning of “curse” and a broader meaning of “disparage.” The parallel root *‫ ארר‬in Exod 22:27, the verse that is the source of *‫קלל‬, indicates that the term should be understood as curse (so C havel , “Lev 24:10–23” [see n. 4], 2143). For the argument of this paper, how the verb is construed is not important. For discussion, see Chavel, Oracular Law (see n. 4), 27–33, 48–49; R. R. Hutton, “The Case of the Blasphemer Revisited (Lev. XXIV 10–23),” VT 49 (1999), 532–541; S. T. K amionkowski , “Leviticus 24,10–23 in Light of Hʼs Concept of Holiness,” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions (ed. S. Shectman and J. S. Baden; ATANT 95; Zürich: Theologischer Verlag, 2009), 73–86, here 76–79; Milgrom, Leviticus 23–27 (see n. 4), 2109; Nihan,

662

David P. Wright

different meanings (“bear” versus “chieftain”). How this may count as evidence of textual dependence will be explained later. The second laws in both collections (Lev 24:16; Exod 23:13) legislate against speaking the ‫שׁם‬, ֵ “name,” of deities. The primary difference is that CCʼs law prohibits speaking the names of other gods whereas Lev 24 sanctions the misuse of YHWHʼs name. This shift or inversion of content will also be explained later. What needs an explanation first is the distance between the two laws in CC. Though separated by nearly a full biblical chapter, the two laws correlate in the structure of CC.24 The final apodictic laws of CC consist of two parallel passages (22:20–30 // 23:9–19), as laid out in table 1. I call these passages strings, to give them a distinctive label for ease of reference. These strings are set around a chiastically formulated core in 23:1–8. Recognition of the string structure helps to make sense of the perplexing redundancy and contextual shifts in the final apodictic laws. What more fully explains the shifting topics as well as the relationship of the final apodictic laws to the initial apodictic laws (20:23–26) is their apparent collective dependence on the thematic sequence in a section of admonitions and exhortations in the epilogue of LH.25 Table 1: Parallel String Structure of the Final Apodictic Laws of the Covenant Code Topic

String I (Exodus 22:20–30)

String II (Exodus 23:9–19)

1. general law about the poor

22:20–23: immigrant, widow, orphan not to be oppressed; Egypt rationale.

23:9: immigrant not to be oppressed; Egypt rationale.

2. two long laws benefiting the poor

(A) 22:24: interest not to be taken from poor; (B) 22:25–26: garment pledge not to be retained.

(A) 23:10–11: the poor eat from produce of seventh-year field; (B) 23:12: poor rest on seventh day.

“Murder” (see n. 4), 215–216; idem, “Révisions” (see n. 4), 142–143; Weingreen, “Case of the Blasphemer” (see n. 4), 119; Wright, Inventing (see n. 1), 195–196. 24  The basic structure was identified by C. C armichael , “A Singular Method of Codification of Law in the Mishpatim,” ZAW 84 (1972), 19–25, with observations developed by J ackson , Wisdom-Laws (see n. 4), 457–458 (see also idem , “Modelling Biblical Law: The Covenant Code,” Chicago-Kent Law Review 70 [1995]: 1745–1827, here 1794–1795, 1801–1802; idem , Semiotics [see n. 4], 218–222). It is more fully articulated in W right , Inventing (see n. 1), 10–16, 51–90, 286–321, 322–332, and connected to stimuli from the epilogue of Hammurabiʼs Laws. 25  For a recent abbreviated charting of these correlations, see Wright, “Origin” (see n. 11), 235–236; for other charts, see Wright, Inventing (see n. 1), 12–13, 82–90 (the latter includes the full texts). For a more detailed charting of these strings and their relationship to LH, see my other essay in this volume, “The Covenant Code Appendix (Exodus 23:20–33), Neo-Assyrian Sources, and Implications for Pentateuchal Study,” pp. 47–85.

Source Dependence and the Development of the Pentateuch

663

(A) 22:27a: God not to be 3. two short laws cursed (*‫;)קלל‬ about speaking about sovereigns (B) 22:27b: chieftain (‫)נָ ִשׂיא‬ not to be cursed.

(A) 23:13bα: names (‫)שׁם‬ ֵ of other gods not to be recalled; (B) 23:13bβ: (name of these gods) not to be heard on lips.

4. cultic laws

23:17–19: three annual festivals to be observed where people appear before/“see” the deity; miscellaneous cultic rules: leaven with sacrificial blood not to be offered, festival offering not to remain until morning, first fruits to be offered, kid not to be boiled in motherʼs ‫חלב‬.

22:28–30: miscellaneous cultic rules: offer first produce, dedicate firstborn humans, offer firstborn animals after remaining with mother a week, carrion not to be eaten because people are holy.

Leviticus 24 appears to have recognized the structural parallelism and thematic relationship between Exod 22:27 and 23:13 to create a double blasphemy law. That H is cognizant of this topical symmetry in CC is found in its laws on sacrificial animals and their offspring in Lev 22:27–28: Leviticus 22:27–28 ‫וּמיֹּום ַה ְשּׁ ִמינִ י וָ ָה ְלאָה יֵ ָר ֶצה ְל ָק ְר ַבּן ִא ֶשּׁה‬ ִ ‫אֹו־עז ִכּי יִ וָּ ֵלד וְ ָהיָ ה ִשׁ ְב ַעת יָ ִמים ַתּ ַחת ִאמֹּו‬ ֵ ‫אֹו־כ ֶשׂב‬ ֶ ‫שֹׁור‬ ‫׃‬ ‫ת־בּנֹו לֹא ִת ְשׁ ֲחטוּ ְבּיֹום ֶא ָחד‬ ְ ‫אֹו־שׂה אֹתֹו וְ ֶא‬ ֶ ‫׃ וְ שֹׁור‬ ‫ַליהוָ ה‬

These laws appear to be a collocation and interpretation of two laws from the group of cultic laws at end of the strings of CCʼs final apodictic laws in Exod 22:28b–29 and 23:19 (topic 4 in table 1, above) and not far after the laws on speaking about deities: Exodus 22:28b–29 ]. . .[ ‫נֹו־לי‬ ִ ‫ם־אמֹּו ַבּיֹּום ַה ְשּׁ ִמינִ י ִתּ ְתּ‬ ִ ‫ן־תּ ֲע ֶשׂה ְלשׁ ְֹרָך ְלצֹאנֶ ָך ִשׁ ְב ַעת יָ ִמים יִ ְהיֶ ה ִע‬ ַ ‫ן־לּי ֵכּ‬ ִ ‫ְבּכֹור ָבּנֶ יָך ִתּ ֶתּ‬ Exodus 23:19

‫א־ת ַב ֵשּׁל גְּ ִדי ַבּ ֲח ֵלב ִאמֹּו‬ ְ ֹ‫ל‬

The two laws in CC, distant from each other as are the laws about speaking about deities, are clearly related. Both have to do with mother animals and their young. Though the term in the mother-kid law may mean – certainly it has come to mean – “milk,” Lev 24 may have understood or creatively read it to mean “fat” and thus read CCʼs rule as prohibiting offering mother and offspring at the same time.26 This understanding of CCʼs thematic structure is not new to the H school. Deuteronomy shows the same collocation of parallel laws from the two strings in CCʼs final apodictic laws in the final verse of its food laws: 26  See J. S asson , “Ritual Wisdom? On ‘Seething a Kid in Its Motherʼs Milk,’ ” in Kein Land für sich allein: Studien zum Kulturkontakt in Kanaan, Israel/Palästina und Ebirnâri für Manfred Weippert zum 65. Geburtstag (ed. U. Hübner and E. A. Knauf; OBO 186; Freiburg: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht, 2002), 294–308, here 305.

664

David P. Wright

Deuteronomy 14:21 ‫אַתּה ַליהוָ ה‬ ָ ‫ר־בּ ְשׁ ָע ֶריָך ִתּ ְתּנֶ נָּ ה וַ ֲא ָכ ָלהּ אֹו ָמכֹר ְלנָ ְכ ִרי ִכּי ַעם ָקדֹושׁ‬ ִ ‫ֹאכלוּ ָכל־נְ ֵב ָלה ַלגֵּ ר ֲא ֶשׁ‬ ְ ‫ל ֹא ת‬ ‫׃‬ ‫א־ת ַב ֵשּׁל גְּ ִדי ַבּ ֲח ֵלב ִאמֹּו‬ ְ ֹ ‫ֹלהיָך ל‬ ֶ ‫ֱא‬

This anthologizes the two dietary laws at the end of CCʼs two strings: Exodus 22:30 Exodus 23:19b

]. . .[ ‫׃‬ ‫אכלוּ ַל ֶכּ ֶלב ַתּ ְשׁ ִלכוּן אֹתֹו‬ ֵ ֹ ‫וּב ָשׂר ַבּ ָשּׂ ֶדה ְט ֵר ָפה לֹא ת‬ ָ ‫וְ אַנְ ֵשׁי־ק ֶֹדשׁ ִתּ ְהיוּן ִלי‬ ‫׃‬ ‫א־ת ַב ֵשּׁל גְּ ִדי ַבּ ֲח ֵלב ִאמֹּו‬ ְ ֹ‫ל‬

These examples of revision point to an enduring background understanding of the structuring principles of CCʼs final apodictic laws in Hebrew scribal culture. Leviticus 24ʼs blasphemy legislation appears to be reacting to a larger contextual web in CC. It solves two problems in CCʼs laws about speaking about deities (Exod 22:27; 23:13). First, it supplies a penalty for the crime. The Covenant Codeʼs prohibitions lack this because they are formulated apodictically. I will come back to influences from CC for the penalties in Lev 24 in a moment. Second, Lev 24 fills a gap in the larger context of CC. The cursing prohibitions in Exod 22:27 and 23:13 are complemented by a positive description of speaking YHWHʼs name in the initial apodictic laws. In the context of prescribing the building of an altar and sacrifice, the deity announces ‫ל־ה ָמּקֹום ֲא ֶשׁר אַזְ ִכּיר‬ ַ ‫ְבּ ָכ‬ ‫וּב ַר ְכ ִתּיָך‬ ֵ ‫ת־שׁ ִמי אָבֹוא ֵא ֶליָך‬ ְ ‫( ֶא‬20:24). Some have argued that this text originally had a second-person verb (‫)ּתזְ ִּכיר‬. ַ Whatever the original reading, this statement correlates with the thematic position of the divine-name laws in the final apodictic laws when lined up with the themes in the admonitions and exhortations in Hammurabiʼs epilogue, and it is the only thematic element that is found in all three stretches of apodictic legislation in CC (i.e., the initial apodictic laws of 20:23–26 and the two strings of the final apodictic laws in 22:20–30 and 23:9–19). The three instances of this motif collectively allow for the use of YHWHʼs name in legitimate cultic contexts (20:24), prohibit the use of the names of other gods (23:13), and prohibit cursing the deity (22:27). What is missing is a prohibition of the misuse of YHWHʼs name. Leviticus 24 supplies this. Leviticus 24 used the larger context of CCʼs talion laws to offer this supplement. It took the motif of men fighting from the beginning of CCʼs miscarriage law, to which CCʼs talion law is attached, to set up the context for blasphemy: Exodus 21:22

]. . .[ ‫יה‬ ָ ‫וְ ִכי־יִ נָּ צוּ ֲאנָ ִשׁים וְ נָ גְ פוּ ִא ָשּׁה ָה ָרה וְ יָ ְצאוּ יְ ָל ֶד‬

Leviticus 24:10–11 ‫ן־ה ִא ָשּׁה‬ ָ ‫׃ וַ יִּ קֹּב ֶבּ‬ ‫] וַ יִּ נָּ צוּ ַבּ ַמּ ֲחנֶ ה ֶבּן ַהיִּ ְשׂ ְר ֵא ִלית וְ ִאישׁ ַהיִּ ְשׂ ְר ֵא ִלי‬. . .[ ‫ן־א ָשּׁה יִ ְשׂ ְר ֵא ִלית‬ ִ ‫וַ יֵּ ֵצא ֶבּ‬ ]. . .[ ‫ת־ה ֵשּׁם וַ יְ ַק ֵלּל‬ ַ ‫ַהיִּ ְשׂ ְר ֵא ִלית ֶא‬

Source Dependence and the Development of the Pentateuch

665

Leviticus 24 took over CCʼs verb ‫ יִ נָּ צוּ‬and put it in narrative tense, ‫וַ יִּ נָּ צוּ‬.27 Though this motif is used to indicate a context of inadvertence in CC (discussed above), Lev 24 does not use it thus. Quite the contrary; though a fight, the speaker has full control over what comes out of his mouth and is capitally liable for it. Leviticus 24 imitated CC further in the next verb, which describes the egregious behavior. In the miscarriage law, the men fight (‫ )יִ נָּ צוּ‬and then knock (‫)וְ נָ גְ פוּ‬ a pregnant woman. In Lev 24, the men fight (‫ )וַ יִּ נָּ צוּ‬and then the half-Israelite man pronounces (‫ )וַ יִּ קֹּב‬the name in a curse.28 Leviticus 24ʼs verb *‫ נקב‬is phonologically similar to *‫( נגף‬nasal dental, palatal stop, and labial). The verb describing the misdeed in the miscarriage law appears to have prompted the verb chosen for describing the blaspheming of the deityʼs name in Lev 24.29 This phenomenon may be called paronymic replacement or echoing.30 That this is an actual hermeneutical technique is indicated by the very next pericope in the H legislation, its revision of CCʼs seventh-year law:  Nihan, “Murder” (see n. 4), 227–228, and idem, “Révisions” (see n. 4), 151, notes the verbal correspondence. It is the basis of his argument that the pericope is a unified composition. Deut 25:11–12, which also uses the verb, reads like an alternate road developing partly from CCʼs miscarriage law, answering what happens if the woman assaults one of the men in some way. But Dʼs law is also similar to Middle Assyrian Laws A 8 (see Wright, Inventing [see n. 1], 110–115). 28  On the meaning of *‫נקב‬, see C havel , Oracular Law (see n. 4), 27–33, 64; H utton , “Blasphemer Revisited” (see n. 23), 534–536; Milgrom, Leviticus 23–27 (see n. 4), 2107– 2109; Nihan, “Murder” (see n. 4), 216. It does not seem to have a negative sense (cf. Isa 62:2 and Num 1:17; Ezra 8:20; 1 Chr 12:32; 16:41; 2 Chr 28:15; 31:19) and may not be connected with *‫קבב‬, “curse.” But contextually and based on sources, the case of pronouncing the name (Lev 24:16) is more severe than cursing (v. 15). See the discussion below. 29  The verb *‫ נגף‬occurs also in CCʼs law on an ox goring an ox (21:35–36; this is one of two verses on animal death with the verb *‫ שלם‬that also has the talion formula for replacement; see also v. 37, with the verb *‫טבח‬, “slaughtering”). As a verb describing assault for humans and animals, *‫ נגף‬is conceptually parallel to the verb ‫ה ָכּה‬, ִ “strike,” that is used in the talion laws of Lev 24:17–18, 21. Doing the action of *‫ נקב‬to the divine name plays against doing *‫ נגף‬to a person or animal. 30  This may be included under Levinsonʼs notion of “lemmatic transformation,” where individual elements in a source text are selected out and transformed quite apart from the original context. See B. M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (Oxford: Oxford University Press, 1998), 36–38. In paronymic replacement, the lemmatic element is abstracted and used to adduce or generate another quite unrelated term and idea. It is more evocative allusion than redefinition or reinterpretation. Fishbane, Biblical Interpretation, 451, refers to “paranomastic associations or elaborations” and gives examples from elsewhere in the Bible (e.g., ‫ ָשׁ ֵקד‬/ ‫ שׁ ֵֹקד‬in Jer 1:11–12; his index, p. 575, lists the term “paranomastic exegesis,” where he refers readers also to his entry “puns”). The phenomenon of name giving where names and verbal roots are paronymically but not, in the strict sense, linguistically ִ ִ‫ ָ;קנ‬Gen 4:1) can be considered part of this broader exegetical and related (e.g., ‫ ַקיִ ן‬and ‫יתי‬ conceptual phenomenon. 27

666

David P. Wright

Exodus 23:10–11 ‫אָכלוּ ֶא ְבי ֹנֵ י ַע ֶמָּך‬ ְ ְ‫יעת ִתּ ְשׁ ְמ ֶטנָּ ה וּנְ ַט ְשׁ ָתּהּ ו‬ ִ ‫׃ וְ ַה ְשּׁ ִב‬ ‫בוּאָתהּ‬ ָ ‫ת־תּ‬ ְ ‫אָס ְפ ָתּ ֶא‬ ַ ְ‫ת־אַר ֶצָך ו‬ ְ ‫וְ ֵשׁשׁ ָשׁנִ ים ִתּזְ ַרע ֶא‬ ‫׃‬ ‫יתָך‬ ֶ ֵ‫ן־תּ ֲע ֶשׂה ְל ַכ ְר ְמָך ְלז‬ ַ ‫אכל ַחיַּ ת ַה ָשּׂ ֶדה ֵכּ‬ ַ ֹ ‫וְ יִ ְת ָרם תּ‬ Leviticus 25:3–4 ‫יעת ַשׁ ַבּת ַשׁ ָבּתֹון‬ ִ ‫וּב ָשּׁנָ ה ַה ְשּׁ ִב‬ ַ ‫׃‬ ‫בוּאָתהּ‬ ָ ‫ת־תּ‬ ְ ‫אָס ְפ ָתּ ֶא‬ ַ ְ‫ֵשׁשׁ ָשׁנִ ים ִתּזְ ַרע ָשׂ ֶדָך וְ ֵשׁשׁ ָשׁנִ ים ִתּזְ מֹר ַכּ ְר ֶמָך ו‬ ‫׃‬ ‫אָרץ ַשׁ ָבּת ַליהוָ ה ָשׂ ְדָך לֹא ִתזְ ָרע וְ ַכ ְר ְמָך לֹא ִתזְ מֹר‬ ֶ ‫יִ ְהיֶ ה ָל‬

Leviticus 25:2–7 draws from and builds on the wording and form of Exod 23:10–11, but replaces the root *‫ שמט‬in CC (Exod 23:11) with the similar sounding root *‫( שבת‬Lev 25:2, 4, 5, 6; post-alveolar fricative, labial, dental stop). Note in particular that Hʼs new root appears in the same syntactic location as CCʼs *‫שמט‬, after the adjective ‫יעת‬ ִ ‫( ַה ְשּׁ ִב‬Lev 25:4; Exod 23:11). H also appears to have been influenced by Dʼs seventh-year law. D for its part took up the root *‫ שמט‬from CC but used it repeatedly as it redefined its meaning (Deut 15:1, 2 [3×], 3, 9). Hʼs multiple uses (seven times) of the root *‫ שבת‬appears to reflect Dʼs multiple uses of *‫שמט‬. The repetition allowed H to further redefine the nature of the seventh year and imbue it with Sabbath ideology.31 Two other cases of paronymic replacement in Lev 24, which will be described later on in this essay, will make the association of *‫ נקב‬and *‫ נגף‬appear more compelling and indicate that this is, in fact, a hermeneutical technique in H.32 Leviticus 24ʼs blasphemy laws also have a broad structural correlation with CCʼs miscarriage legislation. The Covenant Code has two miscarriage laws, one in which only the fetus dies but the woman is uninjured (Exod 21:22, labeled I in the accompanying cited texts) and the second where the woman dies or is injured (v. 23–25, labeled II). Leviticus 24:15–16 imitates this double-law structure. The first law (24:15, labeled I) deals with cursing (*‫ )קלל‬the deity, based on the prohibition in Exod 22:27, as noted earlier. The second concerns pronouncing (*‫ )נקב‬the name YHWH (24:16, labeled II). This law, as observed earlier, fills the gap in CC about wrongly uttering YHWHʼs name and inversely echoes the content of the prohibition against speaking the names of other deities in Exod 23:13. But this law also has a connection with the miscarriage law in using a root that is alliteratively similar to CCʼs *‫נגף‬, as noted just above. The correlation with CCʼs miscarriage law continues further in the placement of the talion law  For the influence of CC and D in the seventh-year law of Lev 25, see J. S tackert , Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007), 117–128, 137; Chavel, Oracular Law (see n. 4), 49–54. For the replacement of *‫ שמט‬with *‫שבת‬, see Stackert, Rewriting, 120–121. Perhaps H makes this change partly because CCʼs “Sabbath” law, which uses the verb *‫שבת‬, immediately follows its seventh-year law (Exod 23:10–12). 32  Here in passing it may be mentioned that this phenomenon makes one wonder if the verb ‫ח ֵּלל‬, ִ “desecrate,” used of the deityʼs name in other H passages (Lev 18:21; 19:12; 20:3; 21:6; 22:2, 32), is partly echoing CCʼs verb ‫ ִק ֵּלל‬in Exod 22:27 in connection with passages about speaking the names of gods (Exod 20:24; 23:14, discussed above). CC uses ‫ ִח ֵּלל‬in another context (but of a sanctum) in Exod 20:25. See Chavel, Oracular Law (see n. 4), 88–92. 31

Source Dependence and the Development of the Pentateuch

667

after the double law structure, the difference being that in CC the talion list is the apodosis of the second miscarriage law (Exod 21:23a–25), whereas in Lev 24 the talion laws are juxtaposed to the two blasphemy laws (Lev 24:17–22): Leviticus 24:10–11, 15–16

Exodus 21:22–23

]. . .[ ‫] וַ יִּ נָּ צוּ ַבּ ַמּ ֲחנֶ ה‬. . .[

‫וְ ִכי־יִ נָּ צוּ ֲאנָ ִשׁים‬

]. . .[ ‫ת־ה ֵשּׁם‬ ַ ‫] ֶא‬. . .[ ‫וַ יִּ קֹּב‬

]. . .[ ‫וְ נָ גְ פוּ ִא ָשּׁה ָה ָרה‬

‫ֹלהיו‬ ָ ‫] ִאישׁ ִאישׁ ִכּי־יְ ַק ֵלּל ֱא‬. . .[ (I)

‫( וְ לֹא יִ ְהיֶ ה אָסֹון‬I)

‫  וְ נָ ָשׂא ֶח ְטאֹו‬

‫  ענֹושׁ יֵ ָענֵ שׁ ַכּ ֲא ֶשׁר יָ ִשׁית‬ ָ ]. . .[ ‫  ע ָליו ַבּ ַעל ָה ִא ָשּׁה‬ ָ

‫( וְ נ ֵֹקב ֵשׁם־יְ הוָ ה‬II)

‫( וְ ִאם־אָסֹון יִ ְהיֶ ה‬II)

]. . .[ ‫יוּמת‬ ָ ‫  מֹות‬

]. . .[ ‫  וְ נָ ַת ָתּה‬

talion appears here, as a separate law in vv. 17–21

talion appears here, as part of apodosis of second law in vv. 23b–25

The penalties of Lev 24ʼs two blasphemy laws have textual stimuli. The cursing prohibition in CC that lies behind Lev 24ʼs blasphemy law actually consists of a double prohibition: ‫ֹלהים לֹא ְת ַק ֵלּל וְ נָ ִשׂיא ְב ַע ְמָּך לֹא ָתאֹר‬ ִ ‫( ֱא‬Exod 22:27). The root *‫ נשא‬in the second prohibition, which comes after the prohibition against cursing the deity, which was influential in Lev 24:15, may have motivated use of the PH idiom about bearing sin with the verb *‫נשא‬. This would be another example of paronymic replacement, though it involves using different forms and senses of the same root rather than different but phonologically similar roots.33 It is similar to the case posited for *‫ נקב‬and *‫ נגף‬in the reflection of sequential motifs from CC (‫ יִ נָּ צוּ‬followed by ‫ וְ נָ גְ פוּ‬in Exod 21:22 // ‫ וַ יִּ נָּ צוּ‬followed by ‫ וַ יִּ קֹּב‬in Lev 24:10–11; ‫] ְת ַק ֵלּל‬. . .[ ‫ֹלהים‬ ִ ‫ ֱא‬followed by ‫ נָ ִשׂיא‬in Exod 22:27 // ‫ֹלהיו‬ ָ ‫יְ ַק ֵלּל ֱא‬ followed by ‫ וְ נָ ָשׂא ֶח ְטאֹו‬in Lev 24:15), and similar to the example of paronymic replacement in Lev 25 (‫בוּאָתהּ‬ ָ ‫ת־תּ‬ ְ ‫אָס ְפ ָתּ ֶא‬ ַ ְ‫ת־אַר ֶצָך ו‬ ְ ‫ וְ ֵשׁשׁ ָשׁנִ ים ִתּזְ ַרע ֶא‬followed by ָ ‫ת־תּ‬ ְ ‫אָס ְפ ָתּ ֶא‬ ַ ְ‫] ו‬. . .[ ‫ֵשׁשׁ ָשׁנִ ים ִתּזְ ַרע ָשׂ ֶדָך‬ ‫יעת ִתּ ְשׁ ְמ ֶטנָּ ה‬ ִ ‫ וְ ַה ְשּׁ ִב‬in Exod 23:10–11 // ‫בוּאָתהּ‬ followed by ‫יעת ַשׁ ַבּת ַשׁ ָבּתֹון‬ ִ ‫וּב ָשּׁנָ ה ַה ְשּׁ ִב‬ ַ in Lev 25:3–4). I will return shortly to the meaning of the penalty of “bearing sin.” The death penalty and the participial form that sets up the law in Lev 24:16 were inspired by the form and content of capital crimes in CC (Exod 21:12, 15–16; cf. 22:18):34  Part of the stimulus for using the bearing-of-sin idiom may have been CCʼs first miscarriage law: ‫( ָענֹושׁ יֵ ָענֵ שׁ ַכּ ֲא ֶשׁר יָ ִשׁית ָע ָליו ַבּ ַעל ָה ִא ָשּׁה‬Exod 21:22). Here, placing a penalty on the person is conceptually similar to his bearing a sin. A proverb uses language that mirrors these two idioms: ‫ל־ח ָמה נ ֵֹשׂא עֹנֶ שׁ‬ ֵ ‫( גְּ ָד‬Prov 19:19, MT Qĕre). 34  N ihan , “Murder” (see n. 4), 228, says that the parent-cursing law in Exod 21:17 was particularly influential because of its collocation of cursing and penalty (see also C havel , 33

668 Leviticus 24:16 Exodus 21:12 Exodus 21:16 Exodus 21:17 Exodus 22:18

David P. Wright

‫יוּמת‬ ָ ‫וְ נ ֵֹקב ֵשׁם־יְ הוָ ה מֹות‬ ָ ‫ַמ ֵכּה ִאישׁ וָ ֵמת מֹות‬ ‫׃‬ ‫יוּמת‬ ָ ‫וּמ ָכרֹו וְ נִ ְמ ָצא ְביָ דֹו מֹות‬ ְ ‫וְ גֹנֵ ב ִאישׁ‬ ‫׃‬ ‫יוּמת‬ ‫׃‬ ‫יוּמת‬ ָ ‫אָביו וְ ִאמֹּו מֹות‬ ִ ‫וּמ ַק ֵלּל‬ ְ ָ ‫ם־בּ ֵה ָמה מֹות‬ ְ ‫ָכּל־שׁ ֵֹכב ִע‬ ‫׃‬ ‫יוּמת‬

Both texts set up the case with a participial form followed by the death-penalty formula. This second blasphemy law is followed immediately by the homicide- and animal-killing law (Lev 24:17–18). Recall that vv. 17–18 were also influenced by CCʼs capital crime laws, as described earlier in this essay. The influence of CCʼs capital crimes extends across the two discrete legal sections of Lev 24 (the blasphemy laws and talion laws). This, along with the double-law structure ending in talion as found in CCʼs miscarriage laws, joins them compositionally and conceptually. Thus two double-law structures have been influential in Lev 24:15–16: the two separated but structurally parallel laws about cursing or speaking the names of deities in Exod 22:27 and 23:13 and the two miscarriage laws of Exod 21:22–25. These correlations focus attention and provide data for thinking about the meaning and relationship of the two blasphemy laws in Lev 24:15–16. Do they reflect an increase in severity as found in the two miscarriage laws of Exod 21:22 and 23–25? Or, are the laws independent and parallel, like the prohibitions against cursing God and speaking the names of other deities in Exod 22:27 and 23:13? An argument can be made that Lev 24ʼs second law about pronouncing the name represents a greater crime than the first law apart from its correlations with CCʼs miscarriage laws. This is mainly visible in the penalties. Bearing sin as in Lev 24:15 is a rather standard way by which PH describes guilt, short of specifying a punishment.35 Some cases of sin bearing are rectifiable, but, as Oracular Law [see n. 4], 55–58; Milgrom, Leviticus 23–27 [see n. 4], 2118). Note, however, that Lev 24:15–16 puts the cursing and death-penalty phrases in separate blasphemy laws. That homicide has the ‫יוּמת‬ ָ ‫ מֹות‬penalty in Exod 21:12 and that Lev 24 is primarily drawing an analogy of blasphemy to homicide means that a specific tie to Exod 21:17 against other capital laws in CC is not required. It is unclear, too, as Nihan claims, that the order homicide then cursing (parents) in Exod 21:12–17 is connected to the reverse order blasphemy then homicide in Lev 24:15–17. 35  See B. J. S chwartz , “The Bearing of Sin in the Priestly Literature,” in Pomegranates

Source Dependence and the Development of the Pentateuch

669

Schwartz has noted, bearing sin is often used of “sins from which one can never be disencumbered [. . .]. It is a permanent state.”36 In such cases, the individual would be liable to the penalty of excision (*‫)כרת‬, which is early death and extinction of posterity by divine agency.37 The connection between bearing transgression and excision is found in Lev 19:8; 20:7; Num 9:13. A punishment similar to excision is actually described in CCʼs apodictic laws a few verses before its cursing law. Oppressing the weak members of society, including widows and orphans, arouses divine wrath, such that, measure for measure, the deity will make the oppressorsʼ wives widows and their children orphans (Exod 22:22–23). The consequence of Lev 24ʼs sin bearing in the PH context is essentially an application of the penalty in Exod 22:22–23 to v. 27. In contrast to the implied punishment by the deity in the first blasphemy law, the second law (Lev 24:16) invokes capital punishment, which is administered through the human court system.38 This is a more immediate and, hence, more serious form of punishment. That H thinks about these two forms of punishment in an interrelated and systematic way is attested by the laws on Molech worship in Lev 20:2–5. The people are to apprehend the transgressor and put him to death. But if they do not do this, the deity at least will effect excision. The rhetorical ordering in this passage is the reverse of the blasphemy laws, with the penalty that reflects the gravity of the case stated first. An increase in severity in the second law is also evident in the description of the deity as object of the blasphemy. The first law repeats the title or common noun for deity from CC (‫ֹלהיו‬ ָ ‫ ֱא‬// ‫ֹלהים‬ ִ ‫)א‬. ֱ The second blasphemy law involves the proper name YHWH. If nothing else, this sharpens the conceptual focus, moving from periphrasis to identification. But this formulation is not merely rhetorical. Only in the second law is the divine name actually uttered. This represents a concrete step beyond cursing the deity in general. This correlates with a broader emphasis in H on the sanctity of the deityʼs name or reputation and how this must be protected against desecration.39 An increase in severity is also indicated by the logical relationship of the two laws. If the verb *‫ נקב‬is neutral in its meaning, meaning simply “pronounce,” the law logically cannot be referring to expressing the divine name in any context. and Golden Bells: Studies in Biblical, Jewish, and Near Eastern Ritual, Law, and Literature in Honor of Jacob Milgrom (ed. D. P. Wright et al.; Winona Lake, IN: Eisenbrauns, 1995), 3–21. 36  See Schwartz, “Bearing of Sin” (see n. 35), 15. 37  See J. Milgrom, Leviticus 1–16 (AB 3; New York: Doubleday, 1991), 457–460. For the correlation of bearing sin and excision, see Milgrom, Leviticus 23–27 (see n. 4), 2117. 38  J. Milgrom, Studies in Levitical Terminology, I (University of California Publications, Near Eastern Studies 14; Berkeley: University of California Press, 1970), 5–8. 39  For the nameʼs holiness, see Lev 20:3; 22:2, 32; for its desecration, Lev 18:21; 19:12; 20:3; 21:6; 22:2, 32. The name is an abstraction for the divine reputation, and the concept is related to the sanctification of the deity (cf. Exod 29:43; Lev 10:3; Num 20:12, 13). See Chavel, Oracular Law (see n. 4), 88–92; Kamionkowski, “Leviticus 24,10–23” (see n. 23).

670

David P. Wright

PH allows the use of the divine name in various ritual performances (e.g., Num 5:19–22; 6:23–27).40 For this reason some argue that the motif of cursing from the first law (v. 15b) is to be carried over into the second law. The narrative appears to reflect this understanding. The actions of the verbs from both laws are attributed to the blasphemer: ‫ת־ה ֵשּׁם וַ יְ ַק ֵלּל‬ ַ ‫ן־ה ִא ָשּׁה ַהיִּ ְשׂ ְר ֵא ִלית ֶא‬ ָ ‫( וַ יִּ קֹּב ֶבּ‬Lev 24:11). The last verb (cursing) may be read as the context in which the utterance of the name occurred. That the individual is called a ‫ ְמ ַק ֵלּל‬elsewhere also points to this as the underlying transgression while pronouncing the name (vv. 14, 23).41 Thus the laws might be rendered as follows: Leviticus 24:15b–16a Any person who curses his god shall bear his sin, and one who (also) speaks the name of YHWH shall be put to death.42 40  See D. P. Wright, “Ritual Speech in the Priestly-Holiness Prescriptions of the Pentateuch and Its Near Eastern Context,” in Mesopotamia in the Ancient World: Impact, Continuities, Parallels (ed. R. Rollinger and E. van Dongen; Melammu Symposia 7; Münster: Ugarit-Verlag, 2015, 107–123); idem, “Law and Creation” (see n. 3), 81–82. 41  For this argument, see Milgrom, Leviticus 23–27 (see n. 4), 2117–2118. A question that arises in an argument about seeing a context of cursing in 24:16a is whether the law entails a broader context for the misuse of the divine name, such as making false oaths. In this regard it may be asked to what extent Exod 20:7 influenced Lev 24:16. (The Decalogue would have been in the text or source that H used in accessing CC.) There is no clear sign that the Decalogue rule was influential in Lev 24. Hʼs correlate of this injunction is Lev 19:12. See Milgrom, Leviticus 17–22 [see n. 21], 1633; Weingreen, “Case of the Blasphemer” [see n. 4], 121–122. Relevant to the how the narrative description in Lev 24 relates to the blasphemy law is the question whether the blasphemerʼs cursing in the narrative was of the deity or of his opponent. The context of the fight in vv. 10–11 suggests that the case involves cursing the opponent by invoking the deity or his name (cf. Josh 6:26; 2 Sam 16:8; 2 Kings 2:24). But CCʼs cursing law (Exod 22:27), the wording of the version of this in Lev 24:15, and passages such as 1 Kgs 21:10; Job 1:5; 2:9 (see also LXX 1 Sam 3:13; Weingreen, “Case of the Blasphemer” [see n. 4], 120) indicate that cursing the deity directly is intended. For discussion see M ilgrom, Leviticus 23–27 (see n. 4), 2107–2109. 42  Nihan, “Murder” (see n. 4), 218–219, and idem, “Révisions” (see n. 4), 144–146, sees an increase in the severity of the laws differently. For him, “his deity” of the first law (v. 15) refers to the respective gods of the Israelite or an immigrant. Hence the penalty is nonspecific and relative to the case. The second law, however, refers specifically to misspeaking YHWHʼs name, and this expressly carries the death penalty. This interpretation is difficult, however, because the term for deity with suffix pronoun (‫ֹלהיו‬ ָ ‫)א‬ ֱ appears elsewhere in PH clearly referring to YHWH (Lev 4:22; 21:7, 12, 17, 21, 22; Num 6:7; see Milgrom, Leviticus 23–27 [see n. 4], 2115, who observes that the term ‫ֹלהים‬ ִ ‫ ֱא‬is chosen in part to echo Exod 22:27). The pronoun seems to mark intimacy or obligatory reverence (i.e., “the God that he should venerate”). The term with suffixes of other persons may relate to this, especially Lev 19:12 and 21:6, which warn against profaning the name of ‫ֹלהיָך‬ ֶ ‫ ֱא‬or ‫יהם‬ ֶ ‫ֹלה‬ ֵ ‫א‬. ֱ These refer to YHWH. Moreover, it is not clear why Lev 24 would proscribe the cursing of other gods, how this relates to a notion of sacrality that may inform the law in Nihanʼs view, or why the immigrant should face possible excision by YHWH (the consequence of unremedied sin bearing). Otherwise on the intensi-

Source Dependence and the Development of the Pentateuch

671

Consideration of CCʼs miscarriage laws gives further weight to the argument that the second transgression is more grave than the first. The influence of the miscarriage law and its context in other respects is undeniable, in the verb ‫וַ יִּ נָּ צוּ‬, the postpositioned talion law, and the formulation of Lev 24:16a after the pattern of CCʼs homicide laws, to which the miscarriage law is a complement. There also seems to be a connection with CCʼs verb *‫ נגף‬used in the miscarriage law. It is reasonable to think that the climaxing structure of the miscarriage laws was also influential. If this is the case, it may require reconsidering the force of the verb *‫ נקב‬intended by H. Even though, from modern linguistic analysis, the verb is not connected with *‫קבב‬, “to curse,”43 perhaps H drew color from the latter root. Perhaps H chose *‫ נקב‬both because of its similarity to *‫ נגף‬and its resonance with *‫קבב‬.

The Influence of Deuteronomy Some motifs in the blasphemy-talion pericope in Lev 24 correlate in a thematically cogent way with laws in D and thus indicate that the latter may also have been a source used in composition. This accords with recent studies that have argued that H in other places has responded to D.44 Some of the H passages where this is visible have a proximate or thematic relationship to Lev 24 and, like that chapter, display dual source influence from both CC and D. For example, the slavery laws of Lev 25:39–46, in the chapter that follows, build on CCʼs debtslave laws of Exod 21:2–6 with intersecting influence from Deut 15:12–18.45 This complements the dependence of the seventh-year laws of the same chapter fication of legal cases, see Chavel, Oracular Law (see n. 4), 64 (he sees a type of formulaic intensification but “when all is said and done, there is no difference between cursing the deity and exclaiming his name”); Kamionkowski, “Leviticus 24,10–23” (see n. 23), 79–80. 43  Num 22:11, 17; 23:8, 11, 13, 25, 27; 24:10; Job 3:8; Prov 11:26; 24:24. 44  See S tackert , Rewriting (see n. 31), 5–12, for a summary of literature and general arguments for Hʼs dependence on D. His book otherwise provides an extended argument that H knew D or parts thereof (and see the list of correspondences on pp. 7–8 and n. 18). Nihan, “Murder” (see n. 4), 237, says that “there is now a growing consensus that H presupposes [. . .] most of the legislation found in Deuteronomy (Deut 12–26).” 45  S tackert , Rewriting (see n. 31), 113–164; B. M. L evinson , “The Manumission of Hermeneutics: The Slave Laws of the Pentateuch as a Challenge to Contemporary Pentateuchal Theory,” in Congress Volume: Leiden, 2004 (ed. A. Lemaire; VTSup 109, Leiden: Brill, 2006), 281–324; idem, “The Birth of the Lemma: The Restrictive Reinterpretation of the Covenant Codeʼs Manumission Law by the Holiness Code (Leviticus 24:44–46),” JBL 124 (2005), 617–639, here 630–633, 635–636. Hʼs grand program of a seventh-year sabbatical plus a release of debts in the jubilee year are a reconfiguration of CCʼs laws on debt-slavery and seventh-year leaving of produce for the poor (Exod 21:2–6; 23:10–11) and Dʼs debt-release and debt-slave laws of Deut 15.

672

David P. Wright

(Lev 25:2–7) on both CC and D, as noted earlier.46 The thematically related laws about homicide asylum in Num 35:9–34 develop the motif of inadvertent homicide from CC (Exod 21:12–14) and articulate the PH view of city asylum in response to Deut 19:1–13.47 In general, H interacted with CC and D differently. It cites or alludes to CC in a more visible manner, as we have seen in Lev 24. This indicates that Hʼs primary hermeneutic interest was in CC, as an established authoritative or prestige text. In contrast, H reflects D in a more incidental or oblique fashion. This gives the impression that it esteemed D differently, perhaps more as a recent competing interpretation of CC than as an authority. Lev 24 has correlations with Dʼs two blocks of laws that deal with witnesses. These include Dʼs main witness laws in Deut 19:15–21,48 which conclude with Dʼs version of the talion law, and 17:2–7, an apostasy law that includes a witness law similar to 19:15. The main witness laws of 19:15–21 set out the requirement of two or more witnesses in cases of accusation, followed by a law about giving false witness: Deuteronomy 19:15–21 ‫ל־פּי‬ ִ ‫ל־פּי ְשׁנֵ י ֵע ִדים אֹו ַע‬ ִ ‫ל־ח ְטא ֲא ֶשׁר יֶ ֱח ָטא ַע‬ ֵ ‫ל־ח ָטּאת ְבּ ָכ‬ ַ ‫וּל ָכ‬ ְ ‫ל־עֹון‬ ָ ‫ל ֹא־יָ קוּם ֵעד ֶא ָחד ְבּ ִאישׁ ְל ָכ‬ ָ ‫׃ ִכּי־יָ קוּם ֵע‬ ‫ה־ע ִדים יָ קוּם ָדּ ָבר‬ ֵ ‫ֹלשׁ‬ ָ ‫ְשׁ‬ ‫ר־ל ֶהם‬ ָ ‫י־ה ֲאנָ ִשׁים ֲא ֶשׁ‬ ָ ֵ‫׃ וְ ָע ְמדוּ ְשׁנ‬ ‫ד־ח ָמס ְבּ ִאישׁ ַל ֲענֹות בֹּו ָס ָרה‬ ‫ד־שׁ ֶקר‬ ֶ ‫יטב וְ ִהנֵּ ה ֵע‬ ֵ ‫׃ וְ ָד ְרשׁוּ ַהשּׁ ְֹפ ִטים ֵה‬ ‫ָה ִריב ִל ְפנֵ י יְ הוָ ה ִל ְפנֵ י ַהכּ ֲֹהנִ ים וְ ַהשּׁ ְֹפ ִטים ֲא ֶשׁר יִ ְהיוּ ַבּיָּ ִמים ָה ֵהם‬ ִ ‫אָחיו‬ ִ ‫יתם לֹו ַכּ ֲא ֶשׁר זָ ַמם ַל ֲעשֹׂות ְל‬ ֶ ‫׃ וַ ֲע ִשׂ‬ ‫אָחיו‬ ִ ‫ָה ֵעד ֶשׁ ֶקר ָענָ ה ְב‬ ‫אָרים‬ ִ ‫׃ וְ ַהנִּ ְשׁ‬ ‫וּב ַע ְר ָתּ ָה ָרע ִמ ִקּ ְר ֶבָּך‬ ‫׃ וְ לֹא ָתחֹוס ֵעינֶ ָך נֶ ֶפשׁ ְבּנֶ ֶפשׁ ַעיִ ן ְבּ ַעיִ ן‬ ‫יִ ְשׁ ְמעוּ וְ יִ ָראוּ וְ לֹא־י ִֹספוּ ַל ֲעשֹׂות עֹוד ַכּ ָדּ ָבר ָה ָרע ַהזֶּ ה ְבּ ִק ְר ֶבָּך‬ ‫׃‬ ‫ֵשׁן ְבּ ֵשׁן יָ ד ְבּיָ ד ֶרגֶ ל ְבּ ָרגֶ ל‬

Lev 24 has several correlations that point to a genetic connection between the texts. Both texts set out measure-for-measure corporal punishment described ִ ‫יתם לֹו ַכּ ֲא ֶשׁר זָ ַמם ַל ֲעשֹׂות ְל‬ ֶ ‫( וַ ֲע ִשׂ‬19:19). with similar language. D prescribes ‫אָחיו‬ Leviticus 24:19–20, in two parallel phrases, prescribes ‫וְ ִאישׁ ִכּי־יִ ֵתּן מוּם ַבּ ֲע ִמיתֹו‬ ‫ ַכּ ֲא ֶשׁר ָע ָשׂה ֵכּן יֵ ָע ֶשׂה לֹּו‬and ‫אָדם ֵכּן יִ נָּ ֶתן בֹּו‬ ָ ‫כּ ֲא ֶשׁר יִ ֵתּן מוּם ָבּ‬. ַ 49 Both describe equally  See at n. 31, above.  Stackert, Rewriting (see n. 31), 31–112; Nihan, “Murder” (see n. 4), 227 n. 52, 240. Besides expanding the system of cities as places of asylum, which is a development from Deut 19 and which itself is a response to CC, Num 35:9–34 reflects CC in its use of X-‫ ַמ ֵכּה‬as technical term for the perpetrator (Exod 21:12 // Num 35:11, 15, 30), the ‫יוּמת‬ ָ ‫ מֹות‬formula (Exod 21:12 // Num 35:16–18, 21, 31, five times reflecting Hʼs tendency toward motif repetition for emphasis), the instrument ‫( ְבּ ֶא ֶבן‬Exod 21:18 // Num 35:17), and the issue of ‫( כּ ֶֹפר‬Exod 21:30// Num 35:31–32). See also Stackert, Rewriting, 165–208, for the dependence of Hʼs tithe law (Num 18:20–32) on Dʼs (Deut 14:22–29). 48  The facts that the witness laws in Deut 19 come almost immediately after Dʼs homicide asylum laws (Deut 19:1–13) and that Dʼs asylum laws were influential on Hʼs asylum law in Num 35 (see n. 47) provide broader contextual support to the argument that Lev 24 knew Dʼs witness laws. 49  Nihan, “Revisions” (see n. 4), 147–48, observes briefly “Cette expression rappelle en partie celle déjà employée dans le TM de Dt 19,19a [. . .], et il est possible que la formulation de Lv 24 renvoie ici à celle de la loi de Dt 19,16–21” (also p. 150). See also the passing comments in N ihan, “Murder” (see n. 4), 225, 226 n. 40. He uses this as an opportunity to 46

47

Source Dependence and the Development of the Pentateuch

673

exacted retribution with a clause headed by the conjunction ‫ ַכּ ֲא ֶשׁר‬that likens what is to be exacted with what was done or planned. Both use the verb *‫עשה‬ to describe exacting the penalty. Both also refer to the victimized Israelite as oneʼs fellow, using terms favored by the respective corpora (‫אָחיו‬ ִ ‫ל‬,ְ ‫)בּ ֲע ִמיתֹו‬. ַ In addition and most prominently, of course, both passages buttress the required penalty with reference to the talion law or principle. Leviticus 24ʼs correlations with the apostasy law of Deut 17:2–7, which contains Dʼs other witness law (v. 6), bolster the impression of a textual connection between the two corpora. It must be observed first that Deut 17:2–7, for its part, builds on the apostasy laws of Deut 13 and, in particular, the second of three cases laid out there (13:7–12), which describes the apostasy and punishment of a family member or fellow Israelite.50 Our particular concern is the penalty in Deut 17:5–7, which fine tunes that of Deut 13:10–12: Deuteronomy 17:5–7 ‫ת־ה ִאישׁ‬ ָ ‫ל־שׁ ָע ֶריָך ֶא‬ ְ ‫ת־ה ָדּ ָבר ָה ָרע ַהזֶּ ה ֶא‬ ַ ‫ת־ה ִא ָשּׁה ַה ִהוא ֲא ֶשׁר ָעשׂוּ ֶא‬ ָ ‫ת־ה ִאישׁ ַההוּא אֹו ֶא‬ ָ ‫את ֶא‬ ָ ‫הֹוצ‬ ֵ ְ‫ו‬ ְ ‫ת־ה ִא ָשּׁה‬ ָ ‫אֹו ֶא‬ ‫יוּמת‬ ַ ‫יוּמת ַה ֵמּת ל ֹא‬ ַ ‫ֹלשׁה ֵע ִדים‬ ָ ‫ל־פּי ְשׁנַ יִ ם ֵע ִדים אֹו ְשׁ‬ ִ ‫׃ ַע‬ ‫וּס ַק ְל ָתּם ָבּ ֲא ָבנִ ים וָ ֵמתוּ‬ ‫׃‬ ‫וּב ַע ְר ָתּ ָה ָרע ִמ ִקּ ְר ֶבָּך‬ ִ ‫אַחר ֹנָ ה‬ ֲ ‫ל־ה ָעם ָבּ‬ ָ ‫׃ יַ ד ָה ֵע ִדים ִתּ ְהיֶ ה־בֹּו ָב ִראשׁ ֹנָ ה ַל ֲה ִמיתֹו וְ יַ ד ָכּ‬ ‫ל־פּי ֵעד ֶא ָחד‬ ִ ‫ַע‬ Deuteronomy 13:10–12 ‫וּס ַק ְלתֹּו ָב ֲא ָבנִ ים וָ ֵמת ִכּי‬ ְ ‫׃‬ ‫אַחר ֹנָ ה‬ ֲ ‫ל־ה ָעם ָבּ‬ ָ ‫ִכּי ָהר ֹג ַתּ ַה ְרגֶ נּוּ יָ ְדָך ִתּ ְהיֶ ה־בֹּו ָב ִראשֹׁונָ ה ַל ֲה ִמיתֹו וְ יַ ד ָכּ‬ ‫׃ וְ ָכל־יִ ְשׂ ָר ֵאל יִ ְשׁ ְמעוּ וְ יִ ָראוּן‬ ‫יאָך ֵמ ֶא ֶרץ ִמ ְצ ַריִ ם ִמ ֵבּית ֲע ָב ִדים‬ ֲ ‫מֹּוצ‬ ִ ‫ֹלהיָך ַה‬ ֶ ‫יחָך ֵמ ַעל יְ הוָ ה ֱא‬ ֲ ‫ִב ֵקּשׁ ְל ַה ִדּ‬ ‫׃‬ ‫א־יֹוספוּ ַל ֲעשֹׂות ַכּ ָדּ ָבר ָה ָרע ַהזֶּ ה ְבּ ִק ְר ֶבָּך‬ ִ ֹ ‫וְ ל‬

The penalties in both laws call for stoning and portray the execution operating in two stages: the “hand” of the witness initiates the stoning; the rest of the people then join in. Deuteronomy 17 inverts the literary order of motifs by prescribing the requirement of stoning first (v. 5), followed by delineation of the two stages of its execution (v. 7). This contrasts with Deut 13, which has the two stages described first (v. 10), followed by the prescription for stoning (v. 11). The inversion in Deut 17 was to facilitate the inclusion of the witness law of v. 6, which corresponds to the law about multiple witnesses in Deut 19:15 (cited above). Deuteronomy 17:7, which follows the witness law, can thus specify that the agents of the first stage of execution are, specifically, the witnesses: ‫יַ ד ָה ֵע ִדים‬ ‫תּ ְהיֶ ה־בֹּו ָב ִראשׁ ֹנָ ה‬. ִ Another difference in Deut 17ʼs description of the penalty in contrast to that in Deut 13 is the requirement to bring the offender out to the gate of the town (‫ל־שׁ ָע ֶריָך‬ ְ ‫] ֶא‬. . .[ ‫ת־ה ִא ָשּׁה ַה ִהוא‬ ָ ‫ת־ה ִאישׁ ַההוּא אֹו ֶא‬ ָ ‫את ֶא‬ ָ ‫הֹוצ‬ ֵ ְ‫ו‬, v. 5).

observe that the literal interpretation of talion was already underway in Deut before Lev 24. He does not argue that Deut may have been an influence on Lev 24. He does not deal in detail with Deut 19 because he says “Deut entretient des rapports plus lointains avec ces deux passages” (“Révisions,” 125). See n. 58, below. 50  See L evinson , Deuteronomy (see n. 30), 98–143, especially 118–123, on how 17:2–7 hermeneutically transforms 13:7–12. For discussion of Deut 17:2–7, see B. Wells, The Law of Testimony in the Pentateuchal Codes (BZABR 4; Wiesbaden: Harrassowitz, 2004), 86–99.

674

David P. Wright

This is a requirement in other cases of execution in D (21:19; 22:24) though not found in Deut 13.51 The basic elements in the prescription for the execution of the blasphemer in Lev 24 correlate with Deut 17:2–7, the expanded apostasy law in D: Deuteronomy 17:5–7 ‫ת־ה ִאישׁ ַההוּא‬ ָ ‫את ֶא‬ ָ ‫הֹוצ‬ ֵ ְ‫ו‬ ‫ת־ה ִא ָשּׁה ַה ִהוא‬ ָ ‫אֹו ֶא‬ ‫ת־ה ָדּ ָבר ָה ָרע ַהזֶּ ה‬ ַ ‫ֲא ֶשׁר ָעשׂוּ ֶא‬ ‫ל־שׁ ָע ֶריָך‬ ְ ‫ֶא‬ ‫ת־ה ִא ָשּׁה‬ ָ ‫ת־ה ִאישׁ אֹו ֶא‬ ָ ‫ֶא‬  ‫וּס ַק ְל ָתּם ָבּ ֲא ָבנִ ים וָ ֵמתוּ׃‬ ְ ‫ֹלשׁה ֵע ִדים‬ ָ ‫ל־פּי ְשׁנַ יִ ם ֵע ִדים אֹו ְשׁ‬ ִ ‫ַע‬ ‫׃‬ ‫ל־פּי ֵעד ֶא ָחד‬ ִ ‫יוּמת ַע‬ ַ ‫יוּמת ַה ֵמּת לֹא‬ ַ ‫יַ ד ָה ֵע ִדים ִתּ ְהיֶ ה־בֹּו‬ ‫ָב ִראשׁ ֹנָ ה‬ ‫אַחר ֹנָ ה‬ ֲ ‫ל־ה ָעם ָבּ‬ ָ ‫וְ יַ ד ָכּ‬ ‫׃‬ ‫וּב ַע ְר ָתּ ָה ָרע ִמ ִקּ ְר ֶבָּך‬ ִ

Leviticus 24:14, 23 ‫ת־ה ְמ ַק ֵלּל‬ ַ ‫הֹוצא ֶא‬ ֵ ‫ל־מחוּץ ַל ַמּ ֲחנֶ ה‬ ִ ‫ֶא‬

‫יהם‬ ֶ ‫ל־השּׁ ְֹמ ִעים ֶאת־יְ ֵד‬ ַ ‫וְ ָס ְמכוּ ָכ‬ ‫ַעל־רֹאשֹׁו‬ ָ ‫וְ ָרגְ מוּ אֹתֹו ָכּ‬ ‫׃‬ ‫ל־ה ֵע ָדה‬ ]. . .[ ‫ת־ה ְמ ַק ֵלּל‬ ַ ‫יֹּוציאוּ ֶא‬ ִ ַ‫ו‬ ‫ל־מחוּץ ַל ַמּ ֲחנֶ ה‬ ִ ‫ֶא‬ ‫אָבן‬ ֶ ‫וַ יִּ ְרגְּ מוּ אֹתֹו‬

Both passages require removal of the offender to the outskirts of the habitation or beyond, using the verb ‫הֹוציא‬, ִ “take out.” The specific place to which the person is taken correlates with the geographical ideology of the respective corpora. Both require stoning of the offender. And, most interestingly, both have a two-staged execution where the hand or hands of the witnesses (“hearers” in Lev 24) are to be on the offender in some way, described with a prepositional phrase that involves the root *‫;על־רֹאשֹׁו( ראש‬ ַ ‫)ב ִראשׁ ֹנָ ה‬. ָ These similarities, along with those charted earlier between Lev 24 and Deut 19, point quite decisively to a genetic connection between H and D. The direction of that relationship is from D to Lev 24. First, with respect to Deut 19, Lev 24 is more elaborate and systematic. It integrates the talion equations into the body of its legislation. Its more elaborate formulation separates off the “life for life” principle from the other talion equations referring to nonlethal injuries and describes the distinctive cases pertaining to each, all as part of a carefully crafted quasi-chiastic structure:52 51  Apart from the relationship of Deut 17 and 13, the interrelationship of the witness laws being considered here is also evident in the deterrence clause (“all Israel/the rest will hear and fear, so they do not further do such an evil thing in your midst”) found in 13:12 and 19:20. The first apostasy law and main witness law also share the phrase “you shall remove evil from your midst” (13:6; 19:19). 52  I say and emphasize “quasi-” because it is not simply chiastic as many have claimed. For critiques, see D. P. Wright, “The Fallacies of Chiasmus,” ZABR 10 (2004), 143–168, here

Source Dependence and the Development of the Pentateuch

675

Leviticus 24:17–21 Case A:

A person who strikes a human being lethally shall be put to death,

Case B:

and one who strikes a beast lethally shall pay it back:

Supporting Principle:

life for life.

Case C:

A person who inflicts a wound on his fellow, as he has done so shall it be done to him:

Supporting Principle:

fracture for fracture, eye for eye, tooth for tooth.

Case C′:

As he inflicted a wound on a man, so shall it be inflicted on him.

Case B′:

But one who strikes a beast (lethally) shall pay it back,

Case A′:

and one who strikes a human being (lethally) shall be put to death.

In contrast, the set of talion equations in Deut 19:21 are appended as a supporting footnote without integration (though v. 21b follows perfectly after v. 19a) and look otherwise like they are solely drawn (with some paraphrase) from CC. More particularly, Lev 24 repeats the requirement for equal retribution in cases of injury (vv. 19, 20b). The second of these, which uses the verb *‫נתן‬, “give, place,” as opposed the first, which uses *‫עשה‬, “do, perform” – the verb of Deut 19:19 – appears to be formulated for clarification: the injuries are to be literally inflicted. In short, the belabored and emphatic formulation in Lev 24 appears secondary relative to D, and Dʼs formulation does not reflect the features of Lev 24. In regard to Deut 17, a linear model of textual development with Lev 24 as the end point, which exhibits a highly ideological and specialized portrayal of ritualized legal practice, makes the best sense. The apostasy law of Deut 13, which, as noted, lies behind Deut 17:2–7, is dependent in part on the Vassal Treaties of Esarhaddon (VTE) § 12 (lines 130–146):53 163–164, n. 37; Chavel, Oracular Law (see n. 4), 23–28, 68–72; idem, “Lev 24:10–23” (see n. 4), 2140–2145; idem, “ ‘Oracular Novellae’ and Biblical Historiography: Through the Lens of Law and Narrative,” Clio 39 (2009), 1–27, here 17. For other observations on the inverted structure, see Kamionkowski, “Leviticus 24,10–23” (see n. 23), 81; Leuchter, “Ambiguous Details” (see n. 4), 443; Milgrom, Leviticus 23–27 (see n. 4), 2128–2133; Nihan, “Murder” (see n. 4), 224–225; T. W illis , “Blasphemy, Talion, and Chiasmus: The Marriage of Form and Content in Lev 24,13–23,” Bib 90 (2009), 68–74. D. P. Wright, “Chiasmus in the Covenant Code Reconsidered” in “Gerechtigkeit und Recht zu üben” (Gen 18,10): Studien zur altorientalischen und biblischen Rechtsgeschichte, zur Religionsgeschichte Israels und zur Religionssoziologie; Festschrift für Eckart Otto (ed. R. Achenbach and M. Arneth; Wiesbaden: Harrassowitz, 2009), 171–181, tempers the extreme skepticism of my earlier critique of the phenomenon of chiasmus. In any case, though the talion law for injuries forms the center of the form in Lev 24, that does not mean it is the focal point thematically. The correlating elements across the entire form may articulate themes or points of equal importance. 53  On Dʼs reliance on VTE (better termed Esarhaddonʼs Succession Treaty), see

676

David P. Wright

If anyone should speak to you of rebellion and insurrection [. . .] you shall seize the perpetrators of insurrection and bring them before Assurbanipal [. . .]. If you are able to seize them and put them to death, then you shall destroy their name and their seed from the land. If, however, you are unable to seize them and put them to death, you shall inform Assurbanipal, the great crown prince designate, and assist him in seizing and putting to death the perpetrators of rebellion.54

The idiom of the hand being upon the offender (X-‫ )יָ ד ִּת ְהיֶ ה ב‬to put him to death in Deut 13:10 refers to the violence involved in prosecuting the execution.55 It is similar to and may be a rendering of VTEʼs description of seizing (ṣabātu) the traitor and putting him to death. Deuteronomy 17 replicates the hand-violence idiom from Deut 13 but adds the requirement that there be at least two witnesses. It is by virtue of the fact that they are witnesses that they begin the execution. Leviticus 24 developed this further. It creatively turned the idiom of a hand being on the offender (‫יַ ד ָה ֵע ִדים ִתּ ְהיֶ ה־בֹּו ָב ִראשׁ ֹנָ ה‬, Deut 17:7) into a ritualized hand placement gesture (‫יהם ַעל־רֹאשֹׁו‬ ֶ ‫ל־השּׁ ְֹמ ִעים ֶאת־יְ ֵד‬ ַ ‫וְ ָס ְמכוּ ָכ‬, Lev 24:14). It kept the same basic order of elements but introduced the verb *‫סמך‬, “place, lean,” used in other cases of PH ritual hand placement.56 Leviticus 24 replaced Dʼs “witnesses” with the more contextually fitting and specific “hearers.” Leviticus 24 changed the temporal adverb ‫ּב ִראשׁ ֹנָ ה‬, ָ “at first,” in 17:7 (also in 13:10) into a concrete locative adverbial phrase referring to the place where ַ “on his head.” This is again an example of the hands are placed, ‫על־רֹאֹשׁו‬, paronymic replacement, similar to the use of the verb *‫ נשא‬in Lev 24:15 based on the noun ‫ נָ ִשׂיא‬in Exod 22:27, noted above, where the same root is used but with an entirely different meaning. Although H could creatively reimage the meaning of Dʼs adverbial ‫ּב ִראשׁ ֹנָ ה‬, ָ it had no use for Dʼs complementary adverb ‫אַחר ֹנָ ה‬ ֲ ‫בּ‬, ָ “next, afterward,” that describes the second stage of execution when the rest, literally, “all,” of the people join in (‫ל־ה ָעם‬ ָ ‫כּ‬, ָ Deut 17:7; cf. 13:10). B. M. Levinson and J. Stackert, “The Limits of ‘Resonance’: A Response to Joshua Berman on Historical and Comparative Method,” JAJ 4 (2013), 310–333; idem, “Between the Covenant Code and Esarhaddonʼs Succession Treaty: Deuteronomy 13 and the Composition of Deuteronomy,” JAJ 3 (2012), 123–140. In my view, influence during the Neo-Assyrian period is the best hypothesis for explaining Akkadian literary motifs in CC and D. See Wright, “Origin” (see n. 11), 242–243. On dating CC to the Neo-Assyrian period, see Nihan, “Révisions” (see n. 4), 126–127, n. 14; 140. 54  Translation S. Parpola and K. Watanabe , Neo-Assyrian Treaties and Loyalty Oaths (SAA 2; Helsinki: Helsinki University Press, 1988), 34. More broadly, see VTE § 10 (lines 108–122) and § 26 (lines 302–317), as well as § 13 (lines 147–152), § 14 (lines 162–172), and § 57 (lines 500–506). 55  So Gen 37:27; Deut 2:15; Josh 2:19; Judg 2:15; 1 Sam 5:9; 7:13; 12:15; 18:17, 21; 24:13, 14; 2 Sam 24:17; Ezek 13:9; 1 Chr 21:17. 56  See D. P. Wright, “The Gesture of Hand Placement in the Hebrew Bible and in Hittite Literature,” JAOS 106 (1986), 433–446; cf. Nihan, “Murder” (see n. 4), 219–20 n. 33.

Source Dependence and the Development of the Pentateuch

677

Leviticus 24 simply omitted this, though it retained the whole communityʼs involvement: ‫ל־ה ֵע ָדה‬ ָ ‫( וְ ָרגְ מוּ אֹתֹו ָכּ‬Lev 24:14b).57 Leviticus 24 may have drawn in particular on Pʼs scapegoat ritual on the Day of Atonement in Lev 16:20–22 for its hand-placement gesture. There, Aaron lays his hands on the goat and confesses the peopleʼs sins, thus effectively transferring them to the animal. The goat is then taken outside the camp. The witnessesʼ performance in Lev 24 is similar: they impose their hands, after which the offender is taken out of the camp. Though the hand placement in Lev 24 may be a way for the witnesses to affirm their testimony – similar to how the witnesses in D affirm their testimony by starting the execution – it appears that H also views its gesture as a means of transferring pollution arising from blasphemy to the offender so that it can be removed from the population. Reversing the relationship of Lev 24 and Deut 17 cannot make sense of the relationship of Deut 17 to Deut 13 and of Deut 13 to VTE. More specifically, it is not clear why Deut 17 (and 13) would retain the motif of a “hand” from Lev 24ʼs ritual gesture (or some similar gesture in traditional legal practice) but use it in a rather banal idiom describing execution, changing the real “head” of the offender to the adverb “first.” One expects movement from the banal or idiomatic to the ideological and specialized rather than the reverse. Leviticus 24 apparently used Dʼs witness laws in the first place because of the talion law that is associated with them (Deut 19:21). This provided the conceptual avenue to Dʼs broader witness legislation. The talion law in 19:21 put Lev 24 in the context of the witness laws of Deut 19:15–21, and 19:15, requiring multiple witnesses, led to Deut 17:2–7, which includes such a requirement (v. 6). A feature in Dʼs talion law that made it good pickings for the hermeneutical task was its association with cases of intentional harm.58 Deuteronomyʼs false witness hatches a plan (‫ )זָ ַמם‬to the detriment of his fellow. What he hoped to do is therefore to be done literally to him (Deut 19:19). This is the essential modality of the injury law in Lev 24, and the passage uses and reiterates Dʼs language for emphasis, calling for performance of literal talion that operates when intent is

57  On the community in H, including Lev 24, see J. J oosten , People and Land in the Holiness Code: An Exegetical Study of the Ideational Framework of the Law in Leviticus 17–26 (VTSup 67; Leiden: Brill, 1996), 36–42. 58  Nihan, “Révisions” (see n. 4), 150–151, 158, and idem, “Murder” (see n. 4), 226, n. 50, sees the literal interpretation in Lev 24 being prepared for by Dʼs interpretation. Part of the broader conceptual development in D on the way to H is allowing for bodily mutilation as a punishment (Deut 25:11–12). CC appears to avoid incorporating corporal punishment from its sources, particularly notable in its not carrying over a literal interpretation of Hammurabiʼs talion laws, LH 196–201 (see W right , Inventing [see n. 1], 190, 264). C havel , Oracular Law (see n. 4), 79–80, observes that the talion law in Deut 19:21 operates as an analogy to the case that it elucidates and that the talion law in Lev 24:17–22 operates in a similar way for the blasphemy laws. See also n. 49, above.

678

David P. Wright

present. The difference is that Lev 24 portrays the harm as directly caused rather than the result of the mechanics of a degenerate legal process.

Conclusion It is difficult to be precise about the conceptual path that generated the blasphemy-talion pericope of Lev 24:10–23. The following, however, seems to capture the main aspects of the compositional or scribal-academic logic. The primary motivation was CCʼs prohibition against cursing the deity in Exod 22:27. Leviticus 24:15 essentially replicated this law. The Covenant Codeʼs prohibition, as an apodictic command, did not have an express penalty, so Lev 24 provided this, perhaps conceptually drawing on the nearby threat of divine punishment in Exod 22:22–23. Leviticus 24 used PH language to evoke this penalty: the one who curses his god will “bear his sin.” The reason for using this particular formula was paronymic association with the root *‫נשא‬, “bear,” that appears immediately after and in the second cursing law in Exod 22:27. Leviticus 24 created a companion law about uttering the name YHWH (v. 16). This sought to fill a gap in CC, which otherwise prohibited uttering the names of other gods (Exod 23:13) and spoke of the legitimate uttering of YHWHʼs name (20:24). The penalty for the new law was capital punishment. The form of and penalty for this new law were drawn from CCʼs capital crimes, which include homicide (Exod 21:12). This allowed Lev 24 to draw also on CCʼs talion laws, which are systemically part of CCʼs homicide and assault legislation (21:23–25). Because the talion laws were part of CCʼs miscarriage laws (21:22–23), this allowed Lev 24 to also use the miscarriage laws as a supplementary model for its two blasphemy laws and talion passage. It replicated the pattern of legal intensification in the miscarriage laws, thereby representing the second case, where YHWHʼs name is uttered, as more severe. Paronymic replacement or echoing, *‫ נקב‬for *‫נגף‬, influenced the choice of verb to describe the crime of the second blasphemy law. Leviticus 24 placed the talion laws after the basic double-law structure but modified this significantly. It detached talion from miscarriage, read the talion laws as arising from intentional assault, and reconstructed them so as to epitomize the scope of most of CCʼs casuistic legislation, including assault against humans (Exod 21:12–32) and animal loss and killing (Exod 21:33–22:14). Thus it was able to characterize and classify blasphemy as a crime and not simply as a transgression against deity as in CCʼs Exod 22:27. Leviticus 24 also used the miscarriage laws in building the narrative that provides a setting for the blasphemy and talion laws. Because Lev 24 now represented the talion crimes as intentional acts, it put the motif of a fight from CCʼs miscarriage law in the blasphemy narrative (Lev 24:10), where, given the nature of the offense, blasphemy could still be portrayed as intentional. The

Source Dependence and the Development of the Pentateuch

679

blasphemerʼs actions were characterized in accord with the blasphemy laws: he misspoke the deityʼs name while cursing. He thus was guilty of the more serious of the two laws and liable to the death penalty. The Covenant Code, however, did not provide the building materials for constructing the judicial process for prosecuting and executing the offender. For this Lev 24 went to D. The specific avenue for accessing D was its talion law in Deut 19:21, which was part of the witness laws of vv. 15–21. Deuteronomyʼs talion and associated witness laws (vv. 16–21) were helpful to Lev 24 in two ways. First, they provided a precedent and a catalyst for considering talion from the point of view of intentional acts. Deuteronomyʼs novelty over CC was that a false accuser was to suffer the punishment he sought to have inflicted on another. Leviticus 24 took this up but applied it to injuries caused directly. This, of course, was theoretically consistent with CCʼs requirement of capital punishment for intentional homicide. Leviticus 24 also picked up language from D about literally replicating the injuries on the offending party (24:19a, 20b; Deut 19:19). Second, Dʼs witness laws provided a way for representing the trial requirements and execution procedure in the blasphemy narrative. The law adjoining the false-witness law in Deut 19:15 called for multiple witnesses, which was instantiated in the apostasy law of Deut 17:2–7 (v. 6). Leviticus 24 represented Dʼs multiple ‫ע ִדים‬, ֵ “witnesses,” as ‫ׁש ְֹמ ִעים‬, “hearers,” to fit the new context (v. 14). More remarkably, it adapted the execution procedure from Deut 17:5–7, where the offender is taken to the outskirts of the town and stoned, first by the witnesses and then by the community. Leviticus 24 retained a two-stage execution process but turned the idiom of the witnessesʼ “hands at first being on” the offender into a ritual gesture of hand placement performed by the witnesses, followed by communal lapidation. It used paronymic association to change the adverb “at first” (‫)ּב ִראשׁ ֹנָ ה‬, ָ which refers to the first stage of the execution, to “on his head” (‫)על־רֹאשֹׁו‬, ַ which refers to where the witnesses place their hands. The rite of hand placement in the scapegoat ritual of Lev 16:21 provided a model for the gesture in the blasphemy pericope. The performance, as with the scapegoat, was a means of transferring miasma to the recipient that can then be removed from the habitation. Influences from sources run throughout Lev 23:10–21, in the narrative and in the laws as well as in the two sections of laws (blasphemy and talion). This adds significant evidence to Nihanʼs claim that the passage is a compositional whole.59  See Nihan, “Révisions” (see n. 4), 141–142, 152; idem, “Murder” (see n. 4), 212–215, 228–230. For the unity on the basis of thematic coherence, see M ilgrom, Leviticus 23–27 (see n. 4), 2120; Chavel, “Oracular Novellae” (see n. 57), 19. The standard hypothesis, as noted by Nihan, is that vv. 10–14, 23 have been combined in an artificial way with the prescriptions of vv. 15–16 (blasphemy) and vv. 17–22 (talion). See, for example, O tto , Kontinuum (see n. 18), 224–242. A. Cholewiński, Heiligkeitsgesetz und Deuteronomium: Eine vergleichende Studie (AnBib 66; Rome: Biblical Institute Press, 1976), 96–100, sees a core of Lev 24:15b–22 that was expanded by a narrative of 10–15a, 23. See also K. E lliger , Leviticus (HAT 4; 59

680

David P. Wright

The passage does not appear to have started with a core, either legal or narrative, that was subsequently expanded.60 The genre shifts, stylistic differences, and jarring juxtaposition of legal topics all arise from working material together from different sources (CC and D) on different topics (cursing deity, talion, miscarriage, witnesses, apostasy) and with different genres (casuistic and apodictic law) to create a new composition with a complex nature and new goals. Though the conclusions here cannot be mechanically applied to other passages, it suggests that the possible use of sources be added to the toolbox of redactional analysis for explaining textual complexity. This study also provides some data for thinking about the blasphemy-talion passage in its context. It appears to be a later addition.61 But it complements the use of CC and sometimes D visible in surrounding passages. The neighboring laws on the seventh year and debt release in the jubilee in Lev 25 display a use of both CC and D similar to what is found in the blasphemy-talion pericope. The festival calendar of Lev 23 relates topically to CC (Exod 23:14–19) and shows correspondences with Dʼs festival legislation (Deut 16:1–17).62 When it is remembered that the blasphemy-talion passage epitomizes much of CCʼs casuistic legislation, Lev 23–25 as a whole covers nearly the entire topical scope of CC. Other parts of the so-called Holiness Code (HC; Lev 17–26) fill in some of the remaining gaps. As we saw above, 22:27–29 reflect the sacrificial legislation of Tübingen: Mohr Siebeck, 1966), 278, 283–284, 329–333; Gabel and Wheeler, “Redactorʼs Hand” (see n. 4). R. R. Hutton, “Narrative in Leviticus: The Case of the Blaspheming Son (Lev 24,10–23),” ZABR 3 (1997), 145–163, here 147–149, 151–153, 162, argues that the talion laws in vv. 15b–17 are an addition to a basic narrative. The unity of the composition is demonstrated by both CCʼs and Dʼs influence on both sets of laws (blasphemy and talion) and on the narrative and by use of the same hermeneutical techniques in these different sections. 60  The motifs attributable to CC or D are a fight (v. 10); the entire concern about uttering the divine name (vv. 11, 16); mispronouncing the name, described with *‫( נקב‬vv. 11, 16); cursing, described with *‫ קלל‬with the object ‫ֹלהים‬ ִ ‫( ֱא‬vv. 11, 15); the root *‫ נשא‬in the idiom for bearing sin (v. 15); the formula for the capital penalty (v. 16); homicide and death penalty (vv. 17, 21); killing an animal and repayment (v. 18, 21); talion equations for “life” and bodily parts (vv. 18, 20); the requirement of literal talion (vv. 19, 20); and the execution procedure, including taking the offender outside the camp, the witnessesʼ hands on the offenderʼs head, the communityʼs participation, and stoning (vv. 14, 16, 23). Interestingly, this accounts for most of the novel or substantial material in the passage. Most of what is left are motifs from legal perspectives of the H school or incidental and structural elements: the identities of the fighting men, bringing the offender to Moses, putting him in detention, introductions of speech, application of laws to Israelite and immigrant, and the fulfillment clause. This raises doubts about whether there is an underlying tradition that has been taken up in that passage, as is argued by Leuchter, “Ambiguous Details” (see n. 4), esp. 436–437. 61  For a review of views, see Nihan, “Murder” (see n. 4), 212–215. 62  The PH Passover and Unleavened Bread laws in Lev 23 and Exod 12 have correlations with Deut 16, i.e., Lev 23:8+15 // Deut 16:8–9, where, for example, Hʼs ‫ ֶשׁ ַבע ַשׁ ָבּתֹות‬stands in place of Dʼs ‫שׁ ְב ָעה ָשׁ ֻבעֹות‬, ִ similar to the replacement of *‫ שמט‬with *‫ שבת‬in the seventh-year law; Exod 12:10, 11, 15, 19, 20 // Deut 16:3–4; and Exod 12:8–9 appears to react to Deut 16:7.

Source Dependence and the Development of the Pentateuch

681

Exod 22:28–29; 23:19b (discussed above), Lev 20:9 may reflect Exod 21:17,63 and Hʼs scattered laws or statements on the immigrant (e.g., Lev 19:33 and Exod 22:20) and the holiness of the people (e.g., Lev 20:26 and Exod 22:30) seem to have a precedent in CC along with influence from Dʼs development of these themes.64 Hʼs blasphemy-talion pericope also has ties to the list of priestly defects in Lev 21, which along with the passage on animal blemishes in Lev 22 has ties to D and CC. Leviticus 24 also echoes the larger concern about the sanctity of the divine name found broadly in HC. So, though the blasphemy-talion pericope looks secondary, it fits the contextual plan of this part of PH in terms of legislative motivation and strategy.65 That is, these HC chapters seem intent on digesting and interpreting CC, in connection with Dʼs refraction of CC, in order to expand priestly legislation beyond the cultic scope. This also gives a new explanation for why Lev 24:10–23 is where it is: it is part of the comprehensive scribal recasting of CC within the boundaries of HC.66 As noted, one of the ideological purposes of juxtaposing talion to blasphemy appears to be to classify blasphemy in the system of casuistic criminal law: blasphemy is as serious as murder. Other ideological considerations and legislative goals may be at play. This includes, as recently argued for in detail by Nihan, seeing homicide and injury to humans, who are created in the image of God, as attacks on the deity.67 The transformation of Dʼs hand motif into a hand-placement gesture also indicates that pollution motifs are a concern and therefore that 63  Cf. C havel , “Lev 24:10–23” (see n. 4), 2143. But see W right , Inventing (see n. 1), 193–204, for an argument that a law like Exod 21:17 was part of Israelite tradition. 64  For Hʼs immigrant legislation that reflects CC, compare Lev 19:33–34 and Exod 22:20–23; 23:9. Immigrant legislation runs throughout HC (17:8, 10, 12, 13, 15; 18:26; 19:10; 20:2; 22:18; 23:22; 24:16, 22; 25:23, 35, 47) and is found in H materials outside of HC (see the listing in Nihan, “Murder” [see n. 4], 230, n. 57; see the longer study by C. Nihan, “Resident Aliens and Natives in the Holiness Legislation,” in The Foreigner and the Law: Perspectives from the Hebrew Bible and the Ancient Near East [ed. R. Achenbach et al.; BZABR 16; Wiesbaden: Harrassowitz, 2011], 111–134; see also C havel , Oracular Law [see n. 4], 54). For D legislation on the immigrant (often in connection with the widow and orphan) and the relationship to CC, see Wright, Inventing (see n. 1), 476, n. 44; 507, n. 19. The holiness of the people is a broad theme of HC: Lev 19:2; 20:7, 8, 25, 26; 21:8; 22:32 (holiness and diet in 11:42–45, an H supplement). This is of concern in D as well (see Wright, Inventing, 486–487, n. 15; 506, n. 19). For other connections between H and CC, see Wright, Inventing, 507, n. 22. For examples of Hʼs dependence on D outside of HC, see n. 47. 65  PHʼs use of CC and other non-P materials is broader than just what we see in HC. See Wright, “Profane Slaughter” (see n. 10); idem, “Law and Creation” (see n. 3). 66  For attempts to explain the position of the passage thematically or structurally, see Milgrom, Leviticus 23–27 (see n. 4), 2102–2106, 2128–2133; J. R. Master, “The Place of Chapter 24 in the Structure of the Book of Leviticus,” BSac 159 (2002), 415–424. 67  N ihan , “Murder” (see n. 4), 221, 230–236. C havel , “Lev 24:10–23” (see n. 4), 2140–2145, says the analogy is more direct: blasphemy is like killing the deity. The paronymic reflection of *‫ נגף‬in *‫ נקב‬would support this.

682

David P. Wright

the pollution of blasphemy is somehow related to that of homicide. That both the blasphemy and talion sections are applicable to Israelite and immigrant alike (24:16, 22) also reflect, in part, a broadening of the audience to which CCʼs laws apply.68 But legal classification of blasphemy appears to be the main goal. This makes sense of why the talion laws of Lev 24 deal with a wide range of legal cases, homicide, injury to humans, and killing domesticated animals. Working out textual relationships in such a way is important for reconstructing the development of the Pentateuch. When the textual genealogical tree can be determined, we are in a better position to argue for absolute dating, especially when we can determine an Archimedean point or two as a foundation.69 Assessing textual relationships is also important for the task of intellectual history. By determining the questions that drove the generation of subsequent texts and defining the hermeneutical techniques that scribes used, we get some notion about how ideas and perspectives have evolved over time.70

 Cf. Vroom, “Recasting” (see n. 4).  N ihan , “Murder” (see n. 4), and idem , “Révisions” (see n. 4), 141, 156–158, argues that the ideological picture of the blasphemy-talion pericope, along with other redactional considerations, fits in with the practice of law in Judea and Samaria at the end of the Persian period. On the difficulty of using sociological and ideological data for historical reconstruction, see Wright, Inventing (see n. 1), 97, 127, 169, 174, 389, n. 36; idem, “Ritual Theory, Ritual Texts, and the Priestly-Holiness Writings of the Pentateuch,” in Social Theory and the Study of Israelite Religion (ed. S. Olyan; SBLRBS 71; Atlanta: Society of Biblical Literature, 2012), 195–216, here 199–209, 216; B. D. Sommer, “Dating Pentateuchal Texts and the Perils of Pseudo-Historicism,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 85–108. For the dating of the basic laws of Deuteronomy as a benchmark in the survey of chronology, see n. 53 and the discussion there. 70  I thank my student Dustin Ray for his help preparing the manuscript for this essay. 68

69

Part Seven Historical Geography of the Pentateuch and Archaeological Perspectives

Introduction Jan Christian Gertz How can historical geography and archaeology contribute to our understanding of the formation of the Pentateuch? The connection between historical-topographical and archaeological findings and the textual world of the Bible has a long history. In the early church, the emerging interest in historical commentaries on biblical toponyms was literarily reflected in Eusebiusʼs Onomasticon1 and iconographically in the mosaic map of Madaba.2 For his Onomasticon, Eusebius analyzed the historical books of the Old Testament and also – in part – the Gospels. He listed the names of cities, regions, territories, mountains, plains, rivers, and deserts and gave a brief explanation of both the placeʼs name and its geographical identification. Occasionally, he pointed to changes that had occurred since biblical times. The Onomasticon was thought to serve as an exegetical device, just as the preoccupation with biblical antiquities initially drew on interest in gaining better understanding of the biblical text in terms of history. Within studies looking at the oral prehistory (Überlieferungsgeschichte) of pentateuchal texts from the first half of the twentieth century, historical geography was connected to the literary history of the Pentateuch for the first time. In the toponyms of the book of Genesis, this kind of research saw possible focal points for the formation of preliterary traditions. In more recent redaction-historical research, questions of historical topography and realities have been addressed for a different purpose. These studies have tended to focus on the literary function and significance of place names for the general historical conditions and compositional origins of a text. The inquiries were based on the insight that investigation of toponyms in a passage could yield extensive information on the authorʼs perspective on the history of a given territory and thus on the authorʼs historical horizon. Here the danger of pseudohistoricism arises, however. Ancient authors almost certainly sought to make their stories historically plausible. In this respect, historical consistency does not necessarily say much about the historicity of the narratives or the date of their composition. The situation is different, however, for those texts that can justifiably be identified as more or less explicitly presupposing a certain historical situation  Eusebius, The Onomasticon: Palestine in the Fourth Century A.D. (ed. J. E. Taylor; trans. G. S. P. Freeman-Grenville; Jerusalem: Carta, 2003). 2  H. Donner, The Mosaic Map of Madaba (Kampen: Kok Pharos, 1992). 1

686

Jan Christian Gertz

with respect to territory or culture or reflecting such a situation from certain ideological perspectives. In these cases, topographical studies offer opportunities for the historical classification of texts that otherwise elude dating because they take place in “prehistory.” That is why the demand arises for research that can constructively connect historical-topographical investigations with the analysis of texts. Therefore, literary-topographical studies on the itineraries of the wandering in the wilderness3 or the toponyms of the Exodus narrative4 require the establishment of literary, symbolic, or ideological maps. These mental maps should not reflect the historical-topographical data but the constellation of toponyms as intended in the texts. Because of the sheer amount of material, however, the connection of territorial history and archaeological discoveries with the textual world of the Pentateuch is a difficult task. We have therefore decided to look at the book of Numbers as an example. This choice is based on a remarkable ambiguity in the book. On the one hand, with respect to redaction history, this book is clearly distinguishable from the other books of the Tetrateuch. I recall a rather casual remark by Martin Noth: “If we were to take the book of Numbers on its own, then we would think not so much of ‘continuing sources’ as of an unsystematic collection of innumerable pieces of very varied content, age and character (‘Fragment Hypothesis’).”5 On the other hand, the book of Numbers contains several texts allowing for an approach that should facilitate historical analysis. In contrast to the other books of the Tetrateuch, the book of Numbers repeatedly offers us the chance to correlate its textual world and its genesis with texts outside of the Bible (e.g., from Ketef Hinnom or Deir Alla). Moreover, the book of Numbers contains a multitude of geographical information that almost leaps off the page to grant access to its territorial history. Within the literary-historical play of a relative chronology, should it not be Numbers, of all the books of the Pentateuch, that enables us to find benchmarks for an absolute chronology based on external evidence? In any case, it is methodologically necessary to take into account that source criticism and redaction history on the one hand and topography and archaeology on the other are distinct approaches that can sometimes be connected by their results and sometimes cannot.  See J. M. M iller , “The Israelite Journey through (around) Moab and Moabite Toponymy,” JBL 108 (1989), 577–595; T. B. Dozeman, “Geography and Ideology in the Wilderness Journey from Kadesh through Transjordan,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 173–189; A. R. Roskop, The Wilderness Itineraries: Genre, Geography, and the Growth of Torah (History, Archaeology, and Culture of the Levant 3; Winona Lake, IN: Eisenbrauns, 2011). 4  F. V. Greifenhagen, Egypt on the Pentateuchʼs Ideological Map: Constructing Biblical Israelʼs Identity (JSOTSup 361; Sheffield: Sheffield Academic Press, 2002). 5  M. Noth, Numbers: A Commentary (OTL; London: SCM Press, 1968), 4–5; German original, Das vierte Buch Mose: Numeri (ATD 7; Göttingen: Vandenhoeck & Ruprecht, 1966), 8. 3

Biblical Criticism from a Geographerʼs Perspective “Transjordan” as a Test Case David Ben-Gad HaCohen Introduction The use of geography and place names is one of the prime means used to fashion the stage on which the biblical narrative is played out.1 The scholarly field of biblical historical geography has developed in an attempt to understand this geographical background. At its best, historical geography should describe the geography as it was in the past, the flora and the natural landscape, along with manʼs actions and his contribution to the landscape, all in relation to a defined historical period.2 The geographer who engages in historical research should examine the processes that led to landscape change in the past, both the natural landscape and that fashioned by man. Accordingly, biblical historical geography is meant to reconstruct the lands of the Bible, with a series of maps showing the changes that occurred during subperiods. The ancient literature, however, does not provide sufficient data for a complete reconstruction of its periods, thereby limiting this discipline to identifying toponyms and sketching the boundary lines and journey routes mentioned in the Bible and in the literature of the ancient Near East. These lines, which are not evident on the ground, are actually historical blueprints overlaying the geography. Thus, geographical tools are employed for historical needs, but this is not geography, and certainly not historical geography. I wish to indicate a different realm of the study of geography and the Bible, neither the historical geography of the biblical period nor biblical toponymy but rather geographical criticism of the Bible. Biblical toponymy searches for continuity and similarity in place names in order to enable the identification and location of sites mentioned in the Bible. The research method employed in geographical criticism, in contrast, is identical to that used in biblical textual criticism: finding the differences in place names and the inconsistencies in lists 1  See Y. Amit, “The Function of Topographical Indications in the Biblical Story,” Shnaton 9 (1987), 15–30 (Hebrew). 2  See Y. Ben-Arieh, “A Geographical Approach to Historical Geography,” Studies in the Geography of Israel 7 (1970), 138–160, esp. 147–152 (Hebrew).

688

David Ben-Gad HaCohen

of toponyms or journey lists. Place names in the land of Israel remained almost without change for thousands of years. Even places whose names were changed by the various conquerors returned to their original names centuries later, when the conquerors had left the land.3 Based on the general continuity and stability of geographical names, any toponymy change, even the most minor, may be highly significant. A name change might constitute proof of regime change or an ideological upheaval. Thus, biblical textual criticism must relate to toponym change as indicative of a change in authorship. Textual criticism successfully identifies duplications and inconsistencies in the biblical narrative, but the interpretation given to those problems in the text is subjective and prone to the inclinations of the scholars. This leads to broad disagreements regarding the exact attribution of verses and even the rejection of the entire Documentary Hypothesis. In contrast to this multiplicity of possible understandings, the geographical differences are unequivocal. Since we are dealing with geography and toponyms, no texts could be better for presenting this topic than Num 21 and Num 33, which are replete with geographical names. In 2002, Thomas Dozeman surveyed more than twenty studies of the Israelitesʼ journeys in Transjordan, in an article entitled “Geography and Ideology in the Wilderness Journey from Kadesh through the Transjordan.”4 Though this study should have been concerned with geography, the studies Dozeman cites are occupied with the history of Transjordan and not with the toponyms that represent the geography in which these narratives unfold. As if to stress his lack of interest in geography, Dozeman removes from his discussion the main geographical source: Numbers 33.5 Identification of most of the toponyms connected with Transjordan was not clear to scholars; some regarded the entire list as an imaginary geography or geography mobilized for ideological needs.6 Even imaginary geography, however, is subject to geographical rules and must exhibit internal logic. Even in an imaginary geography, one looks from the mountain peak to the depths of the valley, and when one goes northward, one sees the sun rising to the right. This article will seek to demonstrate that geography, even if imaginary, enables us to draw important conclusions in the study of the Bible.

3  See A. F. Rainey, “Name, Place Names,” Entziklopediyah Mikra‍ʾ it, 1–8 (ed. H. Tadmor; 8 vols.; Jerusalem: Mossad Bialik, 1972–1982), 8:11–29, esp. 20 (Hebrew); E. A. K nauf , “Toponyms and Toponymy,” ABD 6:601–605, esp. 603; Y. E litsur , Ancient Place Names in the Holy Land: Preservation and History (trans. D. Louvish; Jerusalem: Magnes; Winona Lake, IN: Eisenbrauns, 2004), xi, 1. 4  See T. B. Dozeman, “Geography and Ideology in the Wilderness Journey from Kadesh through the Transjordan,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 173–189. 5  See Dozeman, “Geography and Ideology” (see n. 4), 173, n. 1. 6  See Dozeman, “Geography and Ideology” (see n. 4), 174.

Biblical Criticism from a Geographerʼs Perspective

689

1 The Journey of the Israelites in Transjordan The description of the route the Israelites took on their journey from Egypt, presented as a continuous list of uniformly formulated journeys, appears in the itinerary list in Num 33:3–37, 41–49. A similar list is to be found as itinerary notes that appear intermittently in the books of Exodus and Numbers.7 There is a great deal of overlap between site descriptions and names in the itinerary notes and those in the itinerary list. The description of the section of the route through Transjordan from Kadesh to the steppes of Moab is exceptional in this respect. In the description of this journey section, the itinerary list includes ten sites that were traversed in nine journeys: Kadesh, Mount Hor, Zalmonah, Punon, Oboth, Iye-abarim, Dibon-gad, Almon-diblathaim,8 the hills of Abarim (‫)הרי העברים‬,9 and the steppes of Moab (Num 33:37, 41–48). In contrast, the itinerary notes (Num 20:22–22:1) include twelve journeys and seventeen sites: Kadesh, Mount Hor, the “way of the Sea of Reeds,” “the land of Edom,” Oboth, Iye-abarim, “the wilderness bordering on Moab to the east,” Wadi Zered, “beyond the Arnon, that is, in the wilderness that extends from the territory of the Amorites,” Beer,10 Midbar, Mattanah, Nahaliel, Bamoth, “the valley that is in the territory of Moab,” “the peak of Pisgah,” and the steppes of Moab. The difference between the two journey descriptions is greater than the number of journeys and sites. Of the twenty-two sites mentioned in the two parallel itineraries of the Israelite journey from Kadesh to the steppes of Moab, only five sites are common to the itinerary list and the itinerary notes, whereas seventeen sites appear only in one description and not in the other. 7  Bible scholars differ in their analyses of the literary connection between the itinerary list and the itinerary notes. Noth maintained that the itinerary list is the product of the compiler, who collected the various stations that appear in the itinerary notes in Exodus and Numbers in this chapter and added it to the itinerary of a single specific route. See M. N oth , “Der Wallfahrtsweg zum Sinai,” Palästinajahrbuch des Deutschen evangelischen Instituts für Altertumswissenschaft des Heiligen Landes zu Jerusalem 36 (1940), 5–28; idem, Numbers: A Commentary (trans. J. D. Martin; OTL; Philadelphia: Westminster, 1968), 242–243. Other scholars, in contrast, regard the itinerary notes and associated events as a literary expansion of the itinerary list. See F. M. Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), 301–321, esp. 308–309; G. I. Davies, The Way of the Wilderness: A Geographical Study of the Wilderness Itineraries in the Old Testament (SOTSMS 5; Cambridge: Cambridge University Press, 1979), 59–60. For an additional list of scholars who held this view before the 1970s, see idem, “The Wilderness Itineraries and the Composition of the Pentateuch,” VT 33 (1983), 1–13, esp. 8, n. 13. 8  MT reads: ‫דבלתימה‬, which seems to be a toponym with a heh added as a directional article. The version ‫ דבלתים‬is documented in LXX, Δεβλαθαιμ, and in Jer 48:22. 9  In other places this location is called ‫( הר העברים‬Mount Abarim; Num 27:12; Deut 32:49). 10  The text reads “and from there ‫( ”בארה‬Num 21:16), meaning, “from there to Beer.” Consequently, the site was named Beer. See A. J. Ferch, “Beer (place),” ABD 1:640.

690

David Ben-Gad HaCohen

Furthermore, stylistic differences too are evident also in the two descriptions of journeys in Transjordan. The descriptive formula in the itinerary list is fixed: “they set out from X and encamped at Y, they set out from Y and encamped at Z” (Num 33:5–48). This formula, with both journeying and encamping verbs and a starting point and encampment, also appears in the itinerary notes (Num 21:11a). Along with the description employing this formula, the itinerary notes also contain descriptions in a different formulation: a formula with journeying and encamping verbs that does not mention the starting point of the journey but does specify the encampment site (“from there they set out and encamped at Y, from there they set out and encamped at Z” [Num 21:12–13]); and a concise style, without verbs (“from X to Y, and from Y to Z” [Num 21:18–20]). In addition to the descriptions of defined style, the itinerary notes also contain journey sections of undefined style (Num 21:4, 10, 11). The three main itinerary styles in the Pentateuch have Assyrian parallels: the two earliest journey descriptions known to us (ARMT 1.26; SH 809) are from the time of Samsi-Addu I (1813/1809–1781 BCE).11 These descriptions are in the format “from X to Y, from Y to Z.”12 A description in a similar format (SAA XI.14) is known from the time of Tiglath-pileser III (745–727 BCE).13 A description (DeZ 2521) that is to be ascribed to the time of Tukulti-Ninurta I (1243–1207 BCE)14 uses the formulation “from there at Y they will stay the night; from there at Z they will stay the night.”15 Beginning in the time of Adad-nirari II (911–891 BCE), we find extensive use made of sentences that contain both setting out and arrival or staying-the-night verbs.16 Despite the stylistic range present in the Assyrian inscriptions, they do not mix styles. Each scribe chose the style he found suitable and adhered to it throughout the inscription. In contrast, stylistic divergence in depictions of the Israelite route in Transjordan is undisputable, and no interpretation is capable of finding logic in the jumble of the Transjordan itinerary. 11  See G. Dossin, Correspondance de Samsi-Addu et de ses fils (ARMT 1; Paris: Imprimerie Nationale, 1950), 66–69; M. Lurker (ed.), Beiträge zu Geschichte, Kultur und Religion des Alten Orients: In Memoriam Eckhard Unger (Baden-Baden: Körner, 1989), 189–195; J. E idem and J. L æssøe , The Shemshara Archive, 1: The Letters (Copenhagen: Kongelige Danske Videnskabernes Selskab, 2001), 70–72 (= SH). 12  See ARMT 1.26:14–23; SH 809:55–59. 13  See F. E. Peiser, Studien zur orientalischen Altertumskunde (Mitteilungen der Vorderasia­ tischen Gesellschaft 6/3; Berlin: Peiser, 1901), 40–46; L. D. Levine, “K. 4675+: The Zamua Itinerary,” SAAB 3 (1989), 75–92. 14  See W. Rollig, “Ein Itinerar aus Dur-Katlimmu,” Damaszener Mitteilungen 1 (1983), 279–284 (DeZ = Deir ez-Zor Museum). 15  See DeZ 2521:2–15. 16  See A. B aruchi -U nna , “Genres Meet: Itineraries, Prayers and Divine Messages in Assyrian Royal Inscriptions” (PhD diss., Hebrew University of Jerusalem, 2009), 55, ll. 4–9; 60, ll. 13–16 (Hebrew).

Biblical Criticism from a Geographerʼs Perspective

691

Angela Roskop, in her attempt to resolve the mixture of styles in descriptions of the route taken by the Israelites in Transjordan, argues that the journey route is composed of a Priestly route that underwent three redactions.17 The original Priestly route, she maintains, led to the war with Arad (Num 21:1–3) and to the conquest of the land of Israel from the south.18 This route underwent two redactions in order to incorporate the war with Sihon and the Balaam narrative, thereby diverting the route from the Negev to the heart of the kingdoms of Edom and Moab.19 The third redaction was intended to give the impression that the route passed through the wilderness to the east of Moab.20 According to her proposal, the order of the journeys in Num 33 is a compilation of the journeys listed in the itinerary notes, to which were added sites that appear in Joshua and in Deuteronomy.21 It should be recalled that narrative continuity marks the difference between numerous redactions and numerous sources.22 If the passages that were presumably added by various editors are divorced from the narrativeʼs continuity, then these are editorial passages. If, however, these passages are joined together to form a continuous narrative or to form a number of story lines, each of which exhibits logical continuity, then we have a number of sources that a compiler joined together and not a number of redactions. I will show below that the passages that Roskop viewed as editorial additions join three journey routes, each of which possesses its own narrative continuity and geographical logic, and only their joining together removes the geographical logic from the united narrative.23  A. R. Roskop, The Wilderness Itineraries: Genre, Geography, and the Growth of Torah (Winona Lake, IN: Eisenbrauns, 2011), 204–215. 18  See Roskop, Wilderness Itineraries (see n. 17), 193–203. 19  See Roskop, Wilderness Itineraries (see n. 17), 206–208. 20  See Roskop, Wilderness Itineraries (see n. 17), 209–214. 21  See R oskop , Wilderness Itineraries (see n. 17), 224–225. This supposition does not explain the sites Dophkah and Alush (Num 33:12–14), Rithmah, Rimmon-perez, Libnah, Rissah, Kehelath, Mount Shepher, Haradah, Makheloth, Tahath, Terah, Mithkah, Hashmonah (Num 33:18–30), and Zalmonah (vv. 41–42), which appear only in the itinerary list. Nor does it explain why the wilderness of Paran (Num 10:12; 12:16; 13:3, 26), the Wadi Zered (Num 21:12), the Arnon (v. 13), Beer (v. 16), Mattanah, Nahaliel, Bamoth, “the valley that is in the territory of Moab,” “the peak of Pisgah” (vv. 18–20), and Jahaz (v. 23), which appear in the itinerary notes, are not included in the itinerary list. 22  For the importance of narrative continuity in identifying the documents from which the Pentateuch was composed, see B. J. Schwartz, “The Torah: Its Five Books and Four Documents,” in The Literature of the Hebrew Bible: Introductions and Studies (ed. Z. Talshir; Jerusalem: Yad Ben-Zvi, 2011), 161–226, esp. 193, 213 (Hebrew); J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (New Haven, CT: Yale University Press, 2012). 23  Roskop did not examine this possibility, since she maintains that the Documentary Hypothesis cannot be used for the route in Transjordan because the criterion of the name of God is hardly applicable in those chapters. See R oskop , Wilderness Itineraries (see n. 17), 17

692

David Ben-Gad HaCohen

2 The Israelite Routes in Transjordan The war with Sihon occupies a special place in the events of the journey in Transjordan. Sihon, king of the Amorites, appears in the Pentateuch not only as a king who fought Israel but also as a ruler whose earlier conquests in Transjordan are reviewed both in narrative (Num 21:26) and in poetic form (Num 21:27–30). The uniqueness of this king also reveals the importance of Israelʼs war with him, the conquest of his land, and the settlement in it (Num 21:21–25). Fortunately, the war with Sihon is described extensively in three places: in the real-time narrative in Numbers (21:21–30); in Mosesʼs first-person retelling in Deuteronomy (2:24b–36); and in Jephthah the Gileaditeʼs “history lesson” in Judges (11:19–22). There are differences between the descriptions, but the basic message is the same: the Israelite camp sent a delegation to the Amorite king Sihon to ask permission to pass though his country. Sihon refused and gathered his army to make war against Israel. The battle was waged at Jahaz; Israel prevailed and took possession of Sihonʼs land. Scholars have noted the literary connection between the three descriptions and the linkage between the delegation to Sihon and that to the king of Edom (Num 20:14–21; Judg 11:17).24 After establishing this connection, many have sought to identify directions of influence and an inner chronology within the descriptions. Here as well, the inquiry was limited to the historical and literary data, and scholars ignored the list of sites belonging to the narrative of the Transjordan journey. The war with Sihon is a single event among the continuous events and sites from which the Israelite route in Transjordan is constructed, all of which appear in the three descriptions of the war. Jephthahʼs account of the route includes Kadesh, where the Israelites stayed and from which they set out; the journey in the wilderness, around the land of Edom and to the east of Moab; the Arnon; and, of course, Jahaz, the site of the battle with Sihon (Judg 11:17–20). Deuteronomy provides a more detailed route, which includes all the journey elements in Jephthahʼs speech in the identical order: Kadesh (Deut 1:46), the wilderness by way of the Sea of Reeds (2:1a), around the hill country of Seir (2:1b), the road of the Arabah, away from Elath and Ezion-gaber (2:8a), through the wilderness of Moab (2:8b), the Wadi Zered (2:13), the Wadi Arnon (2:24α), the wilderness of Kedemoth (2:26a), and Jahaz (2:32). Most of the sites listed in D also appear in Numbers, in the same order: Kadesh (Num 20:14, 16), the Sea of Reeds around 204. It must be emphasized that the distinction between the various documents cannot be based solely on differences in divine names. 24  See W. A. S umner , “Israelʼs Encounters with Edom, Moab, Ammon, Sihon, and Og according to the Deuteronomist,” VT 18 (1968), 216–228; J. Van Seters, “The Conquest of Sihonʼs Kingdom: A Literary Examination,” JBL 91 (1972), 182–197; J. R. Bartlett, Edom and the Edomites (JSOTSup 77; Sheffield: Sheffield Academic Press, 1989), 180–181.

Biblical Criticism from a Geographerʼs Perspective

693

the land of Edom (Num 21:4aβ), the wilderness to the east of Moab (21:11bβ), the Wadi Zered (21:12), the Wadi Arnon (21:13a), and Jahaz (21:23). If we compose a list tha Numberbs t includes journey items that are listed in at least two of the three descriptions, it would comprise Kadesh, the Sea of Reeds around the land of Edom/the hill country of Seir, east of Moab/the wilderness of Moab, the Wadi Zered, (the Wadi) Arnon, and Jahaz. Our list now includes way stations and a war narrative and recurs in three descriptions. Since this is a record of a journey, we will call it journey A; it may be compared to the parallel journey records and thereby distinguish, define, and ascribe each one to its original source. Another route is recorded in Num 33. This route contains the following stations, which are linked by transit verbs: Kadesh, Mount Hor, Zalmonah, Punon, Oboth, Iye-abarim, Dibon-gad, Almon-diblathaim, the hills of Abarim, and the steppes of Moab (Num 33:37–49). Only Kadesh appears in both this route and in journey A; accordingly, these are two different journey descriptions. For the purposes of the current discussion, we will call the description in Num 33 journey B. Four of the stations in journey B appear in the same order in Num 20–21: Kadesh (20:22), Mount Hor (20:22–27; 21:4), Oboth (21:10–11), and Iye-abarim (21:11). An additional route that appears as a consecutive list includes the stations of Beer,25 Midbar,26 Mattanah, Nahaliel, Bamoth, the valley that is in the country of Moab, and the peak of Pisgah (Num 21:16–19). None of these stations is known to us from Num 33, and so this route is not part of journey B. The third journey ends at “the peak of Pisgah,” and a mountain named “Pisgah,” with a summit and slopes, appears in other verses, as well: Numbers 23:13–14 Then Balak said to him, “Come with me to another place from which you can see them – you will see only a portion of them; you will not see all of them – and damn them for me from there.” With that, he took him to Sedeh-zophim, on the summit of Pisgah. Deuteronomy 3:17 (and, with slight change, 4:49; Joshua 12:3) The Arabah, from the foot of the slopes of Pisgah on the east.

These two verses clearly describe the lofty mountain named “(the) Pisgah”27 as 25  Bacon identifies Beer in Numbers with Beer-elim in Isaiahʼs Moab pronouncement (Isa 15:8); see B. W. Bacon, The Triple Tradition of the Exodus (Hartford, CT: Students Publishing Company, 1894), 160; J. J. S imons , The Geographical and Topographical Texts of the Old Testament (Leiden: Brill, 1959), 262, 436; J. D. W. Watts, Isaiah 1–33 (WBC 24; Waco, TX: Word, 1985), 231. The attribution of both toponyms to Moab is not baseless. Likewise, the adjectival elim meant rulers/chieftains in Moab (Exod 15:15), and it was chieftains who dug the well (Num 21:18). 26  Instead of “and from the wilderness,” LXX reads, “and from the well” (και ἀπὸ Φρέατος). 27  The toponym ‫אשדת הפסגה‬, “the slopes of Pisgah” (Deut 3:17; 4:49; Josh 12:3; 13:20) indicates that (Ha)-Pisgah is a specific name, not a generic one. See E. K lostermann ,

694

David Ben-Gad HaCohen

providing a view of the northern Dead Sea (Deut 3:17) and of the Israelite camp in the steppes of Moab (Num 23:14).28 The geographical data place Pisgah to the northwest of Jahaz, the site of the battle against Sihon.29 This route, belonging to the third journey, led to the top of Pisgah, without regard for or attention to the local king or war against him.30 Accordingly, the route of the journey from Beer to the peak of Pisgah and journey A belong to different sources.31 We will name this additional description journey C. Defining the three routes resolves the problem of verses with journey descriptions of indefinite style (Num 21:4, 10–11). These verses were composed by the compilerʼs combining data from two different lists. Unraveling the “stitches” joining them will enable us to reconstruct the two original lists: “they set out from Mount Hor” (21:4aα), “they encamped at Oboth” (21:10b), and “they set out from Oboth and encamped at Iye-abarim” (21:11a). In terms of its style and station order, this list corresponds to journey B, as defined above. Whatever does not belong to journey B joins the continuous route: “they set out by way of the Sea of Reeds to skirt the land of Edom. But the people grew restive on the journey” (21:4aα–4b); the narrative of the copper serpent (21:5–9), “the Israelites marched on” (21:10a), and “they encamped in the wilderness bordering on Moab to the east. From there they set out and encamped at the Wadi Zered. From there they set out and encamped beyond the Arnon, that is, in the wilderness that extends from the territory of the Amorites. For the Arnon is the boundary Eusebius Onomastikon der Biblischen Ortsnamen (Leipzig: J. C. Hinrichs, 1904), 16, ll. 22–24; R. S. N otley and Z. S afrai , Eusebius, Onomasticon: The Place Names of Divine Scripture (Boston: Brill, 2005), 158 (Φασγα, Pisgah). 28  Haran formulates this similarly. See M. Haran, “The Exodus Routes in the Pentateuchal Sources,” Tarbiz 40 (1971), 113–143, esp. 138 (Hebrew). Scholars generally agree that “the steppes of Moab” is a region northeast of the Dead Sea. See B. M ac D onald , “East of the Jordan”: Territories and Sites of the Hebrew Scriptures (ASOR Books 6; Boston: American Schools of Oriental Research, 2000), 87; Z. Meshel, “Defining the Biblical ‘Araba,’ ” in “Up to the Gates of Ekron”: Essays on the Archaeology and History of the Eastern Mediterranean in Honor of Seymour Gitin (ed. S. W. Crawford et al.; Jerusalem: Albright Institute of Archaeological Research, 2007), 423–435, n. 6. See the extensive bibliographies in both articles. Only P explicitly mentions the Israelite encampment in the steppes of Moab (Num 26:3, 63; 31:12, and more). However, according to the other sources, which have the Israelites enter Canaan near Jericho, the route passed to the northeast of the Dead Sea, in the region that is to be identified with Pʼs “steppes of Moab.” 29  The various references to Jahaz place it to the north of the Arnon and to the east of the settled region, on the border of the wilderness. Ten different identifications have been proposed for this place, all to the south or southeast of Pisgah. The most comprehensive concentration of data regarding Jahaz, the proposed identifications, and an up-to-date discussion are to be found in MacDonald, East of the Jordan (see n. 28), 103–106. 30  See Bacon, Triple Tradition (see n. 25), 210. 31  See M. Noth, “Num. 21 als Glied der ‘Hexateuch’-Erzählung,” ZAW 58 (1940), 161–189, esp. 178; J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (Berlin: G. Reimer, 1889), 111.

Biblical Criticism from a Geographerʼs Perspective

695

of Moab, between Moab and the Amorites” (21:11b–13). The stations in this list are identical in name and order to journey A. This reconstruction demonstrates that the route descriptions without defined style are not the additions of various redactors. The possibility of reconstructing three different routes, each of which represents a continuous route of uniform style, indicates that the compiler of the Pentateuch combined data from three different sources in Num 20–21. The compiler joined the three routes into a single one, which appears uniform in literary terms but which, as a journey description, lacks geographical logic.

3 The Journey Routes and the Pentateuchal Documents The main event in the description of journey A is the war with Sihon. Benjamin Bacon determined that the war with Sihon undoubtedly belongs to E but without citing a single argument in support of this assertion.32 Martin Noth observed that there is only a single point that aids in attributing this event to E, namely, the frequent use of the term Amorite (Num 21:21, 25, 26, 28), in contrast with the term Canaanite that J often employs.33 The criterion used by Noth is applicable to the territory to the west of the Jordan but is irrelevant for the area to the east of the river. There is not a single appearance in the Bible of the term Canaan in connection with Transjordan. This name is reserved exclusively for the territory west of the Jordan. Consequently, when speaking of Transjordan, the term Amorite does not suffice to ascribe the war with Sihon to E.34 Nor can the continuity of the narrative aid in attributing the war to a particular source. The Wadi Arnon marked the southern border of the Amorite kingdom (Num 21:24, 26). This datum, which is incorporated into the narrative of the war with Sihon, already appeared in the itinerary note that described the arrival of the Israelites at the Wadi Arnon (21:13). Accordingly, the note serves to introduce the portrayal of the war with Sihon, and both descriptions that mention the Arnon belong to the same source. This itinerary note is one of the two such notes with the singular wording, “From there they set out and encamped at [. . .]” (21:12–13). The prepositional phrase “from there” does not enable the forging of a clear and unequivocal connection between these two journeys and what came 32  Bacon ascribes the war with Sihon to E based on the inclusion of this war in the route of E, but he offers no reasoning for the identification of this route. See Bacon, Triple Tradition (see n. 25), 210–211. 33  See Noth, Numbers (see n. 7), 162. 34  Noth ascribes the verse “and they dispossessed the Amorites who were there” (Num 21:32b) to J and suggests that this verse was a redactional change, since J apparently followed the usual practice and used the term Canaanite rather than Amorite. See Noth, Numbers (see n. 7), 133. If this verse belongs to J, however, it proves that even J used the term Amorite in reference to Transjordan.

696

David Ben-Gad HaCohen

before them. Alan Wilkin Jenks could therefore state unequivocally that not a single verse between the narrative of the revolt of Dathan and Abiram (Num 16–17) and that of Balaam (Num 22–24) can be ascribed to E.35 As a result, Richard Elliott Friedman could ascribe the journeys and the war with Sihon to J.36 We will now add the geographical aspect to the discussion. The area described as the land of Sihon became the possession of the Israelite kingdom, and Jahaz, the city in which Sihon mustered his troops, is mentioned in the Mesha Stela as the site of the struggle between Moab and the kingdom of Israel.37 From an Israelite perspective, however, the narrative of the journey to the land of Sihon provides a vehicle for a quotation from the Book of the Wars of the Lord that establishes the Arnon as Moabʼs northern boundary: Numbers 21:13–15 From there they set out and encamped beyond the Arnon, that is, in the wilderness that extends from the territory of the Amorites. For the Arnon is the boundary of Moab, between Moab and the Amorites. Therefore, the Book of the Wars of the Lord speaks of “. . . Waheb in Suphah, and the wadis: the Arnon with its tributary wadis, stretched along the settled country of Ar, hugging the territory of Moab . . . .”38

This citation vibrantly attests to the dispute between Moab and Israel concerning the northern boundary of Moab, which was the southern boundary of the Israelite kingdom in Transjordan.39 Therefore, we have to determine that the narrative of the war with Sihon and the Israelitesʼ route leading to the land of Sihon relates to the Israelite kingdom and therefore is to be ascribed to E.

35  See A. W. Jenks, The Elohist and North Israelite Traditions (SBLMS 22; Missoula, MT: Scholars Press, 1977), 55. 36  See R. E. Friedman, The Bible with Sources Revealed: A New View into the Five Books of Moses (San Francisco: HarperSanFrancisco, 2003), 278–279. 37  “Now the king of Israel had built Yahas, and he occupied it while he was fighting against me. But Kemosh drove him out from before me. I took from Moab two hundred men, its entire unit. I took it up against Yahas and captured it to annex (it) to Daibon” (K. P. Jackson, “The Language of the Mesha Inscription,” in Studies in the Mesha Inscription and Moab [ed. J. A. Dearman; ABS 2; Atlanta: Scholars Press, 1989], 98, ll. 18–21). 38  For a geographical explanation of the difficult verses, see D. B en -G ad H a C ohen , “ ‘Waheb in Suphah’ and ‘The Settled Country of ʿAr’: Comments on the Route of the Israelites in Moab,” Cathedra 95 (2000), 8–24 (Hebrew). The current essay strengthens the conclusions I reached in my earlier article. 39  The debate focuses on the question whether the previous boundary on the Arnon was in Wadi Mujib or in Wadi eth-Themed, its northern tributary that reaches Jahaz. There is no way to determine if this was a practical disagreement pertaining to the foreign relations of the kingdom of Israel or merely a theoretical debate that the biblical narrator sought to answer. On the boundary indicated in the Mesha Stela, which passed through Wadi Heidan and which changed names from west to east to Wadi Wala and Wadi eth-Themed, see J. A. D earman , “The Levitical Cities of Reuben and Moabite Toponymy,” BASOR 276 (1989), 55–66, esp. 57, 63–64.

Biblical Criticism from a Geographerʼs Perspective

697

Roskop devotes a lengthy discussion to proving that the journey description that uses the “they set out . . . and they encamped” style (journey B) is based on the genre of Assyrian annals that depict military campaigns of conquest.40 She also points out all the verses that describe the Israelites who left Egypt as an army.41 All these verses are Priestly,42 and therefore Num 33, which is formulated in annals style, is to be ascribed to P.43 The Pentateuch contains two traditions regarding the burial place of Aaron. One situates the event at Moserah (Deut 10:6), and the other places it at Mount Hor (Num 20:22–29; 33:38–39; Deut 32:51). The Mount Hor tradition links the death of Aaron to the episode of the waters of Meribah (Num 20:24; Deut 32:51), a narrative that based on its context, language, and perspective belongs to P:44 the Lord tells Moses to take the “rod,” and he takes “the rod from before the Lord, as he had commanded him” (Num 20:8–9). The rod that is before the Lord is known to us as Aaronʼs staff in the Priestly narrative that precedes that of the waters of Meribah (Num 17:24–25); this staff is meant to be ‫לאות לבני מרי‬, “a lesson to rebels,” which is how Moses and Aaron address the community in the Meribah episode (‫שמעו נא המורים‬, “Hear now the rebels,” Num 20:10b). The wording “If only we had perished when our brothers perished before the Lord” (20:3b), too, refers to the end of the narrative of Aaronʼs staff: “Lo, we perish! We are lost, all of us lost! [. . .] Alas, we are doomed to perish!” (Num 17:27–28). Following the Priestly format, Aaron appears alongside Moses throughout most of the narrative; the term ‫כבוד יהוה‬, “the glory of the Lord” (Num 20:6); the verb ‫הקהל‬, “assemble” (Num 20: 8, 10); and the noun ‫קהל‬, “congregation” (Num 20:4, 6, 12) are clearly Priestly.45 Consequently, the narrative of the waters of Meribah,

 See Roskop, Wilderness Itineraries (see n. 17), 144–149.  See Roskop, Wilderness Itineraries (see n. 17), 150–152. 42  See Exod 6:26; 7:4; 12:17, 41, 51; Num 1–2; 4:3, 23, 30, 35, 39; 8:24–25; 10:12; 26:2; 31:3–7, 14, 21, 27, 32, 36, 42, 48, 53; 32:27. F riedman , Bible with Sources Revealed (see n. 36), assigns Exod 6:26; 12:51 to R and Num 32:27 to J. He assigns the rest of those verses to P. I assign all of them to P. 43  Despite, and in opposition to, the proofs that she adduces, Roskop states that Num 33 is not Priestly but rather is a collection of all the itinerary notes, with the addition of sites from the books of Deuteronomy and Joshua that were reformulated in a style imitative of the annals style. See Roskop, Wilderness Itineraries (see n. 17), 223–232, and above, n. 21. 44  See Bacon, Triple Tradition (see n. 25), 149, 196, 202–204; Noth, Numbers (see n. 7), 145–147; M. Haran, Biblical Collection: Its Consolidation to the End of the Second Temple Times and Changes of Form to the End of the Middle Ages (4 vols.; Jerusalem: Mossad Bialik, 1996–2004) 2:169–170 (Hebrew); G. B. G ray , A Critical and Exegetical Commentary on Numbers (ICC 4; Edinburgh: T&T Clark, 1986 [1st ed., 1903]), 256–259. 45  See P. J. Budd, Numbers (WBC 5; Waco, TX: Word, 1984), 216, and additional bibliography. For “the glory of the Lord,” see Exod 16:7, 10; 24:16, 17; 40:34–35; Lev 9:6, 23; Num 14:10, 21; 16:19; 17:7 (all P); for ‫קהל‬, see Gen 28:3; 35:11; 48:4; 49:6; Exod 12:6; 16:3, and more. 40 41

698

David Ben-Gad HaCohen

and together with it Aaronʼs death at Mount Hor, are to be ascribed to P. This connection gains added force in light of the date given for Aaronʼs death: “in the fortieth year after the Israelites had left the land of Egypt, on the first day of the fifth month” (Num 33:38b), in characteristic P format.46 The death of Aaron at Mount Hor, and the annals style of Num 33 suffice to ascribe the itinerary list, and with it journey B, to P.47 Journey C looks like an independent list not being connected to any previous event or journey.48 Nevertheless, the list has a continuation. Two of the sites on this route – Bamoth and the summit of Pisgah – are locations where Balak builds altars for Balaam (Num 22:41, 23:14). Accordingly, we can naturally assume that journey C and these passages in the Balaam narrative belong to a single source. Two sources are intertwined in the Balaam narrative.49 The third place to which Balak took Balaam is the summit of Pisgah (Num 23:14), which is known to us from journey C (Num 21:20), together with the description “overlooking the wasteland.” A similar description is attached to the peak of Peor in the Balaam episode: “Balak took Balaam to the peak of Peor, which overlooks the wasteland” (Num 23:28). We cannot assume that a single author used the same characterization for different peaks50 or that he gave different names to the same mountain. As both Pisgah and the peak of Peor appear in the Balaam narrative, we may reasonably assume that the author who used the toponym “the summit of Pisgah” is not the same author who preferred the toponym “the peak of Peor.”  See the dates included in the five itinerary notes that belong to P: Exod 16:1; 19:1; Num 10:11; 20:1; 33:3. The date format usually contains the year, the month, and the day, along with the definition that this date is in relation to the time of the exodus. 47  Noth attributed Kadesh, Mount Hor, Oboth, and Iye-abarim to P. See N oth , “Der Wallfahrtsweg” (see n. 7), 171; idem , Numbers (see n. 7), 159. Davies argued that Num 33 is an independent source. In order to negate its attribution to P, he made the unsubstantiated assertion that Mount Hor was copied here from Num 21:42, which he ascribed to a nonPriestly tradition. See Davies, “Wilderness Itineraries” (see n. 7), 1–8, esp. 8, 11. As we have shown here, the mentions of Mount Hor prove that these texts belong to P. 48  Elsewhere I have proposed viewing the opening verse of Deuteronomy (Deut 1:1) as a route leading from the Edomite plateau to north of the Arnon. Journey C is the only route that could be the continuation of this journey. See D. Ben-Gad HaCohen, “Kadesh in the Pentateuchal Narratives” (PhD diss., Hebrew University of Jerusalem, 2010), 143–145 (Hebrew). 49  Wellhausen listed a series of reasons for the division of the narrative. See Wellhausen, Die Composition des Hexateuchs (see n. 31), 110–111. Only the geographical argument has withstood criticism. For a summation of the objection to the method of Wellhausen and his school regarding the Balaam narrative, see A. Rofé , The Book of Balaam: Numbers 22.2– 24.25; A Study in Methods of Criticism and the History of Biblical Literature and Religion (Jerusalem: Simor, 1979), 13–26 (Hebrew). 50  We may surmise that various mountains overlooked the wasteland. Once an author applied this description to a specific mountain, however, it characterized this mountain alone in his narrative. 46

Biblical Criticism from a Geographerʼs Perspective

699

One author used the geographical name (Deut 3:27; 4:49; Josh 12:3; 13:20), whereas the other called the mount “the peak of Peor,” after the name of the adjoining settlement, Beth-peor.51 The Balaam narrative opens with the description, “Balak son of Zippor saw all that Israel had done to the Amorites” (Num 22:2). Up to this point in the narrative, only Sihon has been called “king of the Amorites” (Num 21:21, 26, 29, 34).52 Since all the references to the war with Sihon draw their data from E, this verse belongs to E. The verses that depict the Israelites as a people “so numerous” (Num 22:3a, 4a, 5b, 6a), however, are not compatible with the description of the people in E. E does not enumerate the number of the Israelites, but the wording used by Moses both in his address to the king of Edom and to Sihon – “We will follow the kingʼs highway, turning off neither to the right nor to the left” (Num 20:17b, and in abbreviated fashion, 21:22b1) – reveals that, in the E conception, a single highway would have sufficed for all the Israelites, without their spilling over to the sides.53 Furthermore, “chiefs of thousands” was the highest rank in the reorganization of the judicial system, as portrayed in E (Exod 18:21, 25) – all those who went forth from Egypt did not require chiefs of myriads. The people described by E numbered only a few thousand souls, as is indicated also by the promise in E to expel the Canaanites from the land. This would be done “little by little,” in stages, “until you have increased” (Exod 23:29–30). This assurance is based on the assumption of a small number of Israelites. William Flinders Petrie maintained that not more than 5,000–7,500 Israelites left Egypt. He drew this conclusion based on three details: the war with Amalek, in which the two sides seem to be evenly matched (Exod 17:11); the area of the land of Goshen being identified with Wadi Tumilat, which had a population of some 20,000 at the end of the nineteenth century; and the appointment of only seventy elders to lead the people (Num 11:16, 24).54 Flinders Petrie did not distinguish between the various sources from which the Pentateuch was composed and therefore was not aware  Because of the mention of Pisgah, and since Eusebius identifies Peor with the Transjordanian plateau, one might identify Beth-peor with Mount Peor. Eusebiusʼs entry for Peor, however, refers to Mount Peor and not to Beth-peor, whereas the verse speaks not of Pisgah but of ‫“( אשדות הפסגה‬the slopes of Pisgah”), which are the western slopes of the mountain called “Pisgah” (Deut 3:17; with slight variation, 4:49; Josh 12:3). See K lostermann , Eusebius Onomastikon (see n. 27), 48, ll. 3–4 (βεθφογόρ, Betpeor); 168, ll. 25–27 (Φογώρ, Peor). Bethpeor is to be identified with the Iron Age site located at Uyun Musa, in Wadi Uyun Musa, to the north of Ras al-Siyagha – this is the mountain to be identified with “the peak of Pisgah.” See MacDonald, East of the Jordan (see n. 28), 138–139. 52  Og, who is also called “the Amorite king,” is given this title only in Deut 3:8. Furthermore, the Og narrative apparently was copied to Numbers (21:33–35) from D (Deut 3:1–7). See A. R ofé , Introduction to the Literature of the Hebrew Bible (trans. H. N. Block and J. H. Seeligman; Jerusalem Bible Studies 9; Jerusalem: Simor, 2009), 279–280. If this is the case, then E knew of only a single Amorite king, Sihon. 53  On the attribution of these events to E, see above, p. 696. 54  See W. M. Flinders Petrie, Researches in Sinai (London: Murray, 1906), 207, 211, 214. 51

700

David Ben-Gad HaCohen

that both the war with Amalek and the appointment of the seventy elders belong to E.55 In contrast with E, in J the Israelites are numerous: “The people who are with me number six hundred thousand men” (Num 11:21a).56 P, too, enumerates those who left Egypt: “about six hundred thousand men on foot, aside from children” (Exod 12:37; and in proximate formulation, 38:26; Num 1:46; 26:51), but the Balaam narrative does not contain any datum that can be ascribed to P. Consequently, we can draw the following conclusions: 1. two sources – E and J – are intertwined in the Balaam narrative; 2. one author used the toponym “the summit of Pisgah,” whereas another referred to the same summit as “the peak of Peor”; 3. the author of E linked the Balaam episode with the victory over the Amorite king, namely, Sihon; 4. journey C led to the summit of Pisgah, without taking into account the local king (Sihon): he is not addressed nor is war waged against him. Consequently, “the summit of Pisgah” in the Balaam narrative does not belong to E, and the use of this toponym should be attributed to J. The summit of Pisgah leads us directly to the depiction of the death of Moses, in which three sources are intertwined: Deuteronomy 34:1–5 Moses went up from the steppes of Moab to Mount Nebo, to the summit of Pisgah, opposite Jericho, and the Lord showed him the whole land. [. . .] And the Lord said to him, “This is the land of which I swore to Abraham, Isaac, and Jacob, ‘I will assign it to your offspring.’ ” [. . .] So Moses the servant of the Lord died there, in the land of Moab, at the command of the Lord.

The ascent to Mount Nebo (Deut 34:1a) is the implementation of the command in P: “Ascend these heights of Abarim to Mount Nebo, which is in the land of Moab facing Jericho” (Deut 32:49); the reference to Moses as “the servant of the Lord” (34:5) appears elsewhere only in E (Exod 14:31; Num 12:7), and the oath to the patriarchs is quintessentially J.57 We possess three different routes for the journey in Transjordan: (1) Journey A, with the stations Kadesh, the wilderness of Moab, Wadi Zered, “beyond the Arnon,” and Jahaz, belongs to E. This source uses the style “from there 55  Joshua son of Nun is one of the key characters in the war with Amalek and in the episode of the seventy elders (Exod 17:9; Num 11:28). Joshua appears in the early stages of the exodus only in E. 56  The Kibroth-hatta‍ʾavah narrative (Num 11:4–34) is composed of two sources. The episode of the seventy elders belongs to E and the complaint about the manna and the quail to J. See B. D. Sommer, “Reflecting on Moses: The Redaction of Numbers 11,” JBL 118 (1999), 601–624. The verse relating to the number of the Israelites belongs to the quail narrative and, therefore, to J. 57  Only in J does the promise of the land precede that of offspring. See J. S. Baden, The Promise to the Patriarchs (Oxford: Oxford University Press, 2013), 112–119.

Biblical Criticism from a Geographerʼs Perspective

701

they set out and encamped at Y.” (2) Journey B, which belongs to P, has the stations Kadesh, Mount Hor, Zalmonah, Punon, Oboth, Iye-abarim, Dibon-gad, Almon-diblathaim, the hills of Abarim, and the steppes of Moab. P uses the journey formula that has both traveling and encampment verbs and a starting point and an encampment site: “they set out from X and encamped at Y.” (3) Journey C describes the route concisely, without verbs: “from X to Y, and from Y to Z.” This route, which belongs to J, contains the stations Beer, (Midbar,) Mattanah, Nahaliel, Bamoth, “the valley that is in the territory of Moab,” and “the peak of Pisgah.”

4 The Original Journey of E The E version, journey A, is known to us in three descriptions, and we should examine the relationship between them. The description in Deuteronomy, although similar in content and at times even in wording to the E version of the Israelite journey included in Numbers, varies stylistically and exhibits the singular Deuteronomic literary style. The speaker in D is always Moses, and so the narrative is related in the first person. It also incorporates various expansions and statements that accord with the reformist spirit of the Deuteronomic circle.58 This, in itself, should enable us to determine that the Deuteronomic narrative is not the original version. These considerations, however, did not prevent Van Seters from positing the originality of the Deuteronomic Sihon narrative.59 Here as well, the geography is precise and unequivocal. In order to prove this, we will turn to the geography of the Edom narrative. Each of the three versions depicts the negotiations with Edom (Num 20:14–21; Deut 2:4–8; Judg 11:17). The versions in Numbers and in Jephthahʼs speech are the closest to each other. In each, the narrator situates the episode at Kadesh, from which a delegation is dispatched to the king of Edom with the request to pass through his land, a petition that is denied by the Edomite king in both versions. The Israelites do not pass through Edom and sometime later set out from Kadesh on a journey that circumvents this land (Num 21:4; Judg 11:18). Deuteronomy (2:4–8a) presents a tale that differs in the placement of this meeting in its sequence of events, its participants, and its consequences. The venue of the meeting is not Kadesh and  See S. E. Loewenstamm, From Babylon to Canaan: Studies in the Bible and Its Oriental Background (Jerusalem: Magnes, 1992), 42–50; H aran , Biblical Collection (see n. 44), 169–70; Moshe Weinfeld, Deuteronomy and the Deuteronomic School (Oxford: Clarendon, 1972), 51 ff. 59  Van Seters based his conclusion on the dominance of D in the Og narrative. See Van S eters , “Conquest of Sihonʼs Kingdom” (see n. 24), 182–184. His view regarding the war with Og may be accepted, but this proves nothing regarding the war with Sihon and the journey in Transjordan. 58

702

David Ben-Gad HaCohen

occurs after the circumvention of Mount Seir; it mentions neither a delegation nor the king of Edom; and the inhabitants of Seir permit the Israelites to pass through their land and apparently even give them food and water, as can be understood from Mosesʼs petition to Sihon: Deuteronomy 2:28–29 What food I eat you will supply for money and what water I drink you will furnish for money; just let me pass through – as the descendants of Esau who dwell in Seir did for me.

The narrative in Deuteronomy is not self-standing and is part of a sermon with the directives for a journey that the Lord gives Moses. The very presence of such a sermon is indicative of the Deuteronomic school and is unknown in the other sources. Furthermore, the need for the detour is a direct consequence of the Edomite kingʼs refusal of passage through his land (Num 20:18, 20–21; Judg 11:17). D sets forth the narrative of bypassing Seir without giving any reason for the detour (Deut 2:1b). We can imagine why D diverged from the pattern in the parallel depictions. His worldview presented the people of Israel, with the Lord before it, as a body that no power could resist:60 Deuteronomy 2:25 This day I begin to put the dread and fear of you upon the peoples everywhere under heaven, so that they shall tremble and quake because of you whenever they hear you mentioned.

The Numbers-Judges narrative espouses the military superiority of the Edomite king, who was capable of stopping the Israelite march (Num 20:21; Judg 11:17). The Deuteronomic sermon explains why “omnipotent” Israel retreated in the face of the military threat posed by the king of Edom. The version of the delegation to Edom in D was altered by the Deuteronomic narrator. The E narrative, as reflected in Num 20 and in the speech by Jephthah, is concerned with Edom. This toponym does not appear in the D narrative, where it is systematically replaced by Seir. One may assume that the authorʼs desire to be understood by his readership led him to use names that were common in his time, even if they differ from the accepted names during the period of which he writes.61 The toponym shift, from Edom to Seir, attests not only to a different author but also to a different period. When we add to this the differences in the Edom narrative between D and its parallels, and especially the journey around Seir, which in D is superfluous, we realize that the narrative in D of the delegation to Edom was altered by the Deuteronomic narrator. Consequently, journey A in D is not the original version of the E narrative. The speech by Jephthah, which relates to the journey of the Israelites in Transjordan, contains, in addition to the petition to the king of Edom, a request 60 61

 See Sumner, “Israelʼs Encounters” (see n. 24), 221.  See MacDonald, East of the Jordan (see n. 28), 4, 14.

Biblical Criticism from a Geographerʼs Perspective

703

addressed to the king of Moab that does not appear in Numbers or Deuteronomy: “They also sent a mission to the king of Moab, and he refused” (Judg 11:17b). According to its position in the description, the appeal was made at Kadesh. The tangible geographical datum that the Bible supplies for the location of Kadesh is that it is on the border with Edom (Num 20:14). The Aramaic Targumim to the Pentateuch, which always render Kadesh as Rekem,62 allow us to position it with greater precision. Rekem is not only a term employed by the Targumim; it is a real site that appears in the writings of Josephus,63 is mentioned in the Tannaitic literature,64 and was known as far away as China.65 All of these references, as well as the inscription “Petraios son of Threptos,”66 have led scholars to identify Rekem with Petra;67 and from this location, which is some distance from the boundary of Moab, there was no reason to appeal to that landʼs ruler. 62  Kadesh is mentioned in Num 13:26; 20:1, 14–16, 22; 27:14; 33:36–37; Deut 1:46. Tg. Pseudo-Jonathan to Num. 27:14 renders it as ‫ריקם‬, whereas in all the other verses and in all the other Targumim it appears in the form ‫רקם‬. See I. Drazin, Targum Onkelos to Numbers (Hoboken, NJ: Ktav, 1998); idem , Targum Onkelos to Deuteronomy (Hoboken, NJ: Ktav, 1982); E. G. Clarke, Targum Pseudo-Jonathan: Numbers (ArBib 4; Edinburgh: T&T Clark, 1994); idem , Targum Pseudo-Jonathan: Deuteronomy (ArBib 5; Edinburgh: T&T Clark, 1998); A. Díez Macho, Neophyti 1: Targum Palestinese MS de la Biblioteca Vaticana, vol. 4; Numeros (Madrid: Consejo Superior de Investigaciones Cientificas, 1974). 63  Josephus (Ant. 4:161) writes: “the city which bears his name [= Rekem, king of Midian, Num 31:8] ranks highest in the land of the Arabs and to this day is called by the whole Arabian nation, after the name of its royal founder, Rekeme: it is the Petra of the Greeks”; see also 4:82 (Josephus [trans. H. S. J. Thackeray; 9 vols.; London: William Heinemann; Cambridge, MA: Harvard University Press, 1934]). Josephus is followed by Eusebius; Klostermann, Eusebius Onomastikon (see n. 27), 36 l. 13; Notley and Safrai, Eusebius, Onomasticon (see n. 27) 39 (Άρχέμ, Arcem); Klostermann, Eusebius Onomastikon, 144, ll. 7–9; Notley and Safrai, Eusebius, Onomasticon, 136 (Ῥεχέμ, Rekem). 64  See D. Ben-Gad HaCohen, “The Southern Boundary of the Land of Israel in Tannaitic Literature and the Bible,” Cathedra 88 (1998), 15–38, esp. 24–27 (Hebrew), with a bibliography for all the references to Rekem. 65  A series of ancient Chinese writings mentions the land of Ta-ts`in as an important trading center. See F. Hirth, China and the Roman Orient (Leipzig: Hirth; Shanghai: Kelly & Walsh, 1885), 35–96. According to some of the sources, Ta-ts`in was also called Li-Kan/Li-kien/ Li-ken/Li-chien. See Hirth, China, 40, E:1; 44, F:1; 48, I:1; 68, P:5; 77, Q:1; 92, R:1. Hirth concluded from an examination of the geographical data of this land that the reference was to the Nabataean kingdom (172–173) and that Li-kan is the Chinese phonetic transliteration of Rekem = Petra (170, n. 1). 66  Petraios was buried in Gerasa, but the tomb was apparently insufficient for his relatives, who desired to commemorate him also in his city of Petra, which preserved its Semitic/ Nabatean name, Rekem. This is indicated by an inscription discovered in 1964 at the entrance to Petra: “This is the (place of the) soul of Petraios / son of Threptos who lived honorably / in Rekem and died / in Gerasa and was buried there / [in the tomb] that his son Taimu made for him”; see J. Starcky, “Nouvelle Epitaphe Nabateenne,” RB 72 (1965), 95–97. 67  See MacDonald, East of the Jordan (see n. 28), 68–69; Davies, Way of the Wilderness (see n. 7), 16–18; P. C. H ammond, “New Evidence for the 4th-Century A.D. Destruction of

704

David Ben-Gad HaCohen

Who was responsible for the story of the address to the king of Moab? Was this the Deuteronomist narrator, who wrote the book of Judges? Or, perhaps, the Elohist author who wrote the main part of the Jephthah narrative?68 Or, possibly, is the entire speech of Jephthah a later addition that was copied from E, with the incorporation of the king of Moab in order to draw the narrative closer to Ammon? The encounter with Moab in the speech by Jephthah presents a conception different from that set forth in D. The Deuteronomic author maintains that “the Moabites who dwell in Ar” sold food and water to the Israelites (Deut 2:28–29), whereas according to the speech by Jephthah, the king of Moab refused the Israelites passage through his land, and there was no contact between the two peoples.69 It can hardly be assumed that the Deuteronomistic author disagreed with D. Moreover, the speech shows no characteristically Deuteronomistic language. Consequently, the appeal to the king of Moab, and Jephthahʼs entire speech, are not the products of the Deuteronomistʼs quill. George Foot Moore showed that Jephthahʼs address to the king of the Ammonites breaks the continuity of events and separates Jephthahʼs petition to the Lord (Judg 11:11b) from its natural continuation of the vow to the Lord (11:30–31).70 Additionally, the mention of Chemosh as the god of Ammon (11:24a) shows that the passage was written in a period in which the author no longer distinguished between Chemosh, the god of Moab, and Milcom, the deity of Ammon. 71 It also seems that the description of the Israelite journey, which begins Jephthahʼs speech, does not belong to the current narrative: “Israel traveled through the wilderness to the Sea of Reeds and came to Kadesh” (11:16b). According to this verse, the Israelites who had left Egypt proceeded in the direction of the Sea of Reeds, but they did not reach their objective, since they halted at Kadesh.72 This section of the journey is of no concern to the king of the Ammonites and is superfluous to the Jephthah narrative. It does, however, suit E, as it connects Petra,” BASOR 238 (1980), 65–67. 68  Moore argued that the Jephthah narrative is mainly Elohist; see G. F. Moore, The Book of Judges (London: James Clarke, 1898), 22–25, 77–81. Burney, in contrast, maintained that this narrative is the redactional product of two narratives, one from E, concerning the war with Ammon, and the other from J, relating to the war with Moab; see C. F. Burney, The Book of Judges, with Introduction and Notes (New York: Ktav, 1970), 298–304. 69  On Ar, the Moabites who dwell there, and the Moabites who dwell in Moab, see BenGad HaCohen, “Waheb in Suphah” (see n. 38), 9–13. 70  See M oore , Book of Judges (see n. 68), 77; R. S mend , “Beiträge zur Geschichte und Topographie des Ostjordanlandes,” ZAW 22 (1902), 134–137. 71  See Burney, Book of Judges (see n. 68), 299. 72  On the directional meaning of the preposition ʿad, see H. L. Ginsberg, “A Preposition of Interest to Historical Geographers,” ASOR 122 (1951), 12–14. It should be stressed that ‫ים‬ ‫סוף‬, “Sea of Reeds,” mentioned here is not the sea mentioned in Exod 13:17–15:22; the sea here could be identified with Suph in Transjordan (Deut 1:1).

Biblical Criticism from a Geographerʼs Perspective

705

the Dathan and Abiram episode (Num. 16–17) with the events at Kadesh. The insertion of the speech within the Jephthah narrative, the mention of the king of Moab, and the journey toward the Sea of Reeds indicate that the speech was not an integral part of the narrative. Therefore we must conclude that the speech was copied into the Jephthah narrative from E. All three descriptions of the war with Sihon mention the Wadi Arnon. D, according to his usual way of writing, transforms the crossing of the stream into a directive from the Lord (Deut 2:24a): “Up! Set out across the Wadi Arnon!” The two other descriptions are almost identical, with only a small, though highly significant, difference: “and they encamped [va-yahanu] beyond the Arnon” (Num 21:13a2), as opposed to “until they encamped [va-yahanun] on the other side of the Arnon” (Judg 11: 18b2).73 In the speech of Jephthah, the verb ‫חנ״ה‬, “to encamp,” appears with a paragogic nun, and in Numbers, in the regular future. The paragogic nun is an early grammatical form that in all the versions of the Sihon narrative is preserved only in Jephthahʼs speech,74 revealing that grammatical changes were made in the description of journey A in Numbers. Additionally, we learn that the portrayal of the arrival at Kadesh, according to E, appears only in the speech by Jephthah and not in Numbers. We may therefore conclude that the depiction of journey A in Num 20–21 is not identical with the E version that was in the hands of the Pentateuchʼs compiler. Our analysis leads to three conclusions: 1. The proximity between the three depictions of journey A shows that they are not based on an oral tradition but on a true document: E. 2. Each of the three authors of the descriptions of journey A possessed the E document,75 and each introduced changes in copying the narrative, in accordance with the needs of the composition. 3. The quotation from the speech of Jephthah the Gileadite – “Israel traveled through the wilderness to the Sea of Reeds and came to Kadesh” – is an original E passage that is not present in the Pentateuch but that survived in the speech of Jephthah.76 73  The difference between the articles bet and mem preceding the word ʿever is not significant, since this could be a copyistʼs mistake ensuing from the graphic similarity of the two letters. See E. Tov , The Text-Critical Use of the Septuagint in Biblical Research (2nd ed.; Jerusalem: Simor, 1997), 135–39, esp. 137. 74  On the paragogic nun, see J. Hoftijzer, The Function and Use of the Imperfect Forms with Nun Paragogicum in Classical Hebrew (Assen: Van Gorcum, 1985); S. A. K aufman , “Paragogic Nun in Biblical Hebrew,” in Solving Riddles and Untying Knots: Biblical, Epigraphic, and Semitic Studies in Honor of Jonas C. Greenfield (ed. Z. Zevit at al.; Winona Lake, IN: Eisenbrauns, 1995), 95–99. 75  Thus Sumner, “Israelʼs Encounters” (see n. 24), 223: “they are at more than one literary remove from each other.” 76  Noth, who was unsuccessful in identifying the original verse of E, thought that this document has no literary connection between the wilderness and the boundary of Edom.

706

David Ben-Gad HaCohen

One main objection to the Documentary Hypothesis is that there are no independent surviving manuscripts of any of the sources that would predate their incorporation into the Pentateuch.77 The original E verse that survived in Jephthahʼs speech constitutes an important step in refuting the arguments of the skeptics.

5 An Accolade to the Compiler of the Pentateuch Comparisons of the parallel sources that describe the arrival at Kadesh provide us with a rare look at the work of the compiler: Numbers 20:1 The Israelites arrived in a body at the wilderness of Zin on the new moon, and the people stayed at Kadesh. Miriam died there and was buried there.

The combination ‫בני ישראל כל העדה‬, “the Israelites – a body,” appears only one other time, in the narrative of the death of Aaron (20:22), which is evidently Priestly. The Kadesh–wilderness of Zin combination, as well, is to be found in Priestly verses (Num 27:14; 33:36; Deut 32:51). The end of the verse portrays the death of Miriam and does not belong to P. Every appearance of Miriam as an active and lively character either belongs to E or was influenced by it: the Song of Miriam at the sea (Num 15:20–21),78 the episode of the skin affliction (Num 12:1–16),79 This led him to the erroneous conclusion: “This makes it clear that the wilderness theme and the conquest theme in the Pentateuch have not been synchronized with each other from the very beginning, but stand side by side in relative independence and have been only loosely interwoven”; see Noth, Numbers (see n. 7), 223. 77  See J. K. H offmeier , Ancient Israel in Sinai: The Evidence for the Authenticity of the Wilderness Tradition (Oxford: Oxford University Press, 2005), 14; T. T. S tewart , Solving the Exodus Mystery, Volume. 1: Discovery of the True Pharaohs of Joseph, Moses, and the Exodus (Lubbock, TX: Biblemart, 2003), 4–5; A. Rosengren, “Why Is There a Documentary Hypothesis, and What Does It Do to You If You Use It? A Response to David Clines,” SBL Forum, http://www.sbl-site.org/Article.aspx?ArticleId=566 (accessed 07/2006). 78  The subject of this passage is “Miriam the prophetess” (15:20), and as usual, “the narratives depicting ecstatic prophesying [. . .] belong to E”; Haran, Biblical Collection (see n. 44), 157. It should be mentioned that Noth ascribes to J all the references to Miriam. The body of proofs, however, in this and the following notes tips the scales in favor of E. See N oth , Numbers (see n. 7), 92–94. Following his general conception regarding Miriam, Noth also attributes her death (Num 20:1b) to J; see Noth, Numbers, 144–45. 79  The tent of meeting is the venue of the event. This is the Elohist tent, in which the Lord is revealed to Moses (Exod 33:7–11) and which was accessible to “whoever sought the Lord” (v. 7b). Additionally, the claim by Miriam and Aaron – “Has the Lord spoken only through Moses? Has he not spoken through us as well?” (Num 12:2) – defines both as prophets according to the Elohist tradition (see the preceding note). On the tent of meeting as the place of meeting with the Lord, in contrast with the tent that served as the dwelling place of the Lord, see I. K nohl , The Divine Symphony: The Bibleʼs Many Voices (Philadelphia: Jewish Publication Society of America, 2003), 74–77.

Biblical Criticism from a Geographerʼs Perspective

707

and the mention of this affliction (Deut 24:9),80 to which we might add Mosesʼs sister, who appears as an active but unnamed figure in the story of Mosesʼs birth (Exod 2:4–8).81 Only E views Miriam as a historical figure who should be recalled and, therefore, whose death should be recorded as well. Again, the death formulation, “Miriam died there and was buried there,” appears only once more, in an Elohist passage (Deut 10:6): “Aaron died there and was buried there.”82 Thus, the verse that begins the Meribah episode comes from two sources. Its beginning belongs to P and its end to E. How did this combination come about? We may assume that the death of Miriam appeared in E in a verse that began with a report of the journey to Kadesh (Judg 11:16b + Num 20:1b): “Israel traveled through the wilderness to the Sea of Reeds and came to Kadesh. Miriam died there and was buried there.” The compiler of the Pentateuch possessed two sources relating to the Israelitesʼ arrival at Kadesh. As he was already copying the law of the red heifer from P (Num 19:1–22), he continued with the same source, “The Israelites arrived in a body at the wilderness of Zin on the new moon, and the people stayed at Kadesh,”83 to which he added the E information about the death of Miriam. The new compiled verse lacks the E information about the journey toward the Sea of Reeds. The missing information is documented in other two verses: Numbers 14:25b Start out, then, tomorrow and march into the wilderness by way of the Sea of Reeds. Deuteronomy 1:40 As for you, turn about and march into the wilderness, toward the Sea of Reeds. 80  Although the mention of the skin affliction belongs to D, the Deuteronomic author apparently derived this from E. On E as the source of Dʼs historical data, see Haran, “Exodus Routes” (see n. 28), 308–309; idem, Biblical Collection (see n. 44), 200. 81  The mention of Miriam as the sister of Moses is doubtful, since E refers to her as “Aaronʼs sister” (Exod 1:20), which seemingly severs the family connection between her and Moses. See Noth, Numbers (see n. 7), 94. Miriam is not mentioned in the family genealogy in Exodus (6:20) but is included in the list in Numbers (26:59); both lists belong to P. At any rate, this Miriam is just a name, and P does not infuse it with vitality. 82  This passage is foreign in its formulation and language to D. Thus H aran , Biblical Collection (see n. 44), 72, n. 58. 83  Bacon has suggested that the verse originally also included the year of the exodus from Egypt; see Bacon, Triple Tradition (see n. 25), 202. In every instance in which P specifies a date that includes a month, it also specifies the year since the exodus (Exod 16:1; 19:1; 40:17; Num 1:1; 9:1; 10:11; 33:38). Accordingly, the P version is apparently to be reconstructed as follows: “The Israelites arrived as a body at the wilderness of Zin on the first new moon [of the Xth year after their departure from the land of Egypt], and the people stayed at Kadesh.” The Pentateuch does not contain a date for Miriamʼs death, but a tradition cited by Josephus places her death in the fortieth year after the exodus; see Josephus, Ant. 4.78. It seems to me that because of the earlier arrival at Kadesh, the compiler of the Pentateuch was forced to omit the year of their arrival. On the earlier arrival at Kadesh according to P, see D. Ben-Gad HaCohen, “Kadesh: The Hidden Toponym,” Shnaton 15 (2005), 3–19, esp. 15–16 (Hebrew).

708

David Ben-Gad HaCohen

Despite their similarity, there are two seemingly minor differences between these verses that are of great importance for their chronology: Numbers 14:25b

Deuteronomy 1:40

1

tomorrow, start out, then,

As for you, turn about

2

into the wilderness (hammidbār)

into the wilderness (hammidbārâ)

Jan Joosten has used a series of examples to show that the syntax in which the description of the time appears first, followed by the verb in the imperative, as with (literally) “tomorrow, start out,” is characteristic of Late Biblical Hebrew and is to be found only in Chronicles, Ezra, Nehemiah, Ecclesiastes, Daniel, and Esther.84 He also demonstrates that the directional particle -â in the word hammidbārâ (“into the wilderness,” Deut 1:40) is one of the outstanding markers of Classical Biblical Hebrew.85 The combination of these two differences shows that the verse in Deuteronomy preceded that in Numbers. We have a unique opportunity to view the evolution of a verse in the Pentateuch: “Israel traveled through the wilderness to the Sea of Reeds and came to Kadesh. Miriam died there and was buried there” (Judg 11:16b + Num 20:1b). Based on the geographical data in E, the Deuteronomic author formulated the directive, “As for you, turn about and march into the wilderness, toward the Sea of Reeds,” which he placed in the episode of the spies (Deut 1:40). The compiler of the Pentateuch rejected the E portrayal of the arrival at Kadesh in favor of the similar depiction in P. Kadesh appears in the P verse that was chosen and Miriam was added to it from E, but without mention of the journey “to the Sea of Reeds.” In order to complete the missing datum, which appeared in E but not in P, the compiler took the Deuteronomic variant of the spies narrative, changed it slightly to fit the syntax common in his time, and inserted the revised verse into the speech of the Lord in the J spies narrative: “Start out, then, tomorrow and march into the wilderness by way of the Sea of Reeds” (Num 14:25b). The original E verse was not included in the Pentateuch but survived verbatim in the speech of Jephthah (Judg 11:16b), and its two later unfoldings appear in the spies narrative, in Numbers and in Deuteronomy.

Summary There are three extant depictions of the route of the journey that led the Israelites to the war with Sihon, in Numbers, Deuteronomy, and Judges. The route includes a list of toponyms and events common to the three depictions. Since  See J. J oosten , “The Distinction between Classical and Late Biblical Hebrew as Reflected in Syntax,” Hebrew Studies 46 (2005), 327–39, esp. 337–338. 85  See Joosten, “Distinction” (see n. 84). 84

Biblical Criticism from a Geographerʼs Perspective

709

this is a journey record, it can be compared to parallel journey records. Such a comparison clearly shows that the toponym list of the Sihon journey is distinct from the P list appearing in Num 33, portions of which are also to be found in Num 20–21. The Sihon journey likewise differs from the list of toponyms leading to the peak of Pisgah. These differences between the lists suffice to demonstrate their different authorship in different documents. Despite the difficulty of locating these three routes by means of a literary analysis, they cannot be disregarded when subjected to geographic scrutiny. The close similarity between the three versions of the Sihon narrative shows that they are based on a written document – E. The minor variations between them indicate that none is the original version of E. The original E text was altered by various writers, but its traces can be discerned by a comparison of the three portrayals. E contains a verse describing the arrival of the Israelites at Kadesh: “Israel traveled through the wilderness to the Sea of Reeds and came to Kadesh.” The compiler of the Pentateuch rejected the description of E and used a similar depiction taken from P. Although the original E verse was not included in the Pentateuch, it survived verbatim in the speech of Jephthah (Judg 11:16b), and two later unfoldings appear in the spies narrative, in Numbers and in Deuteronomy. Beyond the specific division of the sources that our analysis has revealed, this article highlights the importance of geographical criticism as a research tool, to gain an awareness of and to distinguish between the sources composing the Pentateuch.

Early North Israelite “Memories” of Moab Israel Finkelstein and Thomas Römer Moab is a unique “laboratory” for studying materials in the Bible that portray what can be described as pre–late-monarchic historical realities. This is so because of the exceptional information in the Mesha Stela;1 the relatively detailed geographical references in the Bible; and the well-preserved late Iron I and Iron IIA sites, which have been dealt with in detail in recent publications.2 In what follows we wish to use these sources in order to reconstruct several “moments”  For recent treatments of historical issues, see, e.g., N. N a ‍ʾaman , “Royal Inscription versus Prophetic Story: Meshaʼs Rebellion according to Biblical and Moabite Historiography,” in Ahab Agonistes: The Rise and Fall of the Omri Dynasty (ed. L. L. Grabbe; London: T&T Clark, 2007), 145–183; A. L emaire , “La Stèle de Mésha et lʼhistoire de lʼancien Israël,” in Storia e tradizioni di Israele: Scritti in onore di J. Alberto Soggin (ed. D. Garrone and F. Israel; Brescia: Paideia, 1991), 143–169, here 153–157; idem , “La dynastie Davidique (Byt Dwd) dans deux inscriptions Ouest-Sémitiques du IXe s. av. J.-C.,” Studi Epigraphici e Linguistici 11 (1994), 17–19; idem , “The Mesha Stele and the Omri Dynasty,” in Ahab Agonistes: The Rise and Fall of the Omri Dynasty (ed L. L. Grabbe; London: T&T Clark, 2007), 135–144; P. B ordreuil , “A propos de lʼInscription de Mesha’ deux notes,” in The World of the Aramaeans, Vol. III: Studies in Language and Literature in Honour of Paul-Eugen Dion (ed. P. M. M. Daviau et al.; Sheffield: Sheffield Academic, 2001), 158–167; E. G ass , Die Moabiter: Geschichte und Kultur eines ostjordanischen Volkes im 1. Jahrtausend v.Chr. (Wiesbaden: Harrassowitz, 2009), 5–64. 2  For instance, B. Routledge, Moab in the Iron Age: Hegemony, Polity, Archaeology (Philadelphia: University of Pennsylvania Press, 2004); J. S. E. Lev-Tov et al., “Measuring Local Diversity in Early Iron Age Animal Economies: A View from Khirbat al-Mudayna al-ʿAliya (Jordan),” BASOR 361 (2011), 67–93; P. M. M. D aviau , “Hirbet el-Mudēyine in Its Landscape, Iron Age Towns, Forts and Shrines,” ZDPV 122 (2006), 14–30; P. M. M. Daviau and P. E. Dion, “Economy-Related Finds from Khirbat al-Mudayna (Wadi ath-Thamad, Jordan),” BASOR 328 (2002), 31–48; P. M. M. Daviau and M. Steiner, “A Moabite Sanctuary at Khirbat al-Mudayna,” BASOR 320 (2000), 1–21; C.-H. Ji, “Khirbet ʿAratuz: An Interim Overview of the Ten Years of Archaeological, Architectural Findings,” ADAJ 55 (2011), 561–579; idem, “The Early Iron Age II Temple at Hirbet Atarus and Its Architecture and Selected Cultic Objects,” in Temple Building and Temple Cult Architecture and Cultic Paraphernalia of Temples in the Levant (2.–1. mill. B.C.E.): Proceedings of a Conference on the Occasion of the 50th Anniversary of the Institute of Biblical Archaeology at the University of Tübingen (28–30 May 2010) (ed. J. Kamlah; Wiesbaden: Harrassowitz, 2012), 203–221; I. F inkelstein and O. Lipschits, “Omride Architecture in Moab: Jahaz and Ataroth,” ZDPV 126 (2010), 29–42; idem , “The Genesis of Moab,” Levant 43 (2011), 139–152. 1

712

Israel Finkelstein and Thomas Römer

in the history of Moab and Israel and shed light on the nature of the early memories on Moab that are embedded in the Bible.

Three Snapshots Let us start with three archaeological/historical snapshots: 1. A system of well-preserved, fortified sites south of the Arnon3 dates – according to their pottery and radiocarbon determinations – to the late Iron I, in the late eleventh and tenth centuries BCE.4 They were abandoned before 900 BCE and hence cannot be connected either to the Omride expansion north of the Arnon or to the time of Mesha. They create a line on the Arnon in the north and along Wadi el-Lejjun – its major tributary in the east5 – and should probably be interpreted as fortresses that protected the borders of an early Moabite polity south of the Arnon.6 The hub of this formation was perhaps located at Khirbet Balua, where a similar fort may be reconstructed.7 The Balua Stela8 probably 3  Three of them have been excavated: Khirbet el-Medeineh el-Muʿarrajeh (E. Olàvarri, “Sondeo Arqueologico en Khirbet Medeineh junto a Smakieh [Jordania],” Annual of the Department of Antiquities of Jordan 22 [1977–1978], 136–142; idem, “La campagne de fouilles 1982 à Khirbet Medeinet al-Muʿarradjeh près de Smakieh [Kerak],” Annual of the Department of Antiquities of Jordan 27 [1983], 165–178), Khirbet el-Medeineh ʿAliya (B. R outledge , “Seeing through Walls: Interpreting Iron Age I Architecture at Khirbat al-Mudayna al-ʿAliya,” BASOR 319 [2000], 37–70; idem, Moab in the Iron Age [see n. 2], 101–108; idem, “Thinking ‘Globally’ and Analyzing ‘Locally’: South-Central Jordan in Transition,” in Israel in Transition [ed. L. L. Grabbe; New York: T&T Clark, 2008], 144–176; Lev-Tov et al., “Measuring Local Diversity” [see n. 2], 67–93), and Khirbet el-Muʿmmariyya (F. Ninow, “First Soundings at Khirbat al-Muʿmmariya in the Greater Wadi al-Mujib Area,” Annual of the Department of Antiquities of Jordan 48 [2004], 257–266; idem, “The 2005 Soundings at Khirbat al-Muʿmmariya in the Greater Wadi al-Mujib Area,” Annual of the Department of Antiquities of Jordan 50 [2006], 147–155). 4  For the pottery, see R outledge , “Seeing through Walls” (see n. 3), 37–70; for the radiocarbon dates, see Lev-Tov et al., “Measuring Local Diversity” (see n. 2), 67–93, 72, n. 7. 5  Another site that may belong to this group is Khirbet el-Lehun (D. Homès Fredericq, Découvrez Lehun et la Voie Royale: Les fouilles belges en Jordanie [Brussels: Comité belge de fouilles en Jordanie, 1997]; idem, “The Iron Age II Fortress of al-Lahun (Moab),” in Studies of Iron Age Moab and Neighbouring Areas in Honour of Michèle Daviau [ed. P. Bienkowski; Leuven: Peeters, 2009], 165–182), located on the northern rim of the Arnon; for the possibility of early Moabite territorial formations that had their hub south of the Arnon but that included territory to its north, see below. 6  “Moab proper,” see J.M M iller , “The Israelite Journey through (around) Moab and Moabite Toponymy,” JBL 108 (1989), 577–595, here 577. 7  Finkelstein and Lipschits, “Genesis of Moab” (see n. 2). 8  E.g., B. Routledge and C. Routledge, “The Baluca Stela Revisited,” in Studies on Iron Age Moab and Neighbouring Areas in Honour of Michèle Daviau (ed. P. Bienkowski; Leuven: Peeters, 2009), 71–95, with previous bibliography.

Early North Israelite “Memories” of Moab

713

belongs to this period. Another important site may have been located at Kerak, buried under the medieval fort. The background for the rise of this early Moabite polity is a short period of improved climatic conditions in the Iron I, revealed in pollen residues recently counted in a core of sediments from the Dead Sea,9 and prosperity that stemmed from participation in – or even domination of – copper production in Wadi Faynan south of the Dead Sea.10 2. The Mesha Stela reports that the Omrides established two fortresses in Moab – Jahaz and Ataroth.11 Jahaz is identified with Khirbet el-Mudayna ethThemed12 and Ataroth with Khirbet Ataruz. The two Omride forts were built along Wadi el-Walla – the northern tributary of the Arnon, facing Dibon. Both places13 reveal typical construction methods of Omride sites west of the Jordan: an elevated podium, casemate wall, four-entry gate, elaborate glacis, and moat.14 Mesha names Omri as the king who conquered the area north of Wadi el-Walla – the mishor. It is only logical to ascribe the beginning of Israelite activity in Moab not too early in his days, or, if “Omri” stands for the dynasty, better later, in the days of Ahab or Joram, and its end immediately following the defeat of Israel in the battle of Ramoth-gilead in 842/841 BCE. This means that Omride rule in Moab was short-lived – no more than about thirty years, and perhaps less. 3. Mesha the Dibonite took advantage of the weakening of Israel under pressure from Hazael, his suzerain, in order to throw off the yoke of the Northern Kingdom. He took Jahaz, Ataroth, Madaba, and Nebo, that is, the entire mishor. It seems that Mesha also expanded in the opposite direction; lines 31–33 in the inscription apparently speak of his takeover of Horonaim, certainly located, according to geographical references in the Bible (in connection with Zoar; Isa 9  D. Langgut et al., “Dead Sea Pollen Record and History of Human Activity in the Judean Highlands (Israel) from the Intermediate Bronze into the Iron Ages (~2500–500 BCE),” Palynology 38/2 (2014), 1–23. 10  On all this, see Finkelstein and Lipschits “Genesis of Moab” (see n. 2); on the Faynan copper industry, see, e.g., T. E. Levy et al., “Reassessing the Chronology of Biblical Edom: New Excavations and 14C Dates from Khirbat en-Nahas (Jordan),” Antiquity 78 (2004), 865–879; idem , “High-Precision Radiocarbon Dating and Historical Biblical Archaeology in Southern Jordan,” Proceedings of the National Academy of Sciences 105/43 (2008), 16460–16465. 11  For the historical information in the inscription, see, e.g., N. Na‍ʾaman, “King Mesha and the Formation of the Moabite Monarchy,” IEJ 47 (1997), 83–92; idem, “Royal Inscription” (see n. 1), 145–183; Lemaire, “La Stèla de Mésha” (see n. 1); idem, “La dynastie Davidique” (see n. 1); Gass, Die Moabiter (see n. 1), 5–61. 12  J. A. Dearman, “The Location of Jahaz,” ZDPV 100 (1984), 122–126. 13  For Khirbet Mudeyne eth-Themed, see D aviau and S teiner , “Moabite Sanctuary” (see n. 2), 1–21; D aviau and D ion , “Economy-Related Finds” (see n. 2), 31–48; D aviau , “Hirbet el-Mudēyine” (see n. 2), 14–30; for Khirbet Ataruz, see Ji, “Khirbet ʿAratuz” (see n. 2), 561–579; idem, “Early Iron Age II Temple” (see n. 2), 203–221. 14  Finkelstein and Lipschits, “Omride Architecture” (see n. 2), 29–42; I. Finkelstein, “Omride Architecture,” ZDPV 116 (2000), 114–138.

714

Israel Finkelstein and Thomas Römer

15:5; Jer 48:34), south of the Arnon.15 This expansion created the first territorially developed kingdom in Moab, which stretched both north and south of the Arnon.16 His conquests (with a possible later expansion to the area of Heshbon in the north) decided the location of the border between Moab and Israel in the later phases of the Iron Age – north of Heshbon, near the northern edge of the Dead Sea.17 These observations raise several questions: 1. Who ruled south of the Arnon in the half-century or more between the decline of the late Iron I polity and the rise of Mesha? 2. Who ruled north of the Arnon before the Omrides? 3. Why did the Omrides not establish their forts on the northern rim of the Arnon (similar to Meshaʼs Aroer) – a more strategic location? Dibon alone was not strong enough to prevent this. In other words, was Dibon part of a larger polity that included territory south of the Arnon? 4. Whom did Mesha take the territory south of the Arnon from? It seems to us that several biblical texts may help to answer these questions.

The Biblical Material A persistent biblical tradition speaks about the Israelite conquest of Transjordan from two polities: those of Sihon and of Og, who ruled from the Arnon to Bashan (Num 32:33; Deut 3:8–17; 4:47–49; Josh 12:1–5; 13:9–12).18 Numbers 32:33–38 is part of a chapter that deals with the installation of the tribes of Gad, Reuben, half of Manasseh, and (interestingly) Makir (who does not belong to the “official” lists of the twelve tribes19) east of the Jordan. According to these verses, Moses gave to Gad, Reuben, and half of Manasseh the kingdoms of Sihon, king of the Amorites, and Og, king of Bashan. The

15  N. Na‍ʾaman, “The Campaign of Mesha against Horonaim,” BN 73 (1994), 27–30; for different reading of these lines, see Lemaire, “La dynastie Davidique” (see n. 1); A. F. Rainey, “Following Up on the Ekron and Mesha Inscriptions,” IEJ 50 (2000), 116–117. 16  Na‍ʾaman, “King Mesha” (see n. 11), 83–92. 17  On the affiliation of Heshbon in late monarchic times, see A. L emaire “Heshbôn = Hisbân?,” Eretz-Israel 23 (1992), 64–70. 18  See references to discussions of the conquest traditions in, e.g., Na‍ʾaman, “King Mesha” (see n. 11), 91, n. 27. 19  The idea of a twelve-tribe Israel is considered by most scholars to be a late Priestly or even post-Priestly construction; see, e.g., C. Levin, “Das System der zwölf Stämme Israels,” in Congress Volume: Paris, 1992 (ed. J. A. Emerton; VTSup 61; Leiden: Brill, 1995), 163–178. The appearance of the name of Makir is probably an indication that we have some older material here and not just late speculations from the Persian period.

Early North Israelite “Memories” of Moab

715

Gadites are said to have (re)built Dibon, Ataroth, Aroer, and other towns; the Reubenites built Heshbon and other places. Nothing is said about Manasseh. The same tradition is related in Deut 3:12–18, a passage that mentions the same tribes, Makir included. Here Moses gives to the Reubenites and Gadites the territory north of “Aroer, which is on the edge of the valley of the Arnon, and half the hill country of Gilead with its cities” (v. 12) or “as far as the valley of the Arnon, with the middle of the wadi as a boundary, as far over as the river Jabbok, the boundary of the Ammonites. The Arabah also, with the Jordan as the boundary, from Chinnereth as far as the sea of the Arabah, the Salt Sea, under the slopes of Pisgah on the east” (vv. 16–17).20 In this passage Bashan, kingdom of Og, is given to Manasseh (Sihon is not mentioned here). Finally, Josh 13:9–12 and 15–31 reflect the same tradition of Transjordanian possessions of Gad, Reuben, and half Manasseh:21 “from Aroer, which is on the edge of the valley of the Arnon, and the city that is in the middle of the valley, and all the tableland of Medeba as far as Dibon; and all the cities of Sihon king of the Amorites, who reigned in Heshbon, as far as the boundary of the Ammonites; and Gilead, and the region of the Geshurites and Maacathites, and all Mount Hermon, and all Bashan to Salecah; all the kingdom of Og in Bashan, who reigned in Ashtaroth and in Edrei” (vv. 9–12).22 These verses anticipate a more detailed description in Josh 13:15–32 and may be an insert into a late text.23 The tradition about towns that the Gadites and Reubenites are supposed to have built in Moabite territories from Heshbon to Ataroth occurs only in Num 32:34–38.24 According to Horst Seebass, these verses belong to the original text of Num 32, which he attributes to a Yahwist from the eighth century BCE,25 who claimed the territories in vv. 33–38* as genuinely Israelite. Even if one disagrees with the presence of a “Yahwist” in the book of Numbers, the list could indeed reflect a northern tradition, which is perhaps more of a “wishful thinking” in 20  According to E. Otto, Deuteronomium 1–11 (HThKAT; Freiburg: Herder, 2012), 480, vv. 14–17 are a later addition to Deut 3:12–13, 18ff. 21  For a comparison between Num 32 and Jos 13, see also Gass, Die Moabiter (see n. 1), 172–173. 22  See also the synopsis in R. A chenbach , Die Vollendung der Tora: Studien zur Redaktionsgeschichte des Numeribuches im Kontext von Hexateuch und Pentateuch (Wiesbaden: Harrassowitz, 2003), 371–372, and for Josh 12–13 and Num 32, see E. A. Knauf, Josua (ZBK 6; Zürich: Theologischer Verlag, 2008), 130–133. 23  Knauf, Josua (see n. 22), 126–135. 24  For identifications of the place names, see M. W üst , Untersuchungen zu den siedlungsgeographischen Texten des Alten Testaments (Wiesbaden: Reichert, 1975), 147; J. A. D earman , “Historical Reconstruction and the Mesha Inscription,” in Studies in the Mesha Inscription and Moab (ed. J. A. Dearman; ABS 2; Atlanta: Scholars Press, 1989), 155–210, here 170–185. 25  H. S eebass , Numeri (BKAT 4; Neukirchen-Vluyn: Neukirchener Verlag, 2004), 346– 347. See also L. Schmidt, Das vierte Buch Moses: Numeri 10,11–36,13 (ATD 7/2; Göttingen: Vandenhoeck & Ruprecht, 2004), 195, who attributes the oldest layer to the “Elohist.”

716

Israel Finkelstein and Thomas Römer

late monarchic or even later times (below) than territorial reality.26 It should be noted that for the six identifiable and excavated places in Num 32:34–38 (Heshbon, Medeba, Ataroth, Jahaz, Dibon, and Aroer) the picture is quite clear: they depict strong Iron II presence, no (or very weak) Persian period, and strong late Hellenistic occupation; the surveys portray a similar picture.27 The parallel statements about the possessions given to the Reubenites and Gadites (and half Manasseh) in Num 32:33; Deut 3:12–18*; and Josh 13:9–12* are clearly interrelated. It could be that Num 32:33 and Deut 3:12–18*, building on the legendary (Dtr) tradition about Sihon and Og,28 belong to literary layers that extend into the book of Joshua and could therefore be qualified as a “hexateuchal redaction.”29 The focus on Transjordanian possessions in the Persian period could thus be explained either as a memory from the Iron II or by the fact that parts of this territory belonged at a certain time to the subprovince of Samaria;30 it may also reflect an old claim from the time of the kingdom of Israel. In a certain way, this tradition contradicts the Dtr (?) description of the Israelitesʼ passage through Moabite and Ammonite territory in Deut 2:9–23*.31 This passage insists that the Israelites are not allowed to conquer Moabite territory,

 According to Wüst, Untersuchungen (see n. 24), 180–183, the list reflects a territorial constellation before the Omrides. But there is not much evidence for that; S chmidt , Das vierte Buch Moses (see n. 25), 201, thinks that the territories correspond to Solomonʼs twelfth district. But the historicity of the biblical presentation of the Solomonic empire should be given up. In fact, this list is not too different from the prophecies on Moab in Isa 15:1–16:1 and Jer 48:1–47; hence one may suppose that it represents knowledge of the late monarchic period telescoped into the past. 27  For Heshbon, Dibon, and Aroer, see entries in The New Encyclopedia of Archaeological Excavations in the Holy Land (ed. E. Stern; 4 vols.; Jerusalem: Israel Exploration Society & Carta; New York: Simon & Schuster, 1993) and bibliography there; for Jahaz and Ataroth, see n. 2, above; for Medeba, see, e.g., T. P. H arrison et al., “Investigating 5,000 Years of Urban History: The Tall Madaba Archaeological Project,” in Crossing Jordan: North American Contributions to the Archaeology of Jordan (ed. T. E. Levy et al.; London: Equinox, 2007), 143–152; for the survey, see the evaluation in I. F inkelstein , “From Sherds to History: Review Article,” Israel Exploration Journal 48 (1998), 120–131, here 121–123. 28  R. Rendtorff, “Sihon, Og und das ‘israelitische Credo,’ ” in Meilenstein: Festgabe für Herbert Donner (ed. M. Weippert and S. Timm; ÄAT 30; Wiesbaden: Harrassowitz, 1995), 198–203. 29  As pointed out by Otto, Deuteronomium 1–11 (see n. 20), 481, and Knauf, Josua (see n. 22), 135, it is impossible to reconstruct precisely the reworking of these traditions and the late Fortschreibungen and attempts to harmonize. These texts are, as Knauf puts it, “eine schriftgelehrte Komposition” (135). 30  U. Hübner, Die Ammoniter: Untersuchungen zur Geschichte, Kultur und Religion eines transjordanischen Volkes im 1. Jahrtausend v.Chr. (ADPV 16; Wiesbaden: Harrassowitz, 1992), 213–215. 31  Different late glosses about the primitive inhabitants have been inserted into this passage; see, e.g., Otto, Deuteronomium 1–11 (see n. 20), 479–482. 26

Early North Israelite “Memories” of Moab

717

since YHWH has given these lands to Lot and his sons (2:9, 19).32 Apparently, the Deuteronomists considered the Jordan to be a “theological” border (see the crossing of the Jordan in Josh 3–4); it seems that they had no interest, after 587 BCE, in claiming Moabite or Ammonite territory.33 But later revisions in the context of the Hexateuch did. However, there was no moment in history when Israel dominated such a territory as described in the above-mentioned texts: in the two expansionist periods of the Northern Kingdom, the Omrides did not rule north of the Yarmuk, and Jeroboam II did not dominate the mishor of Moab. The Transjordanian emphasis in the above texts can therefore be seen either as a utopian wish or, better, as a late claim based on what one could describe as “accumulative memory” in the north, commemorating both the rule of the Omrides in the mishor and the conquest of Jeroboam II in Bashan (Amos 6:13). This observation takes us to Num 21.34 Scholars saw the song of Heshbon, or both the song and the prose, as belonging to the older narrative layers (J/E) in the Pentateuch.35 The relevant part starts with the defeat of Sihon at the hands of the invading Israelites at Jahaz and a statement that “Israel settled in all the cities of the Amorites, in Heshbon, and in all its villages” (v. 25). Then comes the Song of Heshbon, which commemorates a military victory in the mishor. Scholars have debated the age of the song (giving dates ranging from the premonarchic to 32  Apparently this text presupposes the construction of the Abraham cycle, according to which Lot is either Abrahamʼs brother or his nephew. 33  The origin of this Dtr idea may be located in the geopolitical situation of the seventh century BCE; the only old claim for these territories would be Israelite and date before 720, because this area was taken from Israel by Rezin of Damascus, followed by the conquest of Tiglath-pileser III. 34  This is a highly complex chapter, including in the relationship between the prose and the song; see, e.g., J. Van S eters , “The Conquest of Sihonʼs Kingdom: A Literary Examination,” JBL 91 (1972), 182–197, according to whom Deut 2:26–37 is the basis for Num 21:21–(24)25 and hence the latter is secondary (and late) and not historical; in contrast, see J. R. B artlett , “The Conquest of Sihonʼs Kingdom: A Literary Re-Examination,” JBL 97 (1978), 347–351, who argues that the account in Numbers is the original source; see also E. W. Davies, Numbers (NCBC; Grand Rapids, MI: Eerdmans, 1995); Dearman, “Location of Jahaz” (see n. 12), 122–126; B. A. L evine , Numbers 21–36: A New Translation with Introduction and Commentary (AB 4A; New York: Doubleday, 2000), 110; E. Blum, Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 127–129. A middle position is adopted by E. Otto, Das Deuteronomium im Pentateuch und Hexateuch: Studien zur Literaturgeschichte von Pentateuch und Hexateuch im Lichte des Deuteronomiumsrahmen (FAT 30; Tübingen: Mohr Siebeck, 2000), 129–136, who claims that Num 21 stems from the hexateuchal redaction, which, however, used an older, pre-Dtr account. 35  E.g., M. Noth, Numbers: A Commentary (London: SCM Press, 1968), 161–162; Davies, Numbers (see n. 34), 225–226; C. F revel , “Understanding the Pentateuch by Structuring the Desert: Numbers 21 as Compositional Joint” in The Land of Israel in Bible, History, and Theology: Studies in Honour of Ed Noort (ed. J. van Ruiten and J. C. de Vos; VTSup 124; Leiden: Brill, 2009), 111–135.

718

Israel Finkelstein and Thomas Römer

the postexilic periods)36 and whether it relates to an Israelite victory over Hesh­ bon37 or a Heshbonite victory over Moab (emphasized in the tension between vv. 23–26, which deal with a victory over Sihon, and vv. 27–30, which recall a victory of Sihon over Moab).38 We would side with the former, arguing that both the prose regarding Israelite settlement in the region (and possibly also the battle at Jahaz) and the song, describing a north-south campaign, commemorate the conquest of the mishor by the Omrides in the early ninth century BCE; the biblical record and the Mesha Stela describe the same historical situation. It is quite plausible that in the early days this “memory” was transmitted orally; the Song of Heshbon may have been put in writing in the first half of the eighth century,39 possibly in conjunction with an Israelite claim to this territory in the days of Jeroboam II. Had this been the case, it would be possible that this song circulated in the north and was integrated into the book of Numbers by a “pentateuchal” or a Persian-period redactor (as claimed by Ludwig Schmidt and others40). Though in late monarchic times Heshbon41 was part of Moab (Isa 15:4; 16:8–9; Jer 48:2, 34),42 in the biblical tradition Sihon is described as king of the Amorites not of the Moabites (see also Num 21:13). This too may preserve a genuine memory of the pre-Omride situation in this territory: until the expansion 36  See the presentation of the discussion in H.-C. S chmitt , “Das Hesbonlied Num. 21,27aβb–30 und die Geschichte der Stadt Hesbon,” in Theologie in Prophetie und Pentateuch: Gesammelte Schriften (ed. H.-C. Schmitt; BZAW 310; Berlin: de Gruyter, 2001), 131–154, and S. Timm, Moab zwischen den Mächten: Studien zu historischen Denkmälern und Texten (ÄAT 17; Wiesbaden: Harrassowitz, 1989), 75–89. For the text-critical problems, see Timm, Moab, 65–75. 37  E.g., E. M eyer , “Kritik der Berichte über die Eroberung Palaestinas,” ZAW 1 (1881), 117–146; M. W eippert , “The Israelite ‘Conquest’ and the Evidence from Transjordan,” in Symposia Celebrating the Seventy-Fifth Anniversary of the American Schools of Oriental Research (1900–1975) (ed. F. M. Cross; Cambridge, MA: American Schools of Oriental Research, 1979), 15–34; D avies , Numbers (see n. 34), 231; N a ‍ʾaman , “King Mesha” (see n. 11), 90–91; Levine, Numbers 21–36 (see n. 34), 132. 38  Van Seters, “Conquest” (see n. 34), 195–196; R. P. Knierim and G. W. Coats, Numbers (FOTL 4; Grand Rapids, MI: Eerdmans, 2005), 243; Schmidt, Das vierte Buch Moses (see n. 25), 113–4; according to Schmidt, it is only Dtr that make Heshbon the capital of Sihon. He argues that the song reflects the destruction of Heshbon around 539 BCE. 39  On the beginning of significant writing in Israel only at that time, see I. Finkelstein and B. Sass, “The West Semitic Alphabetic Inscriptions, Late Bronze II to Iron IIA: Archeological Context, Distribution and Chronology,” HBAI 2 (2013), 149–220. 40  See Schmidt, Das vierte Buch Moses (see n. 25), 114–116; T imm , Moab (see n. 36), 75–89; Achenbach, Die Vollendung der Tora (see n. 22), 365. 41  Unfortunately, the archaeology of Tell Hisban is not clear enough to produce minute observations regarding biblical traditions. For discussion, see also Timm, Moab (see n. 36), 79–90, who summarizes the results of the excavations. 42  For resolving its appearance in the oracle against Ammon in Jer 49:3, see, e.g., Lemaire, “Heshbôn” (see n. 17), 65.

Early North Israelite “Memories” of Moab

719

of Mesha the mishor was not considered part of Moab but rather was ruled by a late-Canaanite entity. Verse 26, which “paves the way” for the Song of Heshbon,43 is peculiar: “For Heshbon was the city of Sihon the king of the Amorites, who had fought against the former (Heb., ha-rishon = the first)44 king of Moab and taken all his land [from the Jabbok]45 to the Arnon.” This may have originated from an Omride polemic aimed at showing that the mishor had not belonged to Moab – that is, to legitimate its conquest46 – or better, from a polemic related to later Israelite territorial claims,47 perhaps in the first half of the eighth century.48 The latter possibility is strengthened by v. 25: only Heshbon and its villages (literally “its daughters”) are mentioned here, not the entire territory of the mishor. In his inscription Mesha does not speak about the conquest of Heshbon, and hence this town could have been conquered by Moab later, closer in time to the high period of Israel in the first half of the eighth century.49 Judges 11:25–26 repeats the account in Numbers and is probably dependent on it.50 Judges 11:1–11a and 32–33 is an old north Israelite story,51 depicting Jephthah as a leader of an Apiru group that operated in the no manʼs land on the eastern margin of the Gilead; this reality could not have existed anymore with 43  Noth, Numbers (see n. 35), 163; see also Knierim and Coats, Numbers (see n. 38), 243; Seebass, Numeri (see n. 25), 357. 44  This expression may have been added in order to distinguish this Moabite king from Balak, who is mentioned in the following Balaam story (S chmidt , Das vierte Buch Moses [see n. 25], 115). 45  J. M. Miller, “Early Monarchy in Moab?,” in Early Edom and Moab: The Beginning of the Iron Age in Southern Jordan (ed. P. Bienkowski; Sheffield Archaeological Monographs 7; Sheffield: J. R. Collis National Museums and Galleries Merseyside, 1992), 77–91, here 84; idem , “Israelite Journey” (see n. 6), 578; N a ‍ʾaman , “King Mesha” (see n. 11), 91. There is a text-critical problem here: MT has “from his hand” (‫)מּיָ ֹדו‬, ִ LXX has ἀπὸ Αροηρ. Possibly the original text had ‫ ִמּיַ ּבֹק‬as in v. 24, as suggested by the editor of BHS. 46  Na‍ʾaman, “King Mesha” (see n. 11), 90, identifies the “former king of Moab” with the house of Daudoh from Horonaim. 47  Davies, Numbers (see n. 34), 227. 48  See also Gass, Die Moabiter (see n. 1), 198–199, who also gives good arguments for why the tradition behind Num 21 is archaeologically and textually only possible in the ninth-eighth centuries BCE. 49  Another possibility is to see v. 26 as a gloss, entered by a confused late author who no longer understood the earlier situation in the mishor, before the expansion of Moab in the days of Mesha (v. 29b may also be regarded in the same light; see Levine, Numbers 21–36 [see n. 34], 133). 50  B artlett , “Conquest” (see n. 34), 357–351; M iller , “Israelite Journey” (see n. 6), 577–595; W. Gross, Richter (HThKAT; Freiburg: Herder, 2009), 593; for the opposite theory – that the account in Numbers depends on that in Judges – see Van Seters, “Conquest” (see n. 34); idem, “Once Again: The Conquest of Sihonʼs Kingdom,” JBL 99 (1980), 117–119. 51  For Judg 11:1–11, see G ross , Richter (see n. 50), 555–556; for 11:32–33, T. R ömer , “Why Would the Deuteronomists Tell about the Sacrifice of Jephthahʼs Daughter?,” JSOT 77 (1998), 27–38, 28–30.

720

Israel Finkelstein and Thomas Römer

the dense settlement system and strong central government of the Iron IIB in the eighth century BCE. Judges 11:11b–31 and 34–40 is probably a postexilic insertion.52 It seems, then, that the two additions to the Num 21 story in Judg 11:25–26 – that Balak was the contemporary of Sihon, who ruled south of the Arnon, and that the Israelites occupied the mishor for three hundred years – do not come from an old memory but are an attempt of the late redactor of Judges to combine different pentateuchal traditions. Balak was taken from Num 22, and the three hundred years should be seen as an attempt to harmonize the old account in Numbers with the later conquest traditions. Part of the Balaam narrative, too, is traditionally seen as belonging to the old pentateuchal material.53 However, it has often been observed that the Balaam tradition (which, as shown by the Deir Alla Inscription, is indeed an old one) existed independent of other pentateuchal traditions before it was integrated into the book of Numbers. According to Reinhard Achenbach and Rainer Albertz, the Balaam cycle was inserted by a hexateuchal redactor54 who wanted to strengthen Israelʼs claim to possession of the land. The older tradition probably comes from the north55 and is an account about a famous seer who took the side of Israel against Moab. If one follows Bickertʼs reconstruction of the older account, “Israel” did not come out of Egypt (Moses does not appear in this story) but rather settled in parts of Transjordan, in territories that were also claimed by Moab.56 Numbers 22:36* (which according to Bickert and others belongs to the Grunderzählung57) states that Balak son of Zippor, king of Moab, went from his town to meet Balaam “at the city of Moab (Heb., ʿir Moav), on the boundary formed by the Arnon” (the end of v. 36b is probably an explanatory gloss 58). Clearly, Balak reigns from a place located south of the Arnon. It is reasonable to assume that Ar/Ir Moab is the same as Kir Moab (“the City of Moab,” similar in meaning to “Rabbah of the Ammonites”) and should be identified with Khirbet 52  R ömer , “Why,” (see n. 51), 27–38; P. G uillaume , Waiting for Josiah: The Judges (JSOTSup 385; Sheffield: T&T Clark, 2004), 238–240. 53  S. M owinckel , “Der Ursprung der Bilāmsage,” ZAW 48 (1930), 233–271; N oth , Numbers (see n. 35), 171; Davies, Numbers (see n. 34); A. Rofé, Sefer Balaʿam (Jerusalem: Simor, 1979) (Hebrew), and recently, J. M. Robker, “The Balaam Narrative in the Pentateuch/ Hexateuch/Enneateuch,” in Torah and the Book of Numbers (ed. C. Frevel et al.; Tübingen: Mohr Siebeck, 2013), 334–366. 54  A chenbach , Die Vollendung der Tora (see n. 22), 389–394; R. A lbertz , “Das Buch Numeri jenseits der Quellentheorie: Eine Redaktionsgeschichte von Num 20–24,” ZAW 132 (2011), 171–183, 336–347. 55  R. Bickert, “Israel im Lande Moab: Die Stellung der Bileamerzählung Num 22–24 in ihrem redaktionellen Kontext,” ZAW 121 (2009), 189–210. 56  B ickert , “Israel” (see n. 55), 204–205. He imagines that the original narrative was written down under Ahab or Jeroboam II. 57  The beginning of the verse was added by the redactor who integrated the story about Balaam and his she-ass. 58  Noth, Numbers (see n. 35), 180; Schmidt, Das vierte Buch Moses (see n. 25), 118.

Early North Israelite “Memories” of Moab

721

Balua, close to the Arnon.59 Balakʼs seat can then be located somewhat further to the south, at Kerak, possibly the biblical Horonaim,60 which was taken over by Mesha. Kerak produced a ninth-century BCE royal Moabite inscription, probably written by Meshaʼs scribes (after his annexation of Horonaim?), and other finds, all of which indicate the importance of this place at that time.61 The essence of the story is that Balak “saw all that Israel had done to the Amorites,” that is, to Sihon; hence, rather than relating to the passing of the Israelites on the way from the south to Canaan, it probably refers to the historical conquest of the mishor by the Omrides, who came from the north.62 There are several clues for the antiquity of this story. First, Balak sends a delegation to a prophet from Pethor (Pitru of the Assyrian sources63) “which is near the river” – that is, to Aram64 – probably in conjunction with the Moab-Damascus alliance in the days of Mesha and Hazael. Second, as already mentioned, Balaam is referred to in the eighth-century BCE Deir Alla Inscription. Third, the postexilic text in Deut 23:5–6 presupposes the story and introduces a negative attitude toward Balaam.65 Still, the broader Balaam tale is another example of “accumulative memory” – a story that incorporates realities from different periods: at least two of the high places visited by Balaam in order to bless Israel are located in the northwestern sector of the mishor, overlooking the border between Moab and Israel in late monarchic times.66  See discussion in Miller, “Israelite Journey” (see n. 6), 593–595.  N a ‍ʾaman , “King Mesha” (see n. 11); J. M. S prinkle , “2 Kings 3: History or Historical Fiction?,” BBR 9 (1999), 247–270, here 267; surveys of the possibilities and other views in I. Worschech and E. A. K nauf , “Dimon und Horonaim,” BN 31 (1986), 70–95; J. A. D earman , “The Moabite Sites of Horonaim and Luhith,” PEQ 122 (1990), 41–46; C. Ben-David, “The ‘Ascent of Luhith’ and the ‘Road to Horonaim’: New Evidence for Their Identification,” PEQ 133 (2001), 136–144. 61  E.g., Gass, Die Moabiter (see n. 1), 66–69, 263–264. 62  Levine, Numbers 21–36 (see n. 34), 232. 63  Schmidt, Das vierte Buch Moses (see n. 25), 124–125, however, considers this information to be a kind of midrashic gloss, based on the root p-t-r (Gen 40:5ff. and 41:1ff.), which means “to interpret (dreams).” 64  On the reading Aram, not Edom, see N oth , Numbers (see n. 35), 173–174; D avies , Numbers (see n. 34), 245. 65  Rofé, Sefer Balaʿam (see n. 53), 47; see Noth, Numbers (see n. 35), 172, who thinks that both Balak and Balaam echo historical figures. This attitude was also introduced in Num 22–24, especially in the story of the she-ass. 66  In the Ehud and Eglon story in Judg 3, Moab reaches the area of the northern tip of the Dead Sea; see Na‍ʾaman, “Royal Inscription” (see n. 1), 168; E. Gass, “Zur Ehud-Tradition in historisch-topographischer Hinsicht,” ZDPV 124 (2008), 38–50. Alternatively, this tale may be folkloristic and have nothing to do with a “historical” Moab; see U. Hübner, “Mord auf dem Abort? Überlegungen zu Humor, Gewaltdarstellung und Realienkunde in Ri 3, 12–30,” BN 40 (1987), 130–140; J. A. Soggin, “Ehud und Eglon: Bemerkungen zu Richter III 11b–31,” VT 39 (1989), 95–100; E. A. Knauf, “Eglon and Ophra: Two Toponymic Notes on the Book of Judges,” JSOT 51 (1991), 25–44; Miller, “Early Monarchy” (see n. 45), 85. 59

60

722

Israel Finkelstein and Thomas Römer

Another story of conflict between Israel and Moab appears in 2 Kgs 3:4–27 – a “prophetic story”67 about the campaign of Joram, king of Israel, against an unnamed king of Moab. The message of this story is of course theological,68 and at least part of it – including the references to Elisha and Jehoshaphat – is anachronistic.69 But its nucleus (especially vv. 24–25) is probably a chronistic source about a siege of Kir Moab (mocked here as Kir Hareseth, the city of sherds), that is, the capital of Moab, probably Khirbet Balua south of the Arnon,70 by a northern Israelite king of the Omride dynasty. Evidence for the antiquity of this part is strengthened by the report of a human sacrifice by the king of Moab, which is not criticized by the redactors and is presented as provoking the failure of the assault of his enemies.71 Thus these verses seem to preserve a memory of an aborted Israelite attempt – possibly by Joram72 – to invade the heartland of Moab south of the Arnon in order to incorporate it into the territory ruled by the Northern Kingdom north of the Arnon.

The Israel-Moab Conflict in the Ninth Century BCE Let us try to tie all these data – archaeological, biblical, and extrabiblical – into one coherent historical reconstruction.  And hence containing little historical information; see M. C ogan and H. Tadmor , II Kings: A New Translation with Introduction and Commentary (AB 11; New York: Doubleday, 1988), 49. For the term, see M. Noth, Geschichte Israels (7th ed.; Göttingen: Vandenhoeck & Ruprecht, 1969), 216. For a critique of this passage as “prophetic story,” see E. Würthwein, Die Bücher der Könige (ATD 11; Göttingen: Vandenhoeck & Ruprecht, 1977), 285, who shows that the combination with the Elisha story is not original. 68  B. O. Long, “2 Kings III and Genres of Prophetic Narrative,” VT 23 (1973), 337–348; P. D. S tern , “Of Kings and Moabites: History and Theology in 2 Kings 3 and the Mesha Inscription,” HUCA 64 (1993), 2–14; Na‍ʾaman, “Royal Inscription” (see n. 1). 69  For the different parts, see E. Gass “Topographical Considerations and Redaction Criticism in 2 Kings 3,” JBL 128 (2009), 65–84; according to Würthwein, Die Bücher der Könige (see n. 67), the oldest story can be recovered in vv. 4–9a, 20–27*. 70  Attempts to identify this place north of the Arnon (e.g., B. C. Jones, “In Search of Kir Hareseth: A Case Study in Site Identification,” JSOT 52 [1991], 3–24) go against the logic of the text and should be dismissed. 71  Pace Würthwein, Die Bücher der Könige (see n. 67), 285. The “great wrath” that came over the Israelites in v. 27 was probably the wrath of Kemosh in the old story (which was censored later). For more details, see T. R ömer , “Le sacrifice humain en Juda et Israël au premier millénaire avant notre ère,” ARG 1 (1999), 16–26. 72  This is the theory of most commentators (see W ürthwein , Die Bücher der Könige [see n. 67], 284); Bernhardt postulates a campaign waged by Joash around 800 BCE; see K. H. B ernhardt , “Der Feldzug der drei Könige,” in Schalom: Studien zu Glaube und Geschichte Israels: Festschrift A. Jepsen (ed. K. H. Bernhardt; Stuttgart: Calwer, 1971), 11–22. There is, however, no indication of any Israelite attempt in Moab proper after 841 BCE. 67

Early North Israelite “Memories” of Moab

723

An early polity emerged in southern Moab in the late Iron I (late eleventh and much of the tenth century BCE), a period of growing prosperity in copper production at Wadi Faynan,73 and declined in the second half of the tenth century. Judging from the geographical disposition of the fortified sites in this territory, and from the Balua Stela, its seat was probably located at Khirbet Balua. Elsewhere, one of us suggested that the decline of this formation was the result of the monopolization of copper production and transportation by Egyptʼs Twenty-Second Dynasty, following the campaign of Sheshonq I to Canaan.74 Moab probably recovered after a while, possibly as a sovereign or ally of Damascus; its hub was still located south of the Arnon. The mishor, at least north of Dibon, was not part of this territory; it was ruled by a late-Canaanite territorial formation centered at Heshbon. This is in line with the long-term perspective – what we know archaeologically about the relatively dense settlement system and elaborate architecture in this area in both the Late Bronze and the Iron I.75 Sometime in the first half of the ninth century, the Omrides took over the mishor from this polity – referred to in the Bible as “Amorite,” that is, Canaanite – and established their forts facing a Moabite kingdom that was centered south of the Arnon, probably at Ir Moab/Khirbet Balua, but that included the land of Dibon north of it. The Bible preserves a strong memory of this short phase of north Israelite history in Moab. First, the references to Jahaz as the place of the battle with Sihon, king of Heshbon (Num 21:23; Deut 2:32), and the reference to the same not-too-important place in the Mesha Stela as a fort of the Omrides cannot be coincidental. Second, there is consistent reference to the fact that Israel ruled over the towns of the mishor. An attempt by Israel to press further to the south may be referred to in the annalistic description of a siege over Kir – the capital of Moab south of the Arnon – in 2 Kgs 3:24–25. In this sense, the Balak mentioned in Num 22–24 and the king who was besieged at Kir Moab, though not necessarily the same individual, may represent the same period: they were contemporaries of the Omrides; note that in the texts one seems to have ruled at Horonaim/Kerak and the other at Ir (Kir) Moab/Khirbet Balua. Some biblical texts may preserve a memory that supports the information 73  L evy et al., “High-Precision Radiocarbon Dating” (see n. 10), 16460–16465; I. F inkelstein and E. P iasetzky , “14C and the History of Copper Production at Khirbet en-Nahas,” Tel Aviv 35 (2008), 82–95. 74  F inkelstein and L ipschits , “Genesis of Moab” (see n. 2), 139–152; I. F inkelstein , “The Southern Steppe of the Levant ca. 1050–750 BCE: A Framework for a Territorial History,” PEQ 146 (2014), 89–104. 75  For Tell el-Umeiri in the Late Bronze and Iron I, see D. R. C lark , “The Late Bronze and Early Iron Ages at Tall al-ʿUmayri,” in The Madaba Plains Project: Forty Years of Archaeological Research into Jordanʼs Past (ed. D. R. Clark et al.; London: Equinox, 2011), 43–57; for the settlement pattern and territorial disposition in the Late Bronze Age, see I. Finkelstein, “Settlement Patterns and Territorial Polity in the Transjordanian Highlands in the Late Bronze Age,” Ugarit-Forschungen 45 (2014), 143–159.

724

Israel Finkelstein and Thomas Römer

in Meshaʼs description, namely, that the two places were seats of rival families who struggled for dominance over Moab (see below). Moab took advantage of the weakening of the Northern Kingdom under pressure from Hazael, specifically Israelʼs defeat in the battle of Ramoth-gilead in 841 BCE. One may assume that there was rivalry between Mesha, who probably ruled at Ir Moab/Khirbet Balua, and the house of Daudoh at Horonaim.76 With the backing of his Damascene patron, Mesha took over Horonaim (and erected a stela there – the Kerak Inscription) and then expanded from the land of Dibon to the mishor, taking over the Israelite forts of Jahaz and Ataroth as well as Israelite Madaba and Nebo. This was the first time that Moab ruled over the territories both south and north of the Arnon.77 The border between Moab and Israel was established near the northern edge of the Dead Sea, and it remained there until the fall of the Northern Kingdom. This Moab – on both sides of the Arnon – is the one known to the late monarchic author and referred to so accurately in the prophecies of Isaiah and Jeremiah (Isa 15; Jer 48:1–6, 21–24, 30–37). Mesha imposed his deity, Chemosh,78 over this entire territory and because of the expansion to the north may have moved his capital to Dibon.

Ninth-Century North Israelite Memories in Numbers and Kings The existence of ninth-century “memories” in Num 21–22 and 2 Kgs 3, and their incorporation into much later texts, should come as no surprise. We have recently suggested that the earliest layer in the Jacob narrative – which deals with the delineation of the border between Israelites and Arameans in the Gilead – reflects pre-eighth-century Gileadite traditions.79 And outside of the Pentateuch, the same can be said for, e.g., the memory of Shiloh as an early Israelite shrine, which must come from the late eleventh century (archaeology shows that the site had been destroyed then and did not recover in the Iron II) but which was incorporated into Judahite polemics centuries later.80 76  N a ‍ʾaman , “King Mesha” (see n. 11); idem , “Campaign of Mesha” (see n. 15); for a different interpretation, see Lemaire, “La dynastie Davidique” (see n. 1); Rainey, “Following Up” (see n. 15). 77  Na‍ʾaman, “King Mesha” (see n. 11). 78  The Mesha Stela claims that he destroyed a sanctuary of YHWH and perhaps also another one, if dwd refers to a deity as it does in the original text of Amos 8:14; see T. R ömer , Lʼinvention de Dieu (Paris: Éditions de Seuil, 2014), 152–153; English translation, The Invention of God (trans. R. Geuss, Cambridge, MA: Harvard University Press, 2015), 114–115. 79  I. F inkelstein and T. Römer, “Comments on the Historical Background of the Jacob Narrative in Genesis,” ZAW 126/3 (2014), 317–338. 80  Especially in the so-called ark narrative in 1 Sam 4–6 and the oracles about the destruction of the Jerusalem temple in the book of Jeremiah (Jer 7 and 26), where Shiloh appears as the first sanctuary chosen by YHWH.

Early North Israelite “Memories” of Moab

725

The book of Numbers, although the latest in the Pentateuch, 81 therefore preserves shreds of Israelite traditions regarding the conquest of the mishor by a late-Canaanite king who ruled from Heshbon, the later takeover of the area by the Omrides, as well as “memories” about the existence of an early to mid-ninth-century BCE Moabite kingdom further south, which included what we would call (to use a term from the Mesha Stela) the land of Dibon, land of Kir, and land of Horonaim.82 These traditions must have first been transmitted in the Northern Kingdom83 orally (they may have originated from the temple of YHWH at Nebo, referred to in the Mesha Stela) and were probably put in writing in the first half of the eighth century – in Penuel, Bethel, or Samaria. This is when complex writing appears for the first time in Israel – at Kuntillet Ajrud and Tell Deir Alla.84 The compilation/memorization of these traditions may have been carried out during the reign of Jeroboam II, whose days were characterized by reorganization of the cult in Israel and promotion of the two foundation myths of the kingdom.85 Promotion of the memory of Israelite presence in northern Moab could have served northern territorial ambitions in the first half of the eighth century BCE; such territorial ambitions were fulfilled in the area of Ramoth-gilead (the conquest of Lidebir, Amos 6:13), but for reasons unknown to us were not accomplished in Moab. The now-written early northern Israelite traditions regarding Moab came to be known to Judah in the decades after 720 BCE and still later were incorporated into Judahite/Yehudite works and given a southern orientation. This scenario agrees with the idea that a large number of Israelites settled in Judah in the decades after 720 BCE.86 Our colleague and friend Nadav Na‍ʾaman has recently

 For the idea that Numbers was created as a late “bridge” between a priestly “Triteuch” and the book of Deuterononomy, and as the last occasion to integrate different (non-Dtr and non-Priestly) traditions, see T. Römer, “Israelʼs Sojourn in the Wilderness and the Construction of the Book of Numbers,” in Reflection and Refraction: Studies in Biblical Historiography in Honour of A. Graeme Auld (VTSup 113; ed. R. Rezetko; Leiden: Brill, 2007), 419–445; Achenbach, Die Vollendung der Tora (see n. 22). 82  This is a typical territorial disposition in the semiarid areas of marginal agriculture and pastoral activity in southern Transjordan. 83  For a northern origin of the conquest accounts in Numbers, see also D. E. Fleming, The Legacy of Israel in Judahʼs Bible: History, Politics, and the Reinscribing of Tradition (New York: Cambridge University Press, 2012), 114–132. 84  Finkelstein and Sass, “West Semitic Alphabetic Inscriptions” (see n. 39), 149–220. 85  I. Finkelstein, The Forgotten Kingdom: The Archaeology and History of Northern Israel (Ancient Near East Monographs 5; Atlanta: Society of Biblical Literature, 2013), 141–151; Finkelstein and Römer, “Comments” (see n. 79). 86  M. Broshi, “The Expansion of Jerusalem in the Reigns of Hezekiah and Manasseh,” IEJ 24 (1974), 21–26; W. M. Schniedewind, How the Bible Became a Book: The Textualization of Ancient Israel (Cambridge: Cambridge University Press, 2004); I. F inkelstein and N. A. Silberman, “Temple and Dynasty: Hezekiah, the Remaking of Judah and the Rise of 81

726

Israel Finkelstein and Thomas Römer

opposed this theory;87 yet this is the only way to explain the dramatic settlement and demographic transformation of Judah in a short period of time in the late eighth century and early seventh century BCE.88 Several features of Israelite material culture suddenly appear then in Judah. Transmission of Israelite texts to the south should be seen as a no-less-important characteristic of this process. It is also possible that the integration of some of these memories in the late book of Numbers and in passages belonging to pentateuchal or hexateuchal redaction was favored in the middle of the Persian period (or even later 89), when the Pentateuch was promulgated. In this context the emphasis on northern traditions could tentatively be understood as “Samaritan input,” focusing on territories that the Samaritans may have considered to belong to their province. Recent European research rightly points out that the compilation of the Torah cannot be understood as an exclusively Judahite phenomenon. The northern/ Samaritan contribution to it was probably much stronger than texts such as Ezra 7 or Neh 8 want us to believe. There is no doubt that the northern conquest traditions in Numbers are older, but contrary to the Jacob and other northern traditions, their integration can be understood as a concession to the Samaritans or as a claim by the Samaritans in postexilic times. But where were these traditions kept? Unfortunately, we do not have much information about the political and religious situation in the former Northern Kingdom between 722 BCE and the Persian period. Note however that 2 Kgs 17, although written from a southern perspective,90 admits an ongoing (syncretistic) cult of YHWH91 and therefore the the Pan-Israelite Ideology,” JSOT 30 (2006), 259–285; K. van der Toorn, Family Religion in Babylonia, Syria and Israel (Leiden: Brill, 1996), 339–372. 87  N. Na‍ʾaman, “Dismissing the Myth of a Flood of Israelite Refugees in the Late Eight Century BCE,” ZAW 126 (2014), 1–14. 88  I. F inkelstein , “Migration of Israelites into Judah after 720 BCE: An Answer and an Update,” ZAW 127 (2015), 188–206. 89  The context of the Hasmoneans could constitute a perfect historical setting for some of those texts, both ideologically (they expanded to the areas of Jazer and Madaba) and archaeologically (see above on occupation of the six sites mentioned in Num 32). This observation again raises the question of the terminus ante quem for the latest revision of the Torah. The late second century is probably too late for inserts of important textual passages into scrolls of the Pentateuch. The case could be different in the book of Joshua, for which Knauf, Josua (see n. 22), has detected a redactional layer from the Hasmonean period. 90  The diachronic analysis of 2 Kgs 17 reveals an old layer from the seventh century BCE and two or three revisions from Babylonian and Persian times; see B. B ecking , The Fall of Samaria: An Historical and Archaeological Study (SHANE 2; Leiden: Brill, 1992), and J.-D. M acchi , “Les controverses théologiques dans le judaïsme de lʼépoque postexilique: Lʼexemple de 2 Rois 17, 24–41,” Transeuphratène 5 (1992), 85–93. 91  2 Kgs 17:32: “They also worshiped YHWH and appointed from among themselves all sorts of people as priests of the high places, who sacrificed for them in the shrines of the high places.” See also Römer, Lʼinvention (see n. 78), 234–238; idem, Invention (see n. 78), 176–179.

Early North Israelite “Memories” of Moab

727

existence of one or more Yahwistic sanctuaries (2 Kgs 17:28 mentions Bethel, which may be more ideological than historical, and 17:32 a temple [?] and high places).92 Although most of the northern traditions certainly came to the south after 722 BCE, it seems logical that scrolls and traditions continued to be read and kept in the north, too. All this shows that the Bible – the Pentateuch and the Former Prophets – contains traces of early traditions that go back at least to the early days of the Northern Kingdom in the early ninth century BCE. Other materials may contain even earlier memories,93 for instance in the contexts of the exodus narrative,94 the early Jacob traditions,95 traditions about the sanctuary of Shiloh, and positive memories of the house of Saul.96 Most of these early layers come from the Northern Kingdom. They should probably be understood as local traditions, which – at least in the early days – were memorized in regional shrines. We would refer, for instance, to Gileadite (early Jacob) and Ephraimite (later Jacob, still in the Iron Age) texts from Penuel and Bethel, respectively; Balaam, possibly at Succoth; and Omride traditions about Moab at Nebo. In the late eighth and early seventh centuries BCE – and also later – these early Israelite traditions found their way into Judah. There they acquired later layers of Judahite/Yehudite/ Judean (including “pan Israelite”) ideologies. They were integrated into scrolls that became the Torah and the Prophets and are now embedded in what can be described as “accumulative memories” (or “traces of memory,” according to Jan Assmann97), which – as read today – include layers that represent many centuries of transmission – oral and later written – and redaction.

 The MT has the strange expression ‫ּב ֵבית ַה ָּבֹמות‬, ְ which may reflect a dogmatic correction for “in a temple and in high places.” 93  The fact that old traditions can show up in very late contexts can be demonstrated by the case of Manetho, who in the third century BCE refers to events from more than a thousand years before his time, like the Hyksos or Akhenaten; see G. P. Verbrugge and J. M. Wickersham, Berossos and Manetho Introduced and Translated: Native Traditions in Ancient Mesopotamia and Egypt (Ann Arbor: University of Michigan Press, 2000). 94  See, e.g., D. B. Redford, “An Egyptological Perspective on the Exodus Narrative,” in Egypt, Israel, Sinai: Archaeological and Historical Relationships in the Biblical Period (ed. A. F. Rainey; Tel Aviv: Tel Aviv University Press, 1987), 137–161; R. Hendel, “The Exodus in Biblical Memory,” JBL 120 (2001), 601–608; N. Na‍ʾaman, “The Exodus Story: Between Historical Memory and Historiographical Composition,” JANER 11 (2011), 39–69. 95  Finkelstein and Römer, “Comments” (see n. 79). 96  Finkelstein, Forgotten Kingdom (see n. 85). 97  See, for instance, J. Assmann, Moses der Ägypter: Entzifferung einer Gedächntnisspur (Darmstadt: Wissenschaftliche Buchgesellschaft, 1998). 92

The Historical Geography of the Pentateuch and Archaeological Perspectives Thomas B. Dozeman I would like to explore the relationship between historical geography and archaeological perspectives through the study of the genre of geography in the Pentateuch. Thus, the central question of my paper is, What is the genre of pentateuchal geography? The study of the genre of pentateuchal geography has not been the central topic of literary critics. The historical-critical study of the Pentateuch in the modern period has tended to focus on time, chronology, and events rather than on the spatial and geographical dimension of the literature. The consequence, according to D. Newman, is that “the concrete manifestations of places and locations have not been subject to the level of critical enquiry that has been characteristic of the events that were supposed to have occurred within these same places.”1 The result is that most debates about geography in the Pentateuch focus on historical problems, such as the specific location of a city or a region, and not on its unique literary form or function, in which geographical realism and religious worldview are blended. How do we read pentateuchal geography that is realistic in its representation but often not literal? My reflections on this question will separate into three parts: (1) I will begin with a common dichotomy (or impasse) that presently exists in the interpretation of pentateuchal geography; (2) I will shift from the field of pentateuchal studies to the adjacent discipline of geography and religion to provide a broader perspective on the interpretation of pentateuchal geography; and (3) I will briefly sketch out questions and promising avenues for future research.

The Present State of Study: The Four Rivers of Paradise The four rivers of paradise appear within the second story of creation in Gen 2, where the emergent world is envisioned as the garden of Eden. Genesis 2:10–14 1  D. N ewman , “Metaphysical and Concrete Landscapes: The Geopiety of Homeland Socialization in the ‘Land of Israel,’ ” in Land and Community: Geography in Jewish Studies (ed. H. Brodsky; Studies and Texts in Jewish History and Culture: The Joseph and Rebecca Meyerhoff Center for Jewish Studies, University of Maryland 3; College Park: University Of Maryland, 1998), 153–182, here 157.

730

Thomas B. Dozeman

describes the river that irrigates the garden of Eden along with its four branches: the Pishon flows to the land of Havilah, a land of pure gold and precious stones; the Gihon flows around the land of Cush; the Tigris flows east of Assyria; and the fourth, the Euphrates, receives no further description. The four rivers progress from fantastic locations (the Pishon) to a river so well-known to the author that it requires no description (the Euphrates). The mixing of geographical references from the fantastic to the mundane creates a problem for interpreters. How does one interpret and represent pentateuchal geography? The following representations of the four rivers of paradise represent two common solutions that move in opposite directions.

Map 1: MacMillian Bible Atlas.

Map 2: Anchor Yale Bible Dictionary.

Map 1 is a rendition of the four rivers of paradise from The Macmillan Bible Altas.2 The authors have emphasized the realism of the text rather than the fantasy. The interpretation begins with the Tigris and Euphrates in Mesopotamia, since they can be clearly located on the map. The representation is realistic, conforming to the standards of modern mapmaking. The map includes precise renditions of the bodies of water in the region (now known as the Mediterranean Sea, the Red Sea, and the Persian Gulf), along with the exact flow of the rivers over the topography of the land. Once the region has been realistically reproduced, the authors address the problem of locating the Pishon and Gihon, which they concede may represent symbolic geography. Yet they conclude, 2  Y. A haroni et al., The Macmillan Bible Atlas (3rd ed.; New York: Macmillan, 1993), 21, fig. 14.

The Historical Geography of the Pentateuch and Archaeological Perspectives

731

“since Havilah (the land associated with the Pishon river) is one of the regions of Cush (Gen 10:7), it would seem that the two major branches of the Nile (the Blue and the White) are intended.”3 The result is a very realistic rendition of Gen 2:10–14 suggesting the genre of historical geography. The four rivers of paradise provide fresh water for Mesopotamia and North Africa, the ancient centers of civilization. It is noteworthy that there is no place in such a realistic map for a mythological garden, even though the garden of Eden is the main geographical topic of the story. Map 2 is a drawing of the four rivers of paradise by P. S. Alexander from the Anchor Yale Bible Dictionary.4 The map lacks all realistic topography, focusing instead on the fantastic or cosmological features of the text. As a result, Alexander provides an abstract drawing of the textual map without recognizable details. The central focus of the map is the mythological garden of Eden. In representing the four rivers of paradise, Alexander begins with the more obscure rivers, the Pishon and the Gihon, since they are described first in the text. He notes that the rivers are arranged from east to west. Thus, the Tigris and the Euphrates Rivers are located on the west side of the Pishon and the Gihon Rivers, just the opposite from their position in The Macmillan Bible Atlas. The four rivers of paradise represent cognitive mapping, according to Alexander, in which the author does not write geography based on experience but instead seeks to impose order on space. Yet, Alexander is quick to add that the geography in Gen 2:10–14 is not fiction, because the biblical author is seeking to anchor the garden of Eden in real geographical space, where it is located at the headwaters of four great rivers, which flow from north to south.5 Each map clarifies important aspects of the text. But, it is important to note that the two maps are strikingly different in their representations of the four rivers of paradise. They would appear to reflect the contrasting hermeneutical perspectives of symbolic mapping and historical geography. And they produce very distinct reactions in the reader. Yet it is difficult to evaluate which map represents the stronger interpretation, since each underscores a different aspect of the text. The Macmillan Bible Atlas translates the four rivers of paradise by locating the rivers firmly within historical geography. The map begins with the Tigris and Euphrates and only then brings the Pishon and the Gihon Rivers into conformity with them. The article by Alexander in the Anchor Yale Bible Dictionary emphasizes, instead, the role of the text within the world of religious symbolism and cosmology. And, as a consequence, this map avoids all realistic topography, even with regard to the Tigris and Euphrates Rivers. The two maps underscore important aspects of the description of the four rivers of paradise, while also falsifying the text. They “lie” in different ways, to  Aharoni et al., Macmillan Bible Atlas (see n. 2), 21.  P. S. Alexander, “Early Jewish Geography,” AYBD 2:977–988, here 979. 5  Alexander, “Early Jewish Geography” (see n. 4), 978. 3 4

732

Thomas B. Dozeman

use a phrase from I. Cornelius.6 The hyperrealism of The Macmillan Bible Atlas masks the cosmological implications of the four rivers of paradise, whereas the rendition in the Anchor Yale Bible Dictionary suppresses the realism of the reference to the Tigris and Euphrates Rivers. Placing the two maps side by side provides a strong visual illustration of the way in which biblical geography functions at the crossroads between religious or cosmological symbolism and topographical realism. The contrast between symbolisim and realism has given rise to debate among contemporary interpreters about the differences between fact and fiction, myth and history, or most recently minimalist and maximalist interpretations of pentateuchal geography.7 But such a simple dichotomy or impasse between fact and fiction will not capture the dynamic role of geography in the writing of the Pentateuch. The two perspectives must rather be related in a more nuanced geographical interpretation that is able to account for both the realism and the religious or symbolic presuppositions of the text. This, in a nutshell, is the problem of genre in the interpretation of pentateuchal geography. The problem is not limited to the more mythological texts like the four rivers of paradise in Gen 2; it permeates all geographical representations in the Pentateuch, including travelogues, itinerary lists, boundaries, descriptions of frontiers, and so forth. What does a literary critic and an archaeologist do with this material?

Pentateuchal Geography and the Discipline of Geography and Religion The study of the genre of pentateuchal geography is best located within the discipline of geography and religion, not pentateuchal studies. The field of geography and religion investigates the multiple ways in which religious beliefs and practices interact with their environment, forming a cultural landscape. In the book Introduction to Geography and Religion, C. C. Park divides the field into phases, representing two broad historical periods. The first phase encompasses the premodern period. A defining characteristic of the premodern period is that all geographical representations were made to conform to religious presuppositions. Geographers of religion characterize this phase of their discipline as the 6  I. C ornelius , “How Maps ‘Lie’ – Some Remarks on the Ideology of Ancient Near Eastern and ‘Scriptural’ Maps,” JNSL 24 (1998), 217–230; see the discussion of modern map making by M. Monmonier, How to Lie with Maps (2nd ed.; Chicago: University of Chicago Press, 1996). 7  See the summary of the issues in M. B. M oore and B. E. K elle , Biblical History and Israelʼs Past: The Changing Study of the Bible and History (Grand Rapids, MI: Eerdmans, 2011), 33–43.

The Historical Geography of the Pentateuch and Archaeological Perspectives

733

period of “religious geography.” A second phase, “geography of religion,” arose in the modern period, when social-scientific methodologies changed the study of geography to a more rationalistic and historical orientation.8 The contrast between the two phases of Geography and Religion may be illustrated in the following diagram: Religious Geography

Geography of Religion

Period

Premodern

Modern

Worldview

Religious presupposition

Rationalistic presupposition

Geographical Theological, cosmological, and representation symbolic representations of geographical space

Sociological and historical representations of geographical space

Aim

To identify places in the Bible; to produce a historical geography of biblical times

To clarify the working of God in the world; to support legend and mythology

Geographers of religion note the important role of biblical geography in both phases of their discipline. Biblical geography provided the resource for a theological and cosmological representation of the world in the premodern period and also provided the catalyst for a more historically oriented discipline in the modern period.9 The following overview of the role of biblical geography in the distinct phases of geography and religion will underscore the complex character of the genre.

 C. C. Park , Sacred Worlds: An Introduction to Geography and Religion (London: Routledge, 1994), 1–30. See also E. I saac , “Religious Geography and the Geography of Religion,” in Man and the Earth (University of Colorado Studies, Series in Earth Sciences 3; Boulder: University of Colorado Press, 1965), 1–14, here 2; R. W. S tump , “The Geography of Religion – Introduction,” Journal of Cultural Geography 7 (1986), 1–3; M. Büttner, “Religion and Geography,” Numen 21 (1974), 163–196, here 164; L. K ong , “Geography and Religion: Trends and Prospects,” Progress in Human Geography 14 (1990), 355–371, here 356; Y.-F. Tuan , “Geopiety: A Theme in Manʼs Attachment to Nature and to Place,” in Geographies of the Mind: Essays in Historical Geosophy in Honor of John Kirtland Wright (ed. D. Lowenthal and M. J. Bowden; New York: Oxford University Press, 1976), 11–39. J. K. W right (“Terrai Incognitae: The Place of the Imagination in Geography,” Annals of the Association of American Geographers 37 [1947], 1–15) would define religious geography as “geosophy,” meaning the study of geographical knowledge, which presents “subjective conceptions” of the world. He defines geosophy more broadly than religion as “the geographical ideas, both true and false, of all manner of people – not only geographers, but farmers and fishermen, business executives and poets, novelists and painters, Bedouins and Hottentots – and for this reason it necessarily has to do in large degree with subjective conceptions” (12). 9  Isaac, “Religious Geography and Geography of Religion” (see n. 8), 7. 8

734

Thomas B. Dozeman

Religious Geography Geographers of religion agree that most early forms of geography, whether text or map, fall under the rubric of religious geography. E. Isaac notes in his article “Religious Geography and Geography of Religion” that the essential characteristic of religious geography is religion. All geographical representations of the land or of the broader world are made to conform to religious presuppositions.10 The focus of religious geography, according to R. W. Stump, is on religionʼs role in shaping human perceptions of the world and of humanityʼs place within it.11 Theological imagination and religious cosmology play an active role in representing geographical space. M. Büttner clarifies that the aim of religious geography is to “point out the working of God in the world” and to clarify acts of salvation from a geographical perspective.12 The reason for this, according to L. Kong, is that the “linking of geography to cosmology [. . .] was at the heart of early geography.”13 This is true for the geographical representations in the ancient Near East, for the early Greeks, and for biblical geography.14 Y. F. Tuan captures the dynamic and subjective role of religious geography with the term geopiety, which covers a range of emotional bonds between humans, God, and nature.15 Religious geography in the ancient world ranges from the purely cosmological and pictorial to the more nuanced blending of geographical realism and religious presuppositions that is evident in the four rivers of paradise. The Egyptian cosmological maps found in tombs and on sarcophagi provide ancient illustrations of religious geography, reflecting the worldview of the Egyptians. The maps are representations of the Egyptian gods, especially the cosmology  Isaac, “Religious Geography and the Geography of Religion” (see n. 8), 2.  Stump, “Geography of Religion” (see n. 8), 1–3. 12  Büttner, “Religion and Geography” (see n. 8), 164. 13  Kong, “Geography and Religion” (see n. 8), 356. 14  For discussion of religious geography in Egypt, see A. F. S hore , “Egyptian Cartography,” in The History of Cartography, vol. 1 of The History of Cartography: Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean (ed. J. B. Harley and D. Woodward; Chicago: University of Chicago Press, 1987), 117–129; in ancient Greece, see O. A. W. Dilke, Greek and Roman Maps (Baltimore: Johns Hopkins University Press, 1985); I saac , “Religious Geography and Geography of Religion” (see n. 8), 2–3; and J. S. R omm , The Edges of the Earth in Ancient Thought: Geography, Exploration, and Fiction (Princeton: Princeton University Press, 1992); in the ancient Near East, see A. R. Millard, “Cartography in the Ancient Near East,” in The History of Cartography, vol. 1 of The History of Cartography: Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean (ed. J. B. Harley and D. Woodward; Chicago: The University of Chicago Press, 1987), 113–114; W. H orowitz , Mesopotamian Cosmic Geography (Mesopotamian Civilizations; Winona Lake, IN: Eisenbrauns, 1998); and R. N orth , A History of Biblical Map Making (Beihefte zum Tübinger Atlas des Vorderen Orients 32; Wiesbaden: Dr. Ludwig Reichert Verlag, 1979). 15  Tuan, “Geopiety” (see n. 8), 11–39. 10 11

The Historical Geography of the Pentateuch and Archaeological Perspectives

735

Map 3: The Egyptian cosmological map of the creation of the world from The Book of the Dead of Nesitanesbtashru. Photographed by the British Museum; original artist unknown; What Life Was Like on the Banks of the Nile, edited by Denise Dersin, public domain, https://commons.wikimedia.org/w/index.php?curid=10630686.

associated with the Ennead. I have included one of the many examples from A. F. Shoresʼs contribution on Egyptian cartography in the first volume of The History of Cartography (map 3).16 This cosmological map is pictorial. It shows Shu separating the sky goddess from the earth god. The feet of Shu are on the earth, while his arms support the sky. Geb and Nut represent the boundary of the created world. The Babylonian map of the world (map 4) provides a contrast to the Egyptian cosmological map in the way in which it interweaves geometrical design and cosmic Ocean.17 The map is more abstract than the Egyptian cosmological map, yet it remains firmly anchored in the world of religious geography. The map presents an overview of the surface of the earth, dominated by two concentric circles. The inner circle marks the border of Babylon. Geometric shapes indicate significant topographical locations, including mountains, swamps, channels, and  Shore, “Egyptian Cartography” (see n. 14), 120.  The Babylonian map of the world is contained in the British Museum (ANE 92687). For discussion, see, A. R. Millard, “Cartography in the Ancient Near East,” in Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean, vol. 1 of The History of Cartography (ed. J. B. Harley and D. Woodward; Chicago: University of Chicago Press, 1987), 113–114. 16

17

736

Thomas B. Dozeman

Map 4: Babylonian map of the world. Public domain, https://commons.wikimedia.org/w/index.php?curid=1748616.

cities (i.e., Babylon, Urartu, Susa). The outer ring indicates the border of Ocean, and the triangles that protrude from Ocean represent unexplored lands. The map represents the religious worldview of the Babylonians. It is accompanied by commentary on the role in creation of Marduk and of heroic humans, such as Utnapishtim, the hero of the flood, and the legendary kings Sargon and Nur-Dagan. W. Horowitz notes that the purpose of the map may have been to locate and describe distant regions. Much of the commentary describes the strange animals that are assumed to populate the lands across the sea.18 The writing of the Greek philosopher Anaximander of Miletus (611–546 BCE) provides insight into the Greek tradition of geography, which, although more descriptive of topography, remains in the tradition of religious geography. Anaximander is credited with representing land and sea (Ocean) in a description of the known world. He influenced later authors, such as Hecataeus of Miletus, whose Periodos Ges (Journey around the World), written around 500 BCE, may have included a circular map with Ocean surrounding a land mass. The map indicates the significant role of Ocean as a manifestation of the religious principle of boundless space (apeiron). O. A. W. Dilke provides the following conjectural reconstruction of Hecataeusʼs map of the inhabited world (map 5), which also 18

 Horowitz, Mesopotamian Cosmic Geography (see n. 14), 20–42.

The Historical Geography of the Pentateuch and Archaeological Perspectives

737

Map 5: Hecataeusʼs map of the inhabited world surrounded by Ocean. By Bibi Saint-Pol, own work, based on a GIF by Marco Prins and Jona Lendering, http://www.livius.org/a/1/maps/hecataeus_map.gif, public domain, https://commons.wikimedia.org/w/index.php?curid=1455973.

blends geographical realism with religious presuppositions in a manner similar to the four rivers of paradise. The representations of the Egyptian cosmology, the Babylonian map of the world, and the Ocean in the Greek tradition are best categorized as religious geography, aimed at supporting traditional legend and mythology. It is against this background that we must also interpret pentateuchal geography.

Geography of Religion The field of geography and religion underwent a transformation in perspective and in methodology in the modern era, prompting geographers to divide their discipline between religious geography and geography of religion.19 The term geography of religion first appears in the writing of G. H. Kasche in the late

19

 Büttner, “Religion and Geography” (see n. 8), 165; Park, Sacred Worlds (see n. 8), 18.

738

Thomas B. Dozeman

eighteenth century.20 Central to the change in perspective was the separation of geography from theology, along with the reevaluation of religion as a human institution. The result was that the geography of religion is “concerned less with religion per se than with its social, cultural, and environmental associations and effects.”21 It seeks to explore the impact of religion on physical, cultural, and political landscapes, or the reverse, the influence of the environment on the formation of religion.22 The transition from religious geography to geography of religion signals the rise of modern historical geography. E. Isaac has argued that biblical geography was a catalyst in the reorientation of the field of geography and religion in the late nineteenth and early twentieth centuries. The realism of biblical geography provided a resource for restraining the role of symbolic geography, which dominated the medieval mappimundi tradition. The emphasis on realism over symbolism in biblical geography resulted in the “secularizing of the landscape” in the study of the Bible, giving rise to the study of “historical geography of biblical times.”23 The goal of the new enterprise was “to identify places and names in the Bible and to determine their locations” (a practice that was already established in more ancient work, such as that of Eusebius, perhaps as a resource for pilgrims).24 Geographers of religion note a variety of influences for the turn to realism in biblical geography. (1) The emergence of ecclesiastical geography shifted the focus of spatial representation from symbolism to realism as the church sought to map the spread of the Christian religion, often relating historical geography and religious colonialism.25 (2) The rise of the humanistic tradition, with its rationalistic presuppositions, brought advances in Semitics, philology, and exegetical method in the study of Bible. The goal of the new historical-critical methodology was to reclaim the original text, which would provide reliable historical and geopolitical information about past tradition. The result was a growing interest in geopolitical and geographical research within biblical commentaries and the development of a special field of biblical geography inaugurated by J. D. Michaelis.26 (3) The prominent role of nature in nineteenth-century social-scientific methodology further fueled the study of historical geography as a means for un20  G. H. Kasche, Ideen über religiöse Geographie (Lübeck, 1795). See also M. Büttner, “Kasche and Kant on the Physicotheological Approach to the Geography of Religion,” National Geographical Journal of India 33 (1987), 218–228. 21  Stump, “Geography of Religion” (see n. 8), 1. 22  For discussion see Park, Sacred Worlds (see n. 8). 23  Isaac, “Religious Geography and Geography of Religion” (see n. 8), 8–9. 24  Kong, “Geography and Religion” (see n. 8), 357. 25  Isaac, “Religious Geography and Geography of Religion” (see n. 8), 7–10. 26  For discussion see North, History of Biblical Map Making (see n. 14), 115–19.

The Historical Geography of the Pentateuch and Archaeological Perspectives

739

derstanding the environmental forces shaping biblical religion. It was assumed that a realistic interpretation of biblical geography would provide insight into the natural environment that shaped Israelite religion. The emphasis on etiological legends in the research of A. Alt and M. Noth is a direct outgrowth of the new prominence given to the natural environment.27 Central to their hermeneutical perspective was the assumption that biblical legends “could only have grown out of specific localities, since they are an attempt to give meaning to natural phenomena found there.”28 (4) The turn to realism received still further momentum with increasing travel to Palestine and, in its wake, the emergence of archaeology, in which biblical literature often provided the template for environmental, historical, and geographical research.29 The turn to historical geography and archaeology is already evident in the work of C. Ritter on the geography of Sinai, Palestine, and Syria from 1848 through 1855 (published in 1866) and in E. Robinsonʼs summary of travel in Palestine. These works reflect the growing interest of biblical scholars in the environment of Palestine as a resource for interpreting biblical literature. By the early twentieth century, the environment and physical geography of Palestine were firmly established as an important tool for interpreting pentateuchal geography. International archaeological institutes were formed to support the new research focus, including the French École Biblique in 1890, the German Protestant Institute in 1900, the American School of Oriental Research in 1900, and the British School of Archaeology in 1919. The result of the changing perspective in the eighteenth and nineteenth centuries was that historical geography came to dominate the study of pentateuchal geography. And, over time, it also carried religious and hermeneutical significance, since the truth of Scripture was no longer in its symbolic meaning, as was the case in the medieval mappamundi tradition,30 but in the historical reliability of its geographical data. The changing hermeneutical perspective is evident in H. Fischerʼs appraisal of U. Seetzen (1767–1811): “in Palestine, Seetzen was the first to roam the land of the Bible with the Bible in hand and [. . .] he did a lot to prove the enduring truth of Scripture even in its geographical data.”31 The historically oriented hermeneutical perspective continues into the present time in  For discussion see P. Cloke et al., Approaching Human Geography: An Introduction to Contemporary Theoretical Debates (New York: Guilford, 1991). 28  E. Isaac, “Religion, Landscape and Space,” Landscape 9 (1960), 14–18, here 14. 29  E. Robinson, Physical Geography of the Holy Land (London: J. Murray, 1865). 30  For discussion of the medieval mappaemundi tradition see A.-D. den Brincken, “Mappa mundi und chronographia,” Deutsches Archiv für die Erforschung des Mittelalters 43 (1968), 118–186; and D. Woodward, “Medieval Mappaemundi,” in The History of Cartography, vol. 1 of The History of Cartography: Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean (ed. J. B. Harley and D. Woodward; Chicago: University of Chicago Press, 1987), 286–370. 31  For discussion see North, History of Biblical Map Making (see n. 14). 27

740

Thomas B. Dozeman

the language of “minimalist” and “maximalist” in contemporary biblical studies. If I understand the use of these terms correctly, the criterion that separates them is the historical reliability of geographical and geopolitical representations in biblical literature. A broader and more nuanced contrast between religious geography and geography of religion in the study of the genre of pentateuchal geography is excluded in this debate.

Conclusion This brief overview raises a series of questions concerning the genre of pentateuchal geography, which require further research: How might the field of geography and religion provide a broad framework to “explore the relationship between the historical geography of the Pentateuch and archaeological perspectives” – which is the title of this article and of the research section of the conference “The Pentateuch within Biblical Literature: Formation and Interaction” for which it was prepared. Conversely, what problems result from not clearly identifying the distinct methodologies of religious geography and geography of religion? Two tentative conclusions emerge. The first conclusion is from the perspective of literary criticism: the field of geography and religion demands a clear methodological separation between religious geography and geography of religion in all aspects of the study of pentateuchal geography.32 The two orientations cannot be collapsed into a single procedure. This methodological starting point would raise new questions about the title of my article and the title of our section of study in the conference. Is the contrast that we seek to study really the “historical geography of the Pentateuch and archaeology,” or is the central methodological focus of my article as a literary critic more properly described as the “realistic geography of the Pentateuch and its relationship to archaeology (or better, historical geography)”? The terms realistic and historical convey very different meanings, requiring distinct hermeneutical perspectives. The phrase realistic geography remains multivalent, whereas historical geography is not; it seeks a singular meaning, which no longer separates religious geography from geography of religion. In his study Postmodern Geographies, E. Soja cautions that a study limited to

32  For further discussion, see C. N icolet , Space, Geography, and Politics in the Early Roman Empire (Jerome Lectures 19; Ann Arbor: University of Michigan Press, 1991), 2, 3–5, 9, who introduced the important distinction between historical geography and the history of geography for studying ancient history. Historical geography investigates geographical realities, including vegetation, cultivation, roads, land possession, population, settlements, and boundaries. The history of geography explores the awareness of territory and its representation in the literature, uncovering “the mentality and ideology” of the writers.

The Historical Geography of the Pentateuch and Archaeological Perspectives

741

historical geography cannot be “substituted for an explanation of the social (or religious) production of space” that is generated from lived experience.33 The second conclusion is from the perspective of historical geography: the absence of a comprehensive approach to biblical geography, one that identifies the different hermeneutical claims of religious geography and geography of religion, creates methodological confusion in contemporary study. The ground breaking work of Y. Aharoni illustrates the problem: “Historical geography of the Holy Land is a reflection of the mutual relation between God and Israel as understood and interpreted by Israelʼs national faith.”34 Aharoniʼs research goals are within the field of geography of religion; he wishes to describe historical geography. Yet, his remarks are located squarely in the field of religious geography, as spatial and theological commentary on an ancient text that is intended to describe the presence of God on earth, a divinely chosen people, and their holy land. These are the central theological themes of the Pentateuch; but they exceed the boundaries of geography of religion. It is the realistic geography of the Pentateuch that holds the two approaches of religious geography and geography of religion together, even blurring them. And, indeed, both the historical and the theological inquiry of pentateuchal geography are legitimate areas of study. But the two orientations must be distinguished methodologically in order to discern their creative relationship in the construction of pentateuchal geography. If a distinction is not maintained, the biblical authorʼs tactical and theoretical world of realistic geography is simply rendered as the recording of historical places.35 I conclude by returning to the beginning of my article. The separation of religious geography from geography of religion within the discipline of geography and religion provides a broad framework to reexamine the two maps of the four rivers of paradise that we evaluated at the outset. To be honest, my instinct in comparing the two maps is to contrast the symbolic religious geography of Alexanderʼs map in the Anchor Yale Bible Dictionary to the realistic historical geography of The Macmillan Bible Atlas, as though it were part of the modern discipline of geography of religion. But the field of geography and religion clarifies that both maps represent religious geography, contrasting only in their degree of realism. The two maps illustrate the multivalency of pentateuchal geography. The cognitive map of the Anchor Yale Bible Dictionary underscores the way in which biblical authors impose order on space to reflect their religious presuppositions. The more realistic rendition of the four rivers of paradise in The Macmillan Bible Atlas, however, is no less an imposition of order on space  E. W. S oja , Postmodern Geographies: The Reassertion of Space in Critical Social Theory (London: Verso, 1989), 123. 34  Y. A haroni , The Land of the Bible: A Historical Geography (2nd ed.; Philadelphia: Westminster, 1979), ix. 35  For discussion of the “tactical and theoretical world” that emerges in the making of realistic geography, see Soja, Postmodern Geographies (see n. 33), 1 and passim. 33

742

Thomas B. Dozeman

Map 6: Portion of the Madeba Map. By Bernard Gagnon, own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=12687414.

to reflect the religious presuppositions of the authors than the cognitive map of the Anchor Yale Bible Dictionary. The Macmillan Bible Atlas represents a long tradition that goes back at least to the sixth century CE Madeba Map, in which realistic geography is made to conform to religious presuppositions. The Madeba Map in St. Georgeʼs Church and the four rivers of paradise in the Macmillan Bible Atlas represent the same tradition of realistic religious geography. M. Piccirillo concluded that the Madeba Map was intended to be a document of biblical geography, which “had the Onomasticon of Eusebius as its primary source.”36 The map is rich in historical detail. H. Donner notes the details of the mountains east of the Dead Sea, the flow of the Jordan River, and the position of deserts, plains, valleys, and the Nile Delta.37 Yet, in spite of its geographical realism, the Madeba Map remains firmly anchored in the world of religious geography. “The Bible,” writes Donner, “is the main prerequisite for its cartographical representation.”38 Donner clarifies: “The Bible is not so much the source of the details as it is of the general idea” of salvation history. The entire map is oriented to the east, like Christian churches. Jerusalem is central to the 36  M. Piccirillo, “Medeba,” AYBD 4:656–658, here 658, states that the map is “part of the renewed classicism of the Justinian era.” 37  H. Donner, The Mosaic Map of Madaba: An Introductory Guide (Kampen: Kok Pharos, 1992), 18. 38  Donner, Mosaic Map of Madaba (see n. 37), 21.

The Historical Geography of the Pentateuch and Archaeological Perspectives

743

map, and its general details derive from the story of salvation. The influential role of salvation history is particularly evident with the representation of the Nile River, which flows from the east on the map, rather than from the south. The eastward flow of the Nile River is illustrated in the portion of the Madeba Map represented below (map 6). This section of the map represents the Negeb Desert on the left and the Nile Delta on the right. The Nile River flows from the east into the Mediterranean Sea, rather than from the south, a clear misrepresentation of historical geography. The reason for the presentation of the Nile River is not ignorance of topography by the mapmaker, according to Donner, but the conflict between “real geography” and “religious geography.” He explains: “In ancient Christian tradition the Nile was one of the rivers of Paradise, and the Paradise was situated in the east according to Gen. 2: therefore the Nile had to run from the east to the west without any regard for the geographical facts, even though people may have known them.”39 The Madeba Map illustrates the active role of pentateuchal geography in fashioning a symbolic representation of religious geography, one in which “geopiety” takes precedence over historical geography. The same tradition of religious geography continues in The Macmillan Bible Atlas, where the Pishon and Gihon Rivers are also identified with the Nile River. The difference between the two maps is that the Madeba Map made the topography of the Nile River conform to the biblical text by turning the river to flow eastward, whereas The Macmillan Bible Atlas reversed the process and modified the text to conform to the topography of the Nile River, changing the eastward arrangement of the four rivers of paradise from the sequence in the text, Pishon, Gihon, Tigris and Euphrates, to the sequence on the map, Tigris, Euphrates, Pishon, and Gihon. In both cases, whether the mapmaker rearranged the text or the flow of Nile River, the representations are firmly anchored in the discipline of religious geography. The genre of pentateuchal geography requires further study. New insights into the complex genre are emerging. Z. Kallaiʼs work on scribal practice in the use of geography for intellectual purposes in Israelite historiography provides a helpful starting point.40 In “The Wandering-Traditions from Kadesh-Barnea to Canaan: A Study in Biblical Historiography,” Kallai writes: “historiographical and historiosophical considerations exert a profound influence on the formation of Biblical narrative.” The different boundary texts provide an example. Kallai notes the diverse conceptions of the boundaries of the land in the Hebrew Bible, with distinct titles (i.e., promise to the patriarchs, land of Canaan, and land of Israel). He concludes that they are constructed within a scribal tradition of  Donner, Mosaic Map of Madaba (see n. 37), 19, 67.  For a collection of his work, see Z. K allai , Biblical Historiography and Historical Geography: Collection of Studies (BEATAJ 44; Frankfurt am Main: Peter Lang, 1998); see in particular “Territorial Patterns, Biblical Historiography and Scribal Tradition – A Programmatic Survey” (157–164). 39

40

744

Thomas B. Dozeman

historiography to address larger themes like covenant, the relationship of Israel to its neighbors, or the divine right to land. The representation of geography with both realistic and patterned language provides an important starting point for defining the genre of biblical geography.41 In addition, J. Van Setersʼs research on geography and genealogy, as central concerns in the development of antiquarian historiography, also breaks new ground,42 as does the more recent research on borders and frontiers by Nili Wazana,43 the itineraries by Angela Roskop Erisman,44 and the geographical function of the Jordan River by Rachel Havrelock.45 But we need more sustained literary-critical research on the function of realistic geography in the Pentateuch that is informed by the larger field of critical spatial studies.46

41  Z. K allai , “The Wandering-Traditions from Kadesh-Barnea to Canaan: A Study in Biblical Historiography,” JJS (1982), 175–184, here 184. 42  J. Van S eters , In Search of History: Historiography in the Ancient World and the Origins of Biblical History (New Haven, CT: Yale University Press, 1983). 43  N. Wazana, All the Boundaries of the Land: The Promised Land in Biblical Thought in Light of the Ancient Near East (Winona Lake, IN: Eisenbrauns, 2013). 44  A. R. Roskop, The Wilderness Itineraries: Genre, Geography, and the Growth of Torah (History, Archaeology, and Culture of the Levant 3; Winona Lake, IN: Eisenbrauns, 2011). 45  R. Havrelock, River Jordan: The Mythology of a Dividing Line (Chicago: University of Chicago Press, 2011). 46  In addition to the work of S oja , Postmodern Geographies (see n. 33), see also H. L efebvre , The Production of Space (trans. D. N. Nicholson-Smith; Oxford: Blackwell, 1991).

Hezekiah, Moses, and the Nehushtan A Case Study for a Correlation between the History of Religion in the Monarchic Period and the History of the Formation of the Hebrew Bible1 Jan Christian Gertz The historical evaluation and dating of biblical texts is a difficult business. The direct correlation between the history of the formation of the Hebrew Bible and the history of ancient Israel and its religion is hindered by the complex findings of a tradition of literature developed over centuries by authors largely unknown to us. This problem is especially present for those texts that deal with the early history of Israel. It is widely accepted that these texts were written during a later period, but their exact (historical) placement is difficult to achieve. There is, however, some evidence for the correlation between the history of ancient Israel and the history of literature, and it is one small part of such evidence that I discuss here, namely the evidence supporting the thesis that the tradition of Moses and the exodus first originated in the Northern Kingdom and was incorporated into the traditions cherished in Jerusalem only during the later part of the monarchic period. My test case is the story of Hezekiah, Moses, and the Nehushtan, and I will first turn to the story of the “serpent of bronze” in Num 21. This short narrative is one of the most influential texts of the Hebrew Bible. The bronze serpent crafted at the command of Yahweh, its closeness to non-Israelite custom, which was perceived as problematic from early on, and the alleged transgression of the Decalogueʼs prohibition of images have always been thought-provoking for the wise and academically trained.2 We will discover that

1  This article is based on a preliminary presentation in 2008 at the Collège de France, Paris, and in later guest lectures at the Universities of Tübingen and Tel Aviv. 2  For a detailed account of early Jewish and early Christian reception history, see H. M aneschg , Die Erzählung von der ehernen Schlange (Num 21,4–9) in der Auslegung der frühen jüdischen Literatur (Europäische Hochschulschriften 23/157, Frankfurt am Main: Peter Lang, 1981); J. Frey, “ ‘Wie Mose die Schlange in der Wüste erhöht hat . . .’: Zur frühjüdischen Deutung der ehernen Schlange und ihrer christologischen Rezeption in Johannes 3,14f.,” in Schriftauslegung im antiken Judentum und im Urchristentum (ed. M. Hengel and H. Löhr; WUNT 73; Tübingen: Mohr Siebeck, 1994), 153–205.

746

Jan Christian Gertz

this was already true for the biblical authors themselves, which in turn offers us the rare opportunity to seek the historical reality underlying the text.

I. Numbers 21:4–9: A Summary of the Previous Murmuring Stories Numbers 21:4–9 relates how the people of Israel – toward the end of their wanderings in the wilderness – become impatient with the long journey. Weary of the meager food in the wilderness, they complain to God and Moses. In response to this complaint, Yahweh sends poisonous snakes, and their bites kill numerous Israelites. This triggers the remorse of the people, and they confess their sins and ask Moses to intercede on their behalf with Yahweh. As a result of this intercession, Yahweh orders Moses to make a statue of a bronze serpent and place it on a pole, so that everyone who has been bitten may look at it and live. The passage is marked by a certain narrative skill.3 The disobedience of the people is governed by the expression ‫מות‬, “die.” Just as the people complain about the exodus supposedly leading to their death in the wilderness (v. 5a), the result of Yahwehʼs punishment is precisely the death of the people due to snakebite (v. 6b). Corresponding antithetically to this is the help Yahweh provides via the key word ‫חיה‬, “live.” Looking at the snake raised up by Moses for healing will restore to health all those who have suffered a snakebite (v. 8). This is precisely what happens, as the concluding ‫וחי‬, “live” (v. 9), indicates. From the perspective of literary-critical and traditional-historical considerations, the carefully crafted passage is a unity:4 Verse 4a anchors the episode in the itinerary of the journey in the wilderness. More precisely, despite an interruption by Num 21:1–3, the passage is connected to the mention of Mount Hor in the account of the death of Aaron (Num 20:22–29; P) and presupposes the need to go around Edom mentioned just prior (Num 20:14–21; non-P). Edom does not play any role in the narrative, but that is to be expected. The death of Aaron is taken into account to the extent that the extensive murmuring of the Israelites, corresponding to the murmuring in Num  See K. Koenen, “Eherne Schlange und goldenes Kalb,” ZAW 111 (1999), 353–372, here 356–357. 4  The analysis of E. Aurelius, Der Fürbitter Israels: Eine Studie zum Mosebild im Alten Testament (ConBOT 27; Stockholm: Almqvist & Wiksell, 1988), 146–153, is foundational for what follows. For a multilevel history of the text, see at least S. Beyerle, “ ‘Medizin’ – Phänomene im Alten Israel und im antiken Judentum,” in Gesundheit: Humanwissenschaftliche, historische und theologische Aspekte (ed. S. Beyerle et al.; Theologie – Kultur – Hermeneutik 10; Leipzig: EVA, 2008), 45–78, who emphasizes, on the basis of textual-linguistic criteria, a more recent discourse of an original narratio in Num 21:4, 6, 7aαb, 9, as well as H. Seebass, Numeri 2. Teilband 10,11–22,1 (BK 4/2; Neukirchen-Vluyn: Neukirchener Verlag, 2003), 316–317, who sees a later adaptation in Num 21:4a, 5a*, 6aα, 8. 3

Hezekiah, Moses, and the Nehushtan

747

20:5, is directed no longer against Aaron and Moses but against Yahweh and Moses. The following narrative in vv. 4b–9 connects with the itinerary insofar as the impatience of the people “on the way” presupposes everything else. “The way” can only mean “the way to the Red Sea” in v. 4a.5 The first actions are reported in 21:4b and are reminiscent of the structure of Num 11:1–3: the disobedience of the people (21:4b–5) is followed by the punishment of Yahweh (v. 6), which leads to the cry for help (v. 7a), the intercession by Moses (v. 7b), and finally Yahwehʼs coming to their aid (vv. 8–9). Furthermore, Thomas Römer has shown that “Num 21:4–9 and 11:1–3 can be understood as a frame embracing the cycle of rebellion, with the refusal of conquest as its central point.”6 Despite agreement as far as the basic pattern and the structural function are concerned, there are some specifics that distinguish Num 21 from Num 11: Num 11 is a schematic and programmatic text of Deuteronomistic provenance.7 In contrast, Num 21 offers several characteristic narrative traits. First, the issue of murmuring is not only reported but also quoted extensively by stating the addressees. Second, Yahwehʼs response to the intercession and his help are quite forcefully described. The end of the punishment is not simply relief from the disaster but also the provision of an antidote. As such, Num 21 has a narrative interest that moves beyond the pattern in Num 11. Historically, interpreters have always and correctly identified this special interest as the making of the bronze snake.8 More recently, scholars have  See A urelius , Fürbitter (see n. 4), 149; L. S chmidt , Das vierte Buch Mose: Numeri 10,11–36,13 (ATD 7/2; Göttingen: Vandenhoeck & Ruprecht, 2004), 102; C. Berner, “Die eherne Schlange: Zum literarischen Ursprung eines ‘mosaischen’ Artefakts,” ZAW 124 (2012), 341–355, here 344. There is no reason for a literary-critical division in v. 4a, as J. S. Baden, J, E, and the Redaction of the Pentateuch (FAT 68, Tübingen: Mohr Siebeck, 2009), 134, argues. He attributes the beginning of v. 4a to P and the rest (ab*) to E. The connection to Num 20:21, which according to Baden is the immediately preceding verse in E, is, however, not very convincing. 6  T. R ömer , “Egypt Nostalgia in Exodus 14–Numbers 21,” in Torah and the Book of Numbers (ed. C. Frevel et al.; FAT 2/62, Tübingen: Mohr Siebeck, 2013), 66–86, here 69. 7  Num 11 gives no reason for the dissatisfaction of the Israelites, which clearly marks this episode off from the other narratives of murmuring. In the foreground is the schema whose structure corresponds to the Deuteronomistic schema of the book of Judges (see Judg 3:7–11). Because Deut 9:22 lists Taberah and Kibroth-hatta‍ʾavah among the places where the Israelites angered Yahweh, the Deuteronomistic narrative in Num 11 must be older than the late Deuteronomistic extension in Deut 9. On the Deuteronomistic setting of Num 11:1–3, see Aurelius, Fürbitter (see n. 4), 144–146. Pace Baden, J, E, and the Redaction (see n. 5), 174, who attributes the episode to the Yahwist. 8  H. Gressmann, Mose und seine Zeit (FRLANT 18; Göttingen: Vandenhoeck & Ruprecht, 1913), 289–290; M. Noth, Überlieferungsgeschichte des Pentateuch (Stuttgart: Kohlhammer, 1948), 133–134; Aurelius, Fürbitter (see n. 4), 149–151; Koenen, “Eherne Schlange” (see n. 2), 356; Seebass, Numeri (see n. 4), 319. Berner, “Schlange” (see n. 5), 342, has a different 5

748

Jan Christian Gertz

focused on the taxonomy of this object and its classification in the history of religion. According to prevailing opinion, Num 21 provides a “mytho-historical etiology”9 of an apotropaic symbol in the form of a bronze serpent that was, according to 2 Kgs 18:4, destroyed by Hezekiah, king of Judah (ca. 725–697 BCE). Before I turn to this thesis, I will look at the immediate context in order to explain briefly why it is too simplistic to reduce the intention of the narrative to the manufacture of a specific ritual object. In several respects, Num 21:4–9 occupies a prominent position in the larger pentateuchal narrative. The previous chapter contains a murmuring story as well (Num 20:1–1310). This results in Yahweh accusing Moses and Aaron of a lack of trust in him, and they are consequently not allowed to enter the promised land (Num 20:12). The story of the bronze snake forms a counterpoint to this episode. Within the larger narrative frame, Num 21 is the next and also the last possibility for such a counterpoint. Based on a comparable case to Num 20,11 the story of the bronze snake underlines Mosesʼs extraordinary closeness to God and his precise execution of the divine instructions (cf. 21:8 with 21:9).12 Additionally, Num 21:4–9 is the last real wilderness narrative before Israel crosses into farming country.13 The lament about having had to leave Egypt only to die in the desert (21:5a) thus appears to be somewhat out of place. This first impression, however, is mistaken. This is shown by the concretization of the complaint through the triple reproach of “no food” (Num 11:4–6), “no water” view: “The literary results unambiguously argue for the idea that the cultic worship of the image of the serpent articulated in II Kings 18,4b was developed only later on the foundation of the text in Numbers.” The main argument for this thesis is the note “Moses had made” in 2 Kgs 18:4b, which Berner classifies as a literal quote from Num 21:4–9. That this note refers to Num 21:4–9 on the level of the present shape of the Enneateuch is beyond dispute. That the note had to have this reference from the beginning is a methodologically problematic limitation of the context and the body of knowledge of the texts and traditions in the Bible in its final canonical shape. Thus, the common formulation in 2 Kgs 18:4b (‫ )עשה‬certainly does not need any examples. As an allusion to the Mosaic origin (known by the tradents and intended readers) of the artifact destroyed by Hezekiah, it can be understood without this textual reference. According to Berner, 2 Kgs 18:4b derives from a late (post-Chronicles) redactor who retold the story of the bronze snake in Num 21:4–9 on the basis of Exod 32 and Deut 9. 9  R. Schmitt, Magie im Alten Testament (AOAT 313; Münster: Ugarit-Verlag, 2004), 193. 10  On Num 20:1–13, see Römer, “Egypt Nostalgia in Exodus 14–Numbers 21” (see n. 6), 79–81. He takes the episode to be a post-Priestly narrative. 11  The complaint directed at Yahweh and Moses in Num 21:5 unmistakably continues the complaint against Moses and Aaron in Num 20:4–5: “Why have you brought us up out of Egypt to die in the wilderness? For there is no food and no water” (Num 21:5); “Why have you brought the assembly of the Lord into this wilderness for us [. . .] to die here? Why have you brought us up out of Egypt [. . .] and there is no water to drink.” (Num 20:4–5). 12  See Seebass, Numeri (see n. 4), 314. 13  Deut 2:14 marks the end of the wandering in the wilderness with the crossing of the Zered mentioned in the itinerary in Num 21:12.

Hezekiah, Moses, and the Nehushtan

749

(Exod 15:24; Num 20:5), and “putrid bread” (Exod 16:3). This tripartite lament is more than a general expression of “the dissatisfaction of the people with their austere life in the wilderness.”14 It is a summary of the previous murmuring stories, each of which mentions a single plight.15 It is correct, therefore, to join Erik Aurelius in speaking of a schriftgelehrtes Murren (“scribal murmuring”).16 This “scribal murmuring” places the narrative of the bronze snake in a nexus of references with the three crucial episodes of the wilderness wanderings: the announcement of the exodus in Exod 3, the beginning of the period in the wilderness in Exod 14, and the departure after the stay at Sinai in Num 11. The first announcement, i.e., that Moses will lead the people of Israel from Egypt to a land “flowing with milk and honey” (Exod 3:8), conflicts with the peopleʼs reproach that they have been led into the wilderness to die. This fundamental reproach is expressed at the beginning of the wilderness period and again here (Exod 14:11; Num 21:5a).17 Since Yahweh responds first to the complaint that there is no food in the wilderness by sending them bread from heaven (Exod 16:4), the peopleʼs complaint now is that there is no bread or water but only this miserable food (Num 21:5b).18 The complaint that there is nothing to drink 14  M. N oth , Numbers: A Commentary (OTL; London: SCM Press, 1968), 157; German original, Das vierte Buch Mose: Numeri (ATD 7; Göttingen: Vandenhoeck & Ruprecht, 1966), 137. 15  See G. W. Coats, Rebellion in the Wilderness: The Murmuring Motif in the Wilderness Traditions of the Old Testament (Nashville: Abingdon Press, 1968), 120; Aurelius, Fürbitter (see n. 4), 147, and finally Römer, “Egypt Nostalgia in Exodus 14–Numbers 21” (see n. 6), 82, who refers to Deut 34:10–12 as an analogy. 16  Aurelius, Fürbitter (see n. 4), 147. 17  The Samaritan Pentateuch and very good manuscripts of the LXX read ‫ ונתאצוה‬or ἐξήγαγες ἡμᾶς, “What have you done to us in bringing us out of Egypt?” The singular reading, which is possible here because of the defective writing of the consonantal text, is also confirmed by the Peshitta. In this case, the complaint of the people is not directed against God and Moses but only against Moses, as in Exod 14:11. The remote connection to Exod 14:11 ָ and the comparatively mild punishment perhaps actually suggested the singular (‫)למּות ַּב ִּמ ְד ָּבר‬ reading. Perhaps this also explains the choice of the verb ἐξάγω/‫ יצא‬instead of ἀναβιβάζω/ ‫עלה‬. From v. 7 on, a pluralist reading is suggested. In any case, its affinity with Num 20:5 and the other examples of the sarcastic question about the spirit and purpose of the leading out or coming up of the people of Israel from Egypt is undeniable; cf. Exod 14:11–12; 16:3; 17:3; Num 11:18–20; 14:2–4; 16:13. For the motif of “nostalgia for Egypt,” see R ömer , “Egypt Nostalgia in Exodus 14–Numbers 21” (see n. 6), 70–72. 18  The translation of ‫ לחם הקלקל‬is difficult. Generally speaking, the term ‫ֹלקל‬ ֵ ‫ ַה ְקּ‬is thought to derive from ‫קל‬, ‫קלל‬, or ‫ קלה‬II. In this case, one could translate it as “miserable food.” Connecting ‫ קלקל‬with the Akkadian qulqulliānu (HAL II 1034) is also common. In the case of qulqulliānu, it concerns a cassia species, more precisely the cassia alexandrina (formerly: cassia senna) from the subfamily of the caesalpinioideae (like the St. Johnʼs bread) (AHw 927a), which has a laxative effect. But this derivation does not actually take us further, and the connection with the manna-tamarisk tree is questionable, especially because it is quite improbable that the people in the wilderness or the biblical authors would have come up with

750

Jan Christian Gertz

recalls the first case of murmuring, immediately after the beginning of the actual wandering in the wilderness, following the miracle of the Red Sea (see Exod 15:22–24). Structurally, Num 21:4–9 clearly recalls the murmuring of the people immediately after the departure from Sinai in Num 11:1–3, except that the fundamental aspect of the murmuring is now presented not through a general formulation but through a comprehensive listing of their complaints.19 Finally, the complaint is directed against Moses and Yahweh. Because this configuration is presented elsewhere only at the end of the story of the miracle at the sea, it is probably a specific cross-reference that demonstrates the frailness of the assertion in Exod 14:31b: “the people [. . .] put their trust in him and in Moses his servant.”20 Since the “success” of the exodus thus threatens to be called fundamentally into question shortly before the end of the wandering in the wilderness, the situation is saved only through the peopleʼs confession of sin (cf. Judg 10:6–1621). Because the confession of sin has no parallel in the previous instances of murmuring, its significance for the peopleʼs relationship to God is deliberately emphasized.22 such a protorationalist explanation of the miracle of the manna. Worthy of consideration is the proposal of R. Achenbach, Die Vollendung der Tora: Studien zur Redaktionsgeschichte des Numeri­buches im Kontext von Hexateuch und Pentateuch (BZABR 3; Wiesbaden: Harrasso­ witz, 2003), 348, with n. 50. For ‫ קלקל‬he considers a derivation from ‫ קלה‬I, “to roast,” and refers to the Akkadian qalû, “roasted,” and qalītu, “roasted grain” (AHw 895b, 894a), and the Hebrew ‫( קלי‬Lev 23:4, etc.). In connection with the note on the unleavened bread in Josh 5:11, there was a clear connection with the unleavened bread of the exodus. One can perhaps connect deriving the word from ‫קל‬, ‫קלל‬, or ‫ קלה‬II on the one hand and the etymological closeness to ‫ קלי‬on the other in such a way that it concerns a disparaging wordplay with the food eaten on the night of the exodus. In the context, mentioning the exodus and the echo of the questions of its spirit and purpose, this is even probable. In any case, the polemical undertone, which is also underlined by the tautological self-contradiction of this complaint, is clear: “Nothing to eat and drink except for this miserable food.” For a different view, see Berner, “Schlange” (see n. 5), 346–347. He sees in the wording of Num 21:6 a “compressed” reproduction of the complaint about the manna in Num 11:6. 19  See A urelius , Fürbitter (see n. 4), 146–147; A chenbach , Die Vollendung der Tora (see n. 18), 347. 20  See Noth, Das vierte Buch Mose (see n. 14), 138; Aurelius, Fürbitter (see n. 4), 147; E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 124; Römer, “Egypt Nostalgia in Exodus 14–Numbers 21” (see n. 6), 82. 21  On the connection between Judg 10:6–16 and Num 21:4–9, see A urelius , Fürbitter (see n. 4), 148. 22  On the other hand, we should not go so far as to claim that after the disobedience of the people the bronze snake is used for a test of faith, according to which those who believe the signs Yahweh gives through Moses and listen to him will live. There is nothing here to indicate a test of faith, especially since the alternative – some did not look on the snake and died – is not reported. In any case, one may wonder, in accordance with the complaint directed at Moses (and Yahweh), if this also concerns the question of the accreditation of Moses (cf. Exod 4:1–5 and 7:15) – after all, Moses “made” a snake in both cases.

Hezekiah, Moses, and the Nehushtan

751

The narrative of the bronze snake at the end of the wilderness period thus summarizes the fundamental problem of this period. Furthermore, it displays the tension in the figure of Moses as torn between Yahwehʼs gracious attention and deliverance on the one hand and the disobedience of the people on the other. All this is more than a minor point – regardless of the dominance of the snake motif, to which I will now turn my attention.

II. The Destruction of the Nehushtan within the Context of Hezekiahʼs Reform (2 Kgs 18:4) As mentioned above, the passage in Num 21 is often characterized as an etiology for an apotropaic symbol in the form of a snake that was subsequently destroyed as part of the cultic reforms of Hezekiah. The passage in question states: He removed the high places, broke down the pillars, and cut down the sacred pole. He broke in pieces the bronze serpent that Moses had made, for until those days the people of Israel had made offerings to it; it was called Nehushtan. (2 Kgs 18:4)

The verse follows immediately after the opening note on Hezekiahʼs reign and justifies the very positive evaluation of his piety. The triad of cultic items removed by Hezekiah – high places, pillars, and sacred pole – belongs to the classic repertoire of Deuteronomistic texts. Only in relation to the bronze snake does the text seem to deviate: the mention of the snake cannot be attributed to Deuteronomistic formulas. Nadav Na‍ʾaman, however, has shown that this argument is not without problems, as a look into programmatic texts such as Deut 7:5 shows:23 But this is how you must deal with them: break down their altars, smash their pillars, hew down their sacred poles, and burn their idols with fire.

If we keep in mind that altars were the central objects of cultic places, Deut 7:5 and 2 Kgs 18:4 correspond to each other regarding the order of the cultic objects perceived as problematic by the Deuteronomists. In 2 Kgs 18:4, then, the bronze snake occupies the position of the idols listed in Deut 7:5. The snake seems to stand pars pro toto for the idols. This impression is intensified by the mention of offerings (‫ קטר‬in 2 Kgs 18:4) made to the bronze snake. In Deuteronomistic literature the term ‫( קטר‬piel) has a predominantly negative connotation, generally referring to the worship of “foreign gods” (cf. 2 Kgs 17:1124). The assessment of the dealings with the Nehushtan is clearly pejorative. In this respect, the 23  N. N a ‍ʾaman , “The Debated Historicity of Hezekiahʼs Reform,” ZAW 107 (1995), 179–195, here 181–182. 24  Therefore, this example is of particular interest because it concerns the cultic crimes of the inhabitants of the Northern Kingdom (“Israelites”).

752

Jan Christian Gertz

episode fits well with the undoubtedly Deuteronomistic context. Additionally, the wəqatal forms in 2 Kgs 18:4 deviate from the common narrative mode that uses the consecutive imperfect. According to the commonly held view, these wəqatal forms point to the postexilic period.25 Also, the archaeological debate tends to assume that a major cultic reform by Hezekiah is the product of Deuteronomistic historiography.26 If one adds the observations on content, philology, and archaeology it is prima facie likely that the note on the reform, including the mention of the Nehushtan, stems from a late Deuteronomistic editor. Despite these observations, scholars – in my opinion correctly – tend to regard the note on the destruction of the Nehushtan as historically reliable. The argument for such an evaluation stems from Tendenzkritk (profile-based criticism): because of their reverence for the authority of Moses, the Deuteronomists 25  The wəqatal forms in 2 Kgs 18:4 (cf. 2 Kgs 23:4–20.) related to a single event in the past are a much-discussed problem with respect to their classification and meaning in terms of linguistic history. They are often viewed as a postexilic Aramaism. See M. Weippert, “Die Petition eines Erntearbeiters aus Məṣad Ḥăšavyāhū und die Syntax althebräischer erzählender Prosa,” in Die Hebräische Bibel und ihre zweifache Nachgeschichte: Festschrift für Rolf Rendtorff zum 65. Geburtstag (ed. E. Blum; Neukirchen-Vluyn: Neukirchener Verlag, 1990), 449–466, here 453. This explanation, of course, presupposes the adoption of an entire verbal system from another language, which is improbable for that time even for the later writings of the Old Testament. Furthermore, it is hardly to be overlooked, that – in spite of the growing (Imperial) Aramaic influence on Hebrew – wəqatal forms are nothing more than exceptions within the (undisputed) later writings of the Old Testament, whereas wayyiqtol forms remain the common narrative tense. This state of affairs objects quite clearly to the assumption that later redactors would “accidentally” employ Aramaizing wəqatal forms and thus use two different syntactical constructions at the same time (wayyiqtol forms beside wəqatal forms) to express one and the same affair (narrative continuity in the past). This must be considered particularly when dealing with texts in which both forms exist alongside each other, such as the account of the reform in 2 Kgs 23:4–20. Therefore it becomes necessary to ask for the specific function of the wəqatal forms. See W. von Soden, “Gab es bereits im vorexilischen Hebräisch Aramaismen in der Bildung und der Verwendung von Verbalformen,” ZAH 4 (1991), 36–45; K. K och “Gefüge und Herkunft des Berichts über die Kultreformen des Königs Josia,” in Alttestamentlicher Glaube und Biblische Theologie: Festschrift für Horst Dietrich Preuß zum 65. Geburtstag (ed. J. Hausmann and H.-J. Zobel; Stuttgart: Kohlhammer, 1992), 80–92; C. Hardmeier, “König Joschija in der Klimax des DtrG (2Reg 22f.) und das vordtr Dokument einer Kultreform am Residenzort (23,4–15*): Quellenkritik, Vorstufenrekonstruktion und Geschichtstheologie in 2Reg 22f.,” in Erzählte Geschichte: Beiträge zur narrativen Kultur im alten Israel (ed. R. Lux; Biblisch-theologische Studien 40; Neukirchen-Vluyn: Neukirchener Verlag, 2000), 81–145, here 124–125, 138–139; M. Pietsch, Die Kultreform Joisias (FAT 86; Tübingen: Mohr Siebeck, 2013), 223–238. At this point it is not necessary to come back to the discussion about the coexistence of wayyiqtol forms and wəqatal forms in 2 Kgs 23:4–20. Concerning the account of actions in 2 Kgs 18:4, one can say that it simply juxtaposes different reform measures of the king in the form of a list without representing a specific course of events. Being syntactically “compliant to the rules,” this is expressed by the sequence of the perfect forms, which are connected by a waw (perfectum copulativum). 26  N. Na‍ʾaman, “Historicity” (see n. 23), 179–195.

Hezekiah, Moses, and the Nehushtan

753

would hardly have attributed an item to him that later needed to be liquidated. Since the destruction of the high places, pillars, and sacred pole would have been enough to portray Hezekiah as a reformer, an additional profile at the expense of Moses appears unlikely in a Deuteronomistic context. In other words, “If the Deuteronomists themselves trace the obsolete idol back to Moses we undoubtedly have an older [. . .] tradition here.”27 Since the Deuteronomists – in contrast to the Chronicler (2 Chr 31:1) – did not erase the note, one can argue that their criticism refers not to the origin of the idol but to its misuse as an object of veneration.28 The origin of the Deuteronomistic note on the reform in 2 Kgs 18:4 can therefore be reconstructed as follows: the memory of the destruction of the Nehushtan offered a welcome opportunity for the Deuteronomistic historiographers to paint Hezekiah as a precursor of Josiah and to construct him as a foil to his successor Manasseh, the archetypal villain (see 2 Kgs 21:3). In doing so, they attributed the triad of high places, pillars, and sacred pole to a single act and transformed it into a general cultic reform of a respectable extent. Most likely, they aligned the verbal forms of the older note to Deuteronomistic formulas, listing several elements of a cultic reform by King Hezekiah (cf. 2 Kgs 23:4–20). Both older scholarly works and several modern contributions tend to date Num 21:4–9 early, and the text is attributed to an early Yahwist or Elohist. If this is so, the story is fairly simple: toward the end of the eighth century BCE, Hezekiah removed an old cultic object that is now seen as offensive and is traced back to Moses. Unfortunately, things are no longer that simple or certain when it comes to the literary history of the Hebrew Bible. Since Num 21 alludes to the whole presentation of Israelʼs sojourn in the wilderness, a simple chronology based on an early date of Num 21 is impossible. Here it does not matter whether the passage, on the basis of the allusions to the context, belongs to a Deuteronomistic editorial layer of the Pentateuch or, more likely, is even more recent. If the note in 2 Kgs 18:4 is pre-Deuteronomistic, the tradition contained in it must be older than its alleged etiology in Num 21. Taking this into account, let me now investigate what associations the snake held for the author of Num 21:4–9.

III. The Function of the Bronze Serpent according to Numbers 21:4–9 Snakes first appear in v. 6 of Num 21:4–9: as a response to the murmuring of the people, Yahweh sends poisonous snakes (‫)הּנְ ָח ִׁשים ַה ְּׂש ָר ִפים‬. ַ It is likely that  H. S pieckermann , Juda und Assur in der Sargonidenzeit (FRLANT 129; Göttingen: Vandenhoeck & Ruprecht, 1982), 173. 28  See also Berner, “Schlange” (see n. 5), 343, n. 7. 27

754

Jan Christian Gertz

‫ ָׂש ָרף‬is an attribute of ‫ נָ ָחׁש‬here (similarly Deut 8:15) and should be translated as “burning/poisonous.”29 The snakes are introduced with the definite article, which I understand as the authorʼs way of referring to the common knowledge that poisonous snakes live in the desert and especially in the region, which is later called Wadi Nehushtan (Nahal Timna). In addition to Num 21, this fact is known to Deut 8:15; Isa 14:29; Isa 30:6; Herodotus, Hist. II.75–76; III.107–109; and Esarhaddon (fragment from the annals of a campaign against Magan and Meluha, 671 BCE).30 A zoological determination based on these descriptions is, of course, impossible. What is important is that the creatures are ‫ׂש ָרף‬, ָ “burning.”31 The first mention of the bronze snake figure uses this characteristic term ‫ׂש ָרף‬. ָ 32 Right from the beginning its function is stated: the figure serves to heal deadly snakebite. Snake figures are well attested in the eastern Mediterranean throughout antiquity, mostly in the context of preventive measures against snakebite, a plague of snakes, or, less specifically, disaster in general. This use is attested by the various representations of snakes on door handles, vessels, brick capsules beneath door sills, bracelets, kudurrus (border stones), and, last but not least, seals. Apart from the rather numerous seal finds, primarily from the eighth and seventh centuries BCE, with one or two stylized, upright uraei (cobras), the finds for southern 29  See U. R üterswörden , “‫ׂשרף‬,” TDOT 14:218–228, here 223–228; German original, ThWAT 7:882–891, here 887–891, who distinguishes between the winged seraphim presented here (which do not necessarily need to be serpent-like) and the adjectival use here and in Deut 8:15, which certainly refers to snakes but not to winged ones. Because the verb ‫ שרף‬is transitive, the adjective is to be translated “scorching/burning.” 30  For Herodotus and Esarhaddon, see R. Rollinger, “Herodot (II 75f, III 107–109), Asarhaddon, Jesaja und die fliegenden Schlangen Arabiens,” in Ad Fontes! Festschrift für Gerhard Dobesch (ed. H. Heftner and K. Tomaschitz; Vienna: Eigenverlag der Österreichischen Gesellschaft für Archäologie, 2004), 927–946. For the reception of this tradition in Hellenistic literature, see R. Rollinger and M. Lang, “Die fliegenden Schlangen Arabiens: Transfer und Wandlung eines literarischen Motivs in der antiken Überlieferung – Ein Florilegium,” in Eine ganz normale Inschrift – und ähnliches zum Geburtstag von Ekkehard Weber: Festschrift zum 30. April 2005 (ed. F. Beutler and W. Hameter; Althistorisch-Epigraphische Studien 5; Vienna: Eigenverlag der Österreichischen Gesellschaft für Archäologie, 2005), 101–109. In Isa 14:29 it is somewhat different: from the root of a snake (‫ )נחש‬springs a viper (‫ )צפע‬and his/her fruit is a flying seraph (‫)שרף מעופף‬, which can be understood more as a general term. 31  What comes to mind here are the Palestinian saw-scaled viper (echis coloratus), the horned desert viper (cerastes cerastes), the Persian horned viper (pseudocerastes persicus fieldi), the Egyptian desert cobra (cerastes candidus), the Palestinian viper (vipera palaestinae), the mole viper (atractaspis engaddensis), and the Saharan sand viper (cerastes vipera). See P. Riede, Im Netz des Jägers: Studien zur Feindmetaphorik der Individualpsalmen (WMANT 85; Neukirchen-Vluyn: Neukirchener Verlag, 2000), 231–232. 32  Here without the article because it has to do with a still-unknown object. ‫ ָׂש ָרף‬is probably used elliptically. The division of ‫ ַהּנְ ָח ִׁשים ַה ְּׂש ָר ִפים‬in v. 6 into the arrangement in v. 8 (‫)ׂש ָרף‬ ָ and the explanation in v. 9 (‫ )נָ ָחׁש‬is felicitous, whereby the indication of the finished product (‫נְ ַחׁש‬ ‫ )נְ ח ֶֹׁשת‬is aimed at 2 Kgs 18:4 (‫ נְ ַחׁש ַהּנְ ח ֶֹׁשת‬or ‫)נְ ֻח ְׁש ָּתן‬.

Hezekiah, Moses, and the Nehushtan

755

Syria in the Iron Age are somewhat scanty. But there are snake jewelry and decorations on vessels. Snake skulls found in homes in Taʿanakh are possibly connected with this goal of prevention as well.33 If all that failed and one was nevertheless bitten by a snake, snakes were also used for treatment. This is attested especially by the Egyptian incantation texts relating to snakes.34 As far as ancient Israel is concerned, we have some indications of similar forms of ritual medicine and their specialists (cf. Ps 58; Isa 3:3; Qoh 10:11; Sir 12:18; and Avigad and Sass, no. 420).35 In this magic-therapeutic context, the function of the ritual object is best explained as a substitute for the harmful and ominous powers of the snake. This is precisely the reason why Yahwehʼs naming of the figure in Num 21 refers to the harmful power of the snakes – a figure of the completely harmless Asia Minor dwarf racer, for instance, would never be able to heal the bite of a horned viper. Such ideas are evident in the so-called namburbi rituals from Assyria. As part of such a ritual, the harm, announced or already manifested by snakes, would be transferred from the affected person back to the snake. In a similar vein, 1 Sam 6 reports how the Philistines enlist the help of “five golden tumors and five golden mice” to get rid of a plague brought upon them by Yahweh. To understand such practices, it is important to note that, according to the view presented in the entire corpus of ancient Near Eastern apotropaic medicine, they did not have an automatic effect but depended on the will of the gods, given that evil was also brought by the gods. In the area of ritual, only the distinction between the effective powers on the one hand – snake, dog, mouse, etc. – and the corresponding substitutes on the other hand, along with the actual initiator – the personal gods of the people – made it possible to ward off evil.36 Correspondingly, in Num 21 the magical therapeutic reality of the snake figure set up by Yahwehʼs command is linked back to Yahweh himself. Thus, it is clear that the snake figure is not a magical object that is healing in itself. Rather, it is Yahweh who is behind the effect, just as he is also behind the plague of snakes.37 It is striking that Num 21 reports neither an incantation nor a medical action, such as the application of some ointment or the administration of a potion. It would be conceivable that healing by looking at something is viewed as a posi See Schmitt, Magie (see n. 9), 198.  For an overview, see B. S chipper , “Schlangenbeschwörung in Ägypten und im Alten Testament,” in Israel zwischen den Mächten: Festschrift für Stefan Timm zum 65. Geburtstag (ed. F. Hartenstein and M. Pietsch; AOAT 364; Münster: Ugarit-Verlag, 2009), 265–282. 35  See B eyerle , “Medizin” (see n. 4), 71–72; N. Avigad and B. S ass , Corpus of West Semitic Stamp Seals (Jerusalem: Israel Academy of Sciences and Humanities, 1997), 177. 36  See S. M aul , Zukunftsbewältigung: Eine Untersuchung altorientalischen Denkens anhand der babylonisch-assyrischen Löserituale (Namburbi) (BaF 18; Mainz: von Zabern, 1994), 8, 10, 41, and passim. 37  So also Schmitt, Magie (see n. 9), 197, who rightly emphasizes that the Old Testament understanding of magic is not abstracted from its ancient Near Eastern context. 33

34

756

Jan Christian Gertz

tive reversal of the “evil eye” or that as an amplification of the healing referred to elsewhere through mere touch38 it could emphasize the power of Yahweh. But this explanation is not properly convincing because, given the materials for comparison and alleged everyday practice, an incantation would actually be expected. Instead, a simple glance at the apotropaic object appears to be enough. Healing through a gaze is attested for the Greco-Roman world of a later period but not – to my knowledge – for the ancient Near East. The lack of an incantation or a medicinal act was noted quite early and explained theologically. Thus, the Wisdom of Solomon makes a theological virtue out of the necessity of an explanation and states that it was not looking at the snake that was raised up that brought about healing. Rather, the source of healing was the looking to God that was elicited prior to looking at the snake (Wis 16:5–12; v. 7: “For he who turned toward it was saved, not by what he saw but by you, the Savior of all”). Similarly, the targumim are forced to avoid every appearance of the automatic efficacy of the image and to stress the conversion called for by the direction of oneʼs gaze (Tg Ps.-J.: everyone remained alive “who looked at the bronze serpent and directed his heart to the name of the Word Yahweh” and who “raised his face to his Father in heaven”).39 It was recently proposed that Num 21 lacks an incantation because the text is directed against all forms of magic-therapeutic practice.40 To bolster this claim of a decidedly antimagic stance in the Hebrew Bible, reference is made to the law regarding prophecy in Deut 18 and to the designation of Yahweh as physician in Exod 15. Such a view is rather problematic. First of all, the ‫ ְמנַ ֵחׁש‬in Deut 18:4 is – despite the Ugaritic parallels (see mlḫš in KTU 1.100; 1.107) – not a snake charmer but simply a person who performs destructive magic, as is clear from the other use of the term in the Hebrew Bible (see ‫ נַ ַחשׁ‬in Num 23:23) and from the context, which is focused on some form of mantic activity, most likely witchcraft, and not issues of healing. Second, neither Exod 15:26 nor any other text of the Hebrew Bible appears to contain any form of polemic against the medical profession (cf. 1 Kgs 20). Thus, Exod 15:26 is “not a [. . .] clear hint of an ‘environment hostile to medicine’ in ancient Israel – roughly in the sense of focusing the medical professions already widely available elsewhere in the ancient Near East on Yahweh.”41 Rather, the text emphasizes the healing effect of obedience to Yahweh in a specific situation – the exodus from Egypt. This can be seen in the fact that the 38  For necklets and stones, see A. S chuster -B randis , Steine als Schutz- und Heilmittel (AOAT 46; Münster: Ugarit-Verlag, 2008). 39  Note also m. Roš Haš. 3:8, in which Num 21:4–9 is combined with Exod 17:8–13. 40  B. Schipper, “Die ‘eherne Schlange’: Zur Religionsgeschichte und Theologie von Num 21,4–9,” ZAW 121 (2009), 369–387. 41  Beyerle, “Medizin” (see n. 4), 51. See also H. Niehr, “JHWH als Arzt: Herkunft und Geschichte einer alttestamentlichen Gottesprädikation,” BZ 35 (1991), 3–17.

Hezekiah, Moses, and the Nehushtan

757

“diseases [. . .] brought upon the Egyptians” do not describe a view of disease typical for this region but are a narrative link to the plague narratives.42 Also, the oft-cited exhortation in Sir 38:9, “Pray to God for he is the one who heals,” is not evidence for a mood in ancient Israel opposing the medical profession. The quote is to be read in connection with the high praise of the doctor in Sir 38:1–8, which, in reference to Exod 15:26b, endorses the clarification that God nevertheless is the sole subject who performs the healing, and the doctor, despite the recognition people give him, is in the end nothing more than a tool. Sirach 38:9 emphasizes that God is the sole source and cause of healing, but the skill of the doctor may be given a place in this guiding principle within a creation theology.43 The counterexample is King Asa in 2 Chr 16:12. But here as well it is merely going to see the doctor instead of trusting in Yahweh that is problematic (like the military alliance prior to that, which likewise does not explicity prohibit political action). In addition, we also find some magic-therapeutic texts without explicit recitation of incantations in neighboring cultures.44 Especially interesting in this context are Egyptian texts from the realm of personal piety. These texts – in close analogy to Exod 15 – praise Amun-Re as physician and seem to neglect magic-therapeutic knowledge. In the same context, however, documents of medical-ritual contents were found – in some cases on the same piece of papyrus.45 So it becomes clear that the literary intention of the sources has to be evaluated along with their material context. Keeping this in mind, we must take seriously that we are dealing with a literary text when we try to explain the lack of a ritual in Num 21. The following remarks will help us understand the textʼs intention: 1. After Mosesʼs single misdemeanor, Num 21:4–9 emphasizes his extraordinary closeness to Yahweh, likewise stressing that he follows Yahwehʼs command in producing and erecting a snake down to the last detail.

 See N. Lohfink, “ ‘Ich bin Jahwe, dein Arzt’ (Ex 15,26): Gott, Gesellschaft und menschliche Gesundheit in der Theologie einer nachexilischen Pentateuchbearbeitung (Ex 15,25b.26),” in “Ich will euer Gott werden”: Beispiele biblischen Redens von Gott (ed. N. Lohfink; SBS 100; Stuttgart: Verlag Katholisches Bibelwerk 1981), 11–73, here 26. 43  See L. Schrader, “Beruf, Arbeit und Muße als Sinnerfüllung bei Jesus Sirach,” in Der Einzelne und seine Gemeinschaft bei Ben Sira (ed. R. Egger-Wenzel and I. Krammer; BZAW 270; Berlin: de Gruyter, 1998), 117–149, here 140. 44  See N. P. Heessel, “Aus einer neuassyrischen Tafel mit Rezepten gegen Schlangenbisse,” TUAT 5:153–154. 45  Leiden I 350 (ll. 1–2): “Who liberates from doom, who chases away disease, a doctor who heals without medicine. Who opens the eyes and drives out blindness”; Chester-Beatty IV (ll. 179–180): “It is a cure for every eye, a true medicine, that works instantly.” See J. Assmann, Ägyptische Hymnen und Gebete (2nd ed.; OBO; Göttingen: Vandenhoeck & Ruprecht, 1999), 427, 434. 42

758

Jan Christian Gertz

2. The text identifies the period in the wilderness as an epoch during which Yahwehʼs devotion to his people is threatened because of the peopleʼs disobedience. 3. Within its current context, Num 21:4–9 is the reference text for the statement “the bronze serpent that Moses had made” in 2 Kgs 18:4. Thus, we have arrived again at the etiology for the Nehushtan that Hezekiah destroyed. But we can state more precisely that the omission of the institution of a ritual and a note on what happened to the item during a later period transforms Num 21:4–9 into an etiology that decidedly wants to account for nothing but rather explains why something happened that should not have. In other words, in Num 21 an incriminating but obviously difficult-to-erase note regarding an idol of a serpent fashioned by Moses is defused. This is done by drawing a distinction between the original function of the Nehushtan during a specific situation and time, i.e., the end of Israelʼs sojourn in the wilderness, and its misuse as an item of cultic devotion during a later period.

IV. The Identity of the Nehushtan in 2 Kings 18:4 The bronze snake in Num 21 is the result of scribal activity. But the question of the identity of the Nehushtan in 2 Kgs 18:4 remains. Let me offer some brief and rather speculative remarks. As far as Nehushtanʼs history-of-religion birth certificate is concerned, Egyptian snake symbolism of the first millennium BCE has currently supplanted the Canaanite-Phoenician symbol. Othmar Keel, in particular, has stressed the significance of Egyptian bronzes depicting erect cobras on papyrus flowers.46 These bronzes sometimes have brackets for poles, so they can be described as snake standards. The bronzes are only 10–20 centimeters tall, and the snakes are only about 9.5 centimeters.47 These standards were used in processions.48 Their motifs are comparable to the Phoenician-Israelite metal bowl from Nimrud (ca. 750–700 BCE). This bowl depicts two winged uraei on a papyrus stalk flanking a scarab and surrounded by two sphinxes.49 Keel assumes that Hezekiahʼs destruction of the Nehushtan has to be seen as an anti-Egyptian act. Polemicizing against the representation of Yahweh in the form of animals, opposition groups, which had fled from the Northern Kingdom to Jerusalem  O. K eel , Die Geschichte Jerusalems und die Entstehung des Monotheismus (2 vols.; Göttingen: Vandenhoeck & Ruprecht, 2007), 422–429. 47  Keel, Die Geschichte (see n. 46), 422, fig. 302–303; idem, “Das Land der Kanaanäer mit der Seele suchend,” TZ 57 (2001), 245–261. 48  M. Page -G asser , Götter bewohnten Ägypten: Bronzefiguren der Sammlungen “Bibel + Orient” der Universität Freiburg Schweiz (OBO S02; Freiburg: Universitätsverlag, 2001) 133–134. 49  See Keel, Die Geschichte (see n. 46), 427, fig. 300. 46

Hezekiah, Moses, and the Nehushtan

759

after 722 BCE, may have formed the sociopolitical background to this act. If that is so, Hezekiah would have destroyed such an Egyptian cultic item under pressure from prophetic circles and because he was disappointed that he did not receive Egyptian help when besieged by the Assyrians in 701 BCE.50 Keelʼs theory has been widely accepted. A closer look, however, reveals that the theory is more problematic than the iconographic material might suggest. The classification of the Nehushtan as a seraph on a pole rests only on Num 21 in combination with Isa 6, though we can ask if both texts are speaking of the same “thing” when they describe it as ‫ׂש ָרף‬. ָ Second Kings 18:4 neither states that the Nehushtan is a seraph nor says that it was mounted on a banner. Also, if one wants to maintain the Egyptian origin of the Nehushtan, the assumption of iconoclasm by the Jerusalemites disappointed in Egypt reads like a stopgap. After all, it remains ambiguous whether the Nehushtan was part of the old, established inventory of the temple in Jerusalem at all. If we do not follow Keelʼs interpretation, the question of the religiohistorical origin of the Nehushtan remains open. Perhaps it is possible to connect it with the figure of Moses or, better, with the Moses tradition and with the exiles from the Northern Kingdom.51 The reason for doing so may be found in the phrase ‫ּבנֵ י־יִ ְׂש ָר ֵאל‬, ְ who brought offerings to the Nehushtan. In the books of Kings, this term is initially used for the Israelites of the premonarchic period. But unless the term is related to a contemporary group, from 1 Kgs 12 onward it is used only for the inhabitants of the Northern Kingdom. Additionally, Hezekiahʼs anti-Mosaic act would have been difficult to execute had Moses already been part of the salvation-historical inventory of Jerusalem, as is the case three generations later under Josiah. Under Hezekiah, the Moses tradition appears to be a fairly recent addition in Jerusalem and may be connected to refugees from the north. This fits well with the results of studies that locate the origin of the exodus tradition in the north and date its first literary version to the time around the fall of the  See Keel, Die Geschichte (see n. 46), 429.  On the northern origin of the exodus tradition, see Hosea, who is the only eighth-century prophet who mentions it (Hos 11:1; 12:1, 10, 14; 13:4); see also Y. Hoffman, “A North Israelite Typological Myth and a Judean Historical Tradition: The Exodus in Hosea and Amos,” VT 39 (1989), 169–82; R. A lbertz , “Exodus: Liberation History against Charter Myth,” in Religious Identity and the Invention of Tradition (ed. J. W. van Henten and A. Houtepen; STR 3; Assen: Royal Van Gorcum, 2001), 128–143; K. van der Toorn , “The Exodus as Charter Myth,” in Religious Identity and the Invention of Tradition (ed. J. W. van Henten and A. Houtepen; STR 3; Assen: Royal Van Gorcum, 2001), 113–127; J. J. C ollins , “The Development of the Exodus Tradition,” in Religious Identity and the Invention of Tradition (ed. J. W. van Henten and A. Houtepen; STR 3; Assen: Royal Van Gorcum, 2001), 144–155; M. K öckert , “YHWH in the Northern and Southern Kingdom,” in One God – One Cult – One Nation: Archaeological and Biblical Perspectives (ed. R. G. Kratz and H. Spieckermann; BZAW 405; Berlin: de Gruyter, 2010), 357–94; and E. Blum, “Der historische Mose und die Frühgeschichte Israels,” HBAI 1 (2012), 37–63, here 42–46. 50 51

760

Jan Christian Gertz

Northern Kingdom.52 Of course, such a Nehushtan could be of Egyptian origin, even in the north. The connections of the Northern Kingdom to Egypt and the Negev are well documented (cf. Kuntillet ʿAjrud). However, it is equally likely that the Nehushtan has indigenous or Phoenician origins and stands in continuity with small snake figurines that were placed on cult podiums in temples from the Late Bronze Age in Hazor (stratum IB), Megiddo (stratum X), and Timna (thirteenth–twelfth c. BCE). If the theory I have presented is correct, one should connect the Nehushtan and the Moses tradition associated with it to the former inhabitants of the Northern Kingdom. Then we will have uncovered a further indication that the traditions relating to Moses and the exodus originate in the north – and gained thereby a glimpse into the historical reality behind the text.

52

 See J. C. Gertz, “Mose und die Anfänge der jüdischen Religion,” ZTK 99 (2002), 3–20.

For the Border of the Ammonites Was . . . Where? Historical Geography and Biblical Interpretation in Numbers 21 Angela Roskop Erisman Biblical studies has a geography problem. Ever since Wellhausen, we have been inclined to date biblical texts based on the assumption of a straightforward correspondence between the geography in a text and the time in which it was written.1 Numbers 21 frustrates any such effort because the geography is utterly incoherent. Are the Israelites traveling around or through Moab? Why is the northern border of Moab identified in v. 13 as the Arnon but v. 20 has the Israelites camped in Moab north of the Arnon? Why do the Israelites arrive in Moab twice, first in v. 20 and again in v. 1 of chapter 22? Finally, what is a reference to the border of the Ammonites doing here (v. 24), in a text that otherwise does not mention them? The geographical incoherence of Num 21, combined with literary analysis that understands it as an effort to bridge Tetrateuch, Pentateuch, and Enneateuch late in the formation of the literary corpus we now have, has led scholars to ignore geography in favor of ideology and even to attribute the geographical muddle to scribal ignorance of landscape.2 The fact that Num 21 is a big knot makes it an excellent text for working through – and hopefully beyond – this polarized way of treating geography in biblical literature. Thomas Dozeman has led the way by pointing out that in order to unknot this text we must do historical geography and bring in questions  T. B. D ozeman , “Geography and Ideology in the Wilderness Journey from Kadesh through the Transjordan,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 173–189. For one recent example, see I. Finkelstein, “The Historical Reality behind the Genealogical Lists in 1 Chronicles,” JBL 131 (2012), 65–83. 2  See M. Noth, “Num. 21 als Glied der ‘Hexateuch’-Erzählung,” ZAW 58 (1940), 161–189; D. Baly, “The Pitfalls of Biblical Geography in Relation to Jordan,” in Studies in the History and Archaeology of Jordan III (Amman: Department of Antiquities, 1987), 123–124; J. M. Miller, “The Israelite Journey through (around) Moab and Moabite Toponymy,” JBL 108 (1989), 577–595; J. Van Seters, The Life of Moses: The Yahwist as Historian in Exodus– Numbers (Louisville: Westminster John Knox, 1994). The view continues to hold sway in the most recent synthesis of material relating to Ammon in C. W. Tyson, The Ammonites: Elites, Empires, and Sociopolitical Change (1000–500 BCE) (LHBOTS 585; London: Bloomsbury, 2014), 134–135. 1

762

Angela Roskop Erisman

about how ideology influences the representation of geography in literature.3 An integrative approach can help us navigate the Scylla of assuming that biblical literature is merely a vehicle for objectively representing historical and geographical realities and the Charybdis of assuming that “geographical representation [is] no more than a cipher for ideas” and has no relationship to historical and geographical realities.4 The last decade has seen a burgeoning literature that applies critical spatial theory in ways that are transforming how we read biblical texts.5 The specific geographical problems with Num 21, however, can be better addressed by turning to reception theory and relevance theory to understand how geography is used in literature. The simple correspondence model we have long assumed for understanding the relationship between literature and history is a feature of classic historicism, which tends to view literature as a medium for communicating either history or the ideology of a dominant power. New Historicism has since shown that literature does not correspond to history but is embedded in culture in ways that are often very complex.6 Wolfgang Iserʼs theory of reception and its implications for composition can help us understand how this embeddedness comes about. Authors select elements of cultural repertoire and combine them in a text in order to serve a particular set of literary goals, placing constraints on and creating possibilities for interpretation. Cultural repertoire is drawn from a store of knowledge the scribe has, knowledge that is ideally shared by the reader, who taps it in her effort to interpret what she reads.7 Geography is cultural repertoire, drawn from a store of knowledge that entails a wide variety of associations with a place, including ideas and values as well as

 T. B. Dozeman, “Biblical Geography and Critical Spatial Studies,” in Constructions of Space I: Theory, Geography, and Narrative (ed. J. L. Berquist and C. V. Camp; LHBOTS 481; New York: T&T Clark, 2007), 87–108. 4  T. B. Dozeman, “Geography and History in Herodotus and in Ezra-Nehemiah,” JBL 122 (2003), 449–466, here 455. 5  Noteworthy are these volumes: J. L. Berquist and C. V. Camp (eds.), Constructions of Space I: Theory, Geography, and Narrative (LHBOTS 481; London: T&T Clark, 2008); idem (eds.), Constructions of Space II: Theory, Geography, and Narrative (LHBOTS 490; New York: T&T Clark, 2008); J. C. de Vos et al. (eds.), Constructions of Space III: Biblical Spatiality and the Sacred (LHBOTS 540; London: Bloomsbury T&T Clark, 2016); M. K. G eorge (ed.), Constructions of Space IV: Further Developments in Examining Ancient Israelʼs Social Space (LHBOTS 569; London: T&T Clark, 2013); G. T. M. Prinsloo and C. Maier (eds.), Constructions of Space V: Place, Space, and Identity in the Ancient Mediterranean World (LHBOTS 576; London: Bloomsbury T&T Clark, 2013). 6  See A. R. Erisman, “New Historicism, Historical Criticism, and Reading the Pentateuch,” Religion Compass 8/3 (2014), 71–80, DOI: 10.1111/rec312099, and literature cited there, for discussion of New Historicism and its application to pentateuchal studies. 7  W. Iser, The Act of Reading: A Theory of Aesthetic Response (Baltimore: Johns Hopkins University Press, 1978). 3

For the Border of the Ammonites Was . . . Where?

763

information about its location and settlement history.8 Any of these associations can motivate the selection of a place name and its use in combination with other elements of cultural repertoire, sometimes creatively so, to serve a set of literary goals. Iserʼs theory has profound implications for our geography problem: the key to integrating geography and ideology is to recognize that representation of geography in literature is influenced by literary goals, which often involve ideology. So when we encounter a place name in a text, we should frame our questions differently: we should no longer ask, as we often have, With what historical and geographical reality does this place name correspond? but What literary goals does it serve? and What aspect(s) of its cultural context are relevant to those literary goals? Our problem as modern readers is that we do not fully share the scribeʼs store of knowledge – his cultural context, to the extent this knowledge is shared – when it comes to geography, or anything else for that matter. This is not merely because we read at such a remove from the context in which any given biblical text was produced and initially read. No individual ever fully shares context with another; Dan Sperber and Deirdre Wilsonʼs relevance theory, a form of pragmatics, can help us understand why. A particular reader comes to a text with a cognitive environment, or a set of information and assumptions (about how the world works, about the character of a speaker, etc.) that are manifest to her – a pool from which she can construct a context for interpreting the text. Thus two individuals may share a cognitive environment but construct context differently. Relevance theory is built on the notion that an individual will naturally be drawn to those pieces of information and assumptions that are most relevant, that result in a more meaningful and significant interpretation than other available alternatives. When it comes to interpreting ancient texts, we do not even share a cognitive environment with the scribes; what is manifest to us, beyond what is supplied in the immediate discourse context, is gained through study of data available through accident of preservation and discovery. We can learn ancient genres by studying literary remains, and we can learn ancient geography by studying references in literary and administrative texts as well as material remains produced by archaeological excavation and survey.9 What we gain from study of 8  A. R. Roskop, The Wilderness Itineraries: Genre, Geography, and the Growth of Torah (HACL 3; Winona Lake, IN: Eisenbrauns, 2011), 28–31. 9  Relevance theory deals with overt acts of spoken communication, so a second principle is also at play: not only does human cognition tend to maximize relevance (the cognitive principle of relevance), but we can also assume that any overt act of communication “conveys a presumption of its own optimal relevance” (the communicative principle of relevance); see D. W ilson and D. S perber , “Relevance Theory,” in The Handbook of Pragmatics (ed. L. R. Horn and G. Ward; Oxford: Blackwell, 2004), 607–632. For a fuller version, see D. Wilson and D. Sperber, Meaning and Relevance: Communication and Cognition (Cambridge: Cambridge University Press, 2012). We cannot assume the relevance of data learned

764

Angela Roskop Erisman

these remains is not a historical and geographical picture to which a text may correspond but rather access to the scribeʼs cultural repertoire, which gives us the potential to understand the various associations that might have driven his use of it in the text and that might help us comprehend meaning. Our access to ancient cultural repertoire is of course limited, but we do have enough to unknot the literary history and the geography of Num 21 and to articulate some implications for how we use biblical texts to do historical geography.

Literary History of Numbers 21 Literary analysis has often been dismissed by historical geographers as too speculative,10 especially because scholars so often disagree about the literary history of a text, and because different proposals seem arbitrary, it is difficult to determine which is better. Though it is, of course, impossible to eliminate indeterminacy, genre constraints that emerge from ancient literature enable a relatively strong interpretation of the literary history of Num 21. The itinerary genre is an element of cultural repertoire whose features we can grasp by studying extant examples. One thing we learn is that individual itineraries vary in form from one to the other, but no extant itinerary changes form in the middle. This knowledge about how the genre is typically used is highly relevant, because it generates a significant interpretive yield: Num 21 is not coherent, but we can identify precisely where it is not coherent and do so with relative ease.11 Verses 10–11a exhibit one form, with narrative preterites for setting out and camping and repetition of the place name: ‫ויסעו בני ישראל ויחנו באבת ויסעו מאבת ויחנו בעיי‬ ‫העברים‬. This happens to be the same form used in the Priestly itinerary notices throughout the wilderness narrative, beginning in Exod 12, so these two itinerary notices likely belong to the Priestly version of the wilderness narrative.12 Verses 12–13a have a different form, starting with ‫ משם‬instead of repeating the place name and using a perfect for setting out: ‫משם נסעו ויחנו בנחל זרד משם נסעו ויחנו‬ ‫מעבר ארנון‬. Verses 18b–20 have yet a third form, without any verbs at all: ‫וממדבר‬ ‫מתנה וממתנה נחליאל ומנחליאל במות ומבמות הגיא אשר בשדה מואב ראש הפסגה‬ ‫ונשקפה על פני הישימן‬. through study of remains that happen to be preserved for us the way we could an overt act of communication, but the cognitive principle of relevance still applies: we will seek to process what we have so as to maximize its relevance to an interpretive task. 10  J. M. M iller, “Site Identification: A Problem Area in Contemporary Biblical Scholarship,” ZDPV 99 (1983), 119–129, here 124. 11  Something is judged relevant to the extent that its interpretive yield is high and to the extent that this high interpretive yield is attained with relative ease; see Wilson and Sperber, “Relevance Theory” (see n. 9). 12  See Roskop, Wilderness Itineraries (see n. 8), 178, for a list of the itinerary notices that structure the Priestly version of the wilderness narrative.

For the Border of the Ammonites Was . . . Where?

765

Relevance theory involves the idea that inference from context is involved in processing not only the implicatures of an utterance but also its explicit content.13 Place names in itinerary notices – like the references to Wadis Zered and Arnon in verses 12–13a and to Bamoth, Pisgah, and Yeshimon in verses 18b–20 – are “points of access” to related information in the readerʼs cognitive environment. To understand these place names and their significance for interpreting the itinerary notices, we scan our cognitive environment for potentially relevant elements of context until we find one that is highly relevant and can easily be applied. The relevant context here happens to have nothing to do with where the places are or in what periods they might have been occupied but concerns their function as setting in other pieces of literature: Wadis Zered and Arnon figure prominently in the geography of Deuteronomyʼs frame, whereas Bamoth, Pisgah, and Yeshimon figure in the setting of the Balaam narrative. When we combine this with information about the itinerary genre available in our cognitive environment – that itinerary notices connect material set in one place with material set in another in a coherent chain – we can quite readily understand these two sets of itinerary notices as using geography creatively in service of a literary goal: to integrate disparate pieces of literature into the wilderness narrative by bridging the setting. Verses 12–13a integrate Deuteronomy, and verses 18b–20 integrate the Balaam narrative. The different forms of itinerary notice enable us to see the seams, and inferences based on contextual knowledge about the geography and the typical function of the itinerary genre enable us to identify the literary goals.14 These two revisions to the basic Priestly chain of itinerary notices, which ends with Num 21:10–11a, are not particularly messy from a geographical or a literary perspective. Consistent use of the itinerary genre despite the differences in form helps the reader with what Iser called “consistency-building” – in this case, reading the composite as a whole, even as its composite character is readily evident. The setting is also consistently in Moab despite the evidence of diachronic development. We can blame the “geographical hodgepodge” of Num 21, as J. Maxwell Miller called it, on a third revision that involves Deut 2 as well as Num 21.15 The first part of Deut 2 is concerned with land ideology, which limits the Israelite promised land to Cisjordan and assigns the lands of Edom, Moab, and Ammon in Transjordan to the descendants of Esau and Lot. Edom is south

 Wilson and Sperber, “Relevance Theory” (see n. 9).  See Roskop, Wilderness Itineraries (see n. 8), 50–82, 178, 204–209, for a full presentation of the argument I have summarized here. 15  Miller, “Israelite Journey” (see n. 2), 587. See A. R. Erisman, “Transjordan in Deuteronomy: The Promised Land and the Formation of the Pentateuch,” JBL 132 (2013), 769–789, and Roskop, Wilderness Itineraries (see n. 8), 209–215, for a full presentation of the argument I summarize here. 13 14

766

Angela Roskop Erisman

of Wadi Zered, Moab is north of it, and the Israelites go straight through these territories that they are not permitted to conquer. Then we encounter a geographical problem. Deuteronomy 2:16–23 sets up the expectation that the Israelites will encounter the Ammonites, whom they are also not permitted to conquer, once they cross the Arnon.16 But that is not what happens. When they cross the Arnon in v. 24, they encounter Sihon, king of the Amorites, and conquer his territory. And we learn in v. 27 that the border of the Ammonites is up by the Jabbok. So where is the border of the Ammonites? At the Arnon? Or up by the Jabbok? Numbers 21 contains related geographical problems. Verse 20 has the Israelites arrive in Moab, and the place names here anticipate the setting of the Balaam narrative. But they arrive in Moab again for the Bilaam narrative in vv. 1–2 of chapter 22, creating a double arrival that violates a constraint of the itinerary genre – it conveys linear movement – not to mention constraints on interpretation due to our understanding of how bodies move through space and time – we do not arrive at a place twice unless we have left it in the interim. So this narrative is not plausible as it stands. These two geographical problems, as well as the conflict between routes around and through Transjordan in both Num 21 and Deut 2, are the residue of a revision that sought to extend the ideology of Israelʼs promised land to include Transjordan. Accomplishing this literary goal required the scribe to reconceptualize the land between the Arnon and the Jabbok as Amorite land, which is permissible to conquer, and write a conquest narrative for it. In Num 21, this goal is achieved by means of two glosses to the itinerary notices. Verse 13b limits Moab to territory south of the Arnon, as it is in Deut 2, and makes territory north of the Arnon Amorite: ‫אשר במדבר היצא מגבול האמרי כי‬ ‫ארנון גבול מואב בין מואב ובין האמרי‬. Verse 11b takes the Israelites around settled area to prepare for conquest of that Amorite territory, circumventing the route through settled area: ‫במדבר אשר על פני מואב ממזרח השמש‬.17 The Sihon and Og conquest episodes justify Israelite possession of territory in Transjordan as part of the promised land. Finally, the itinerary genre is used in Num 22:1–2 to integrate the revision back into the setting of the Balaam narrative, resulting in the double arrival in Moab. In Deut 2ʼs sketch of Transjordanian geography, Moab already ends at the Arnon, and it is the Ammonites who stand in the way of an Israelite conquest of territory north of the Arnon. To make room for Amorite (i.e., conquerable) territory and the narrative of its conquest from Sihon, the scribe simply pushes Ammon out of the way, indicating in v. 37 that it is up by the Jabbok, despite the conflict this creates with the expectation set up in  Miller, “Israelite Journey” (see n. 2), 583.  H. Seebass, “Edom und seine Umgehung nach Numeri XX–XXI: Zu Num XXI 10–13,” VT 47 (1997), 255–262, here 256, posits that ‫ ממזרח השמש‬may be a “late gloss” based on comparison to Deut 2:8 but offers no clear sense of what the purpose of the gloss might be. 16 17

For the Border of the Ammonites Was . . . Where?

767

Deut 2:16–23 that the land of the Ammonites is just the other side of the Arnon. The revisions in Numbers and Deuteronomy constitute a coordinated effort: they serve the very same literary goal, and each shows cognizance of the other. The strategy of taking the Israelites around Moab is more prominent in Num 21, but it is echoed in Deut 2:8. And the (shifted) border of the Ammonites is part of the strategy in Deut 2, but it gets a reference in Num 21:24 – seemingly out of place, but understandable once one grasps the literary history of these two texts. Numbers 21 is indeed a bridge between Tetrateuch and Pentateuch. But it is not the tradition-historical vacuum Martin Noth and John Van Seters thought. Thanks to our awareness of genre constraints and an understanding of how geography is used in literature that allows us to go beyond a simple correspondence model, we can see the literary history of this most muddled of pentateuchal texts rather clearly: the Priestly wilderness narrative was extended to incorporate Deuteronomy; that composite was further extended to incorporate the Balaam narrative; and the ensuing composite was revised to include Transjordan in an idealized promised land. We can also see the cause of the geographical mess – conflicting routes around and through Moab and conflicting ideas about where the border of Ammon is – as a scribe sacrificed geographical plausibility in the interest of what, to him, must have been a very important ideological goal.

Implications for How We Use Biblical Texts to Do Historical Geography The scribes who wrote and revised Numbers 21 were motivated to anthologize disparate literatures into a single, more-or-less coherent narrative and to weigh in on what territory should be included in an idealized promised land far more than they were motivated to write history.18 A scribeʼs goals will influence the way he uses geography. If he is writing an administrative document in order to stipulate the route to be taken by a caravan, the constraints on his representation of geography will be quite strong; his document does not achieve its goal if the route cannot be followed. Writing history would also entail relatively  We should not entirely dismiss the possibility of historiographical goals in this passage. One motivation for including Transjordan in the idealized promised land could be that Israelites had lived in this area at one or more points of their history; land claims are often legitimized by appeal to a groupʼs history in a particular place, whether historical, fictitious, or a blend of the two. It is, however, impossible to talk about this motivation in any detail based on the text of Num 21, because geography and history from one or more periods may inform this ideologically driven text, and the geography here is not specific enough, nor do we have enough information archaeologically, to permit untangling such complexity. The conquest episode may justify inclusion of this territory in the idealized promised land, but we are not presently in a position to extract from it anything about the actual extent or history of Israelite settlement in any particular period. 18

768

Angela Roskop Erisman

strong constraints on the way geography is represented, involving not only the location of a place but also its settlement history and the events that took place there, even as the narrative may be heavily emplotted.19 But if his goals are influenced primarily by literary and ideological concerns, the constraints on use of geography tend to be looser. Looser constraints do not necessarily mean that literature becomes untethered from geography. They simply mean that the scribe has more flexibility, enabling him to tap any geography he knows and use it in creative ways that draw on relevant context more selectively, as is the case with references to the Wadi Zered, Wadi Arnon, Bamoth, Pisgah, and Yeshimon in the Num 21:12–13a and 18b–20 itinerary notices, and to stretch or even sacrifice plausibility, as the palimpsest created by the glosses in Num 21:11b and 13b illustrates. Literary texts that are highly motivated by ideology may nonetheless give us information we do not happen to get from other extant sources, so we should not write them off for the purposes of historical geography. It is important to use such texts carefully. This is often said, but rarely is it explained what being careful involves, other than being skeptical about the yield. One thing it involves is attention to the details of literary history, as the quality of our literary analysis can affect the quality of information we get. Indeed, Miller pushed for attention to literary analysis despite the difficulties with it, noting that excavation is often no less subjective than literary analysis, and “failure to take [literary analysis] into account is as serious a methodological error as would be failure to take into account the stratigraphy of a tell.”20 But the literary analysis on which he relied – the effort by Van Seters and others to understand Num 21 as a single-handed effort to conflate Deut 2–3 and Judg 11, resulting in garbled geography – is weak in its ability to make sense of the finer literary details (like aberrations in use of the itinerary genre) and especially the geography in the text. Miller framed the insurmountable topographical problem with Num 21:11–12 like this: “[H]ow could the Israelites, supposedly coming from the south, have reached Iye-abarim in the region east of Moab before crossing the Zered?”21 Once we see that Num 21 is not a single-authored text and that the second half of v. 11, which identifies Iye-abarim as being east of Moab, is part of a revision that incorporates Transjordan into the promised land, the problem is easy to solve: in the previous version of the text, prior to the addition of ‫במדבר‬ ‫אשר על פני מואב ממזרח השמש‬, the Israelites go straight into Moab through Wadi Zered. Iye-abarim is not really in the wilderness east of Moab. The impression that it is results from the manipulation of geography required by the scribeʼs strategy for incorporating a conquest narrative for territory north of the Arnon 19  On emplotment of history, see H. White, “The Historical Text as Literary Artifact,” in Tropics of Discourse: Essays in Cultural Criticism (Baltimore: Johns Hopkins University Press, 1978), 81–100. 20  Miller, “Site Identification” (see n. 10), 124. 21  Miller, “Israelite Journey” (see n. 2), 585–587, quote on 588.

For the Border of the Ammonites Was . . . Where?

769

by circumventing the route through Transjordan. We cannot attribute the geographical muddle in Num 21 to scribal ignorance of landscape; rather, it came about as the scribe sacrificed geographical plausibility – creating a palimpsest of different routes through Transjordan and making Iye-abarim appear to be where it is not – in the interest of achieving his ideological goal. Willingness to sacrifice plausibility may have created a messy text, but it is not the same thing as ignorance of or lack of concern for geography. Attention to the literary history at the end of the wilderness narrative more generally allows us to say more: Iye-abarim is not even likely to be in Moab. The place name appears in Num 21:10–11a, last in the series of itinerary notices that structures the Priestly wilderness narrative before the incorporation of Deuteronomy and the Balaam narrative.22 Geographically speaking, the end of this Priestly narrative is hinted at not only in the conquest of Canaanite sites from the south in Num 21:1–3 but also in Num 27, where Mosesʼs death is set at an unidentified mountain of the Abarim range. This episode was truncated and moved to Deut 32 when Deuteronomy was added, and the mountain was specified as Nebo in the Abarim. We have typically assumed, based on this reference, that the Abarim range is in Moab. But the oracle in Jer 22:18–30 lamenting Jehoiakimʼs exile from every side – Lebanon to the north, Bashan to the east, and Abarim – suggests that the Abarim range may instead be south of the land. The discrepancy between Deut 32:49, which locates Mount Nebo in the Abarim range, and Deut 34:1, which locates it in the Pisgah range, is then easily explained. Deuteronomy 32:49 copied ‫ עלה אל הר העברים הזה‬from Num 27:12 in order to retain continuity between the now two installments of Mosesʼs death scene but specified the mountain as Nebo in order to set the scene in Moab, appropriate for the setting of Deuteronomy. Deuteronomy 34 – which is part of the effort to incorporate Transjordan into the idealized promised land, because the land Moses views includes territory in Transjordan – can be understood to repair the reference, situating Nebo properly in the Pisgah range.23 If Abarim is south of the land, Iye-abarim likely is too. Were we to look for it, we should look not east of Moab, and not even in Moab, but in the southern Negev. A stronger literary analysis of Num 21 not only makes both literary and topographical sense of the text but also generates a potentially useful insight for historical geographers – one we do not get from any other extant source.24 22  See Roskop, Wilderness Itineraries (see n. 8), 178, for a list of Priestly itinerary notices, and 193–203, for discussion of how itinerary notices were used to marginalize conquest from the south (Num 21:1–3) and incorporate Joshua into the Priestly wilderness narrative, with its conquest from the east. 23  R oskop , Wilderness Itineraries (see n. 8), 272–274; E risman , “Transjordan in Deuteronomy” (see n. 15), 782–787. On reference repair, see H. H. C lark , Using Language (Cambridge: Cambridge University Press, 1996), 284–285. 24  The only other reference to Iye-abarim available to us is Num 33:44, which is dependent on Num 21; see R oskop , Wilderness Itineraries (see n. 8), 223–232, for discussion of the

770

Angela Roskop Erisman

Just as we learn something about the location of Iye-abarim by peeling away the revisions in Num 21 that distorted it, we can also look beneath the revision to Deut 2 that shifted the Ammonite border north to the Jabbok in order to accommodate a conquest narrative and see that the underlying text understands Ammonite territory to begin instead at the Arnon. This insight is less useful for two reasons. First, though it is generally accurate, it does not add anything to what we already know in more detail from other sources. We start to see a distinctive Ammonite material culture emerge in this area during Iron IIC as secondary state formation takes place under the Assyrian umbrella:25 distinctive pottery, distinctive iconography (e.g., royal statues with the atef-crown), and both native and Assyrian seals and texts that refer to Ammon and its kings.26 Archaeologists differ in their analysis of Ammonʼs geographical extent: Raz Kletter proposes a more limited Ammon based on his interpretation of the towers in the area around Amman (ancient Rabbath Ammon),27 whereas Larry G. Herr proposes a more extensive Ammon by tracing Ammonite material culture as far south as Madaba, as far west as the Jordan (e.g., Nimrin), and as far north as the Jabbok (or just north of it, e.g., Deir Alla, Tell el-Mazar).28 Border ambiguities

relationship between the Num 33 itinerary and the wilderness narrative. 25  A. H. Joffe, “The Rise of Secondary States in the Iron Age Levant,” JESHO 45 (2002), 425–467. On Assyrian involvement in Transjordan, see M. Weippert, “The Relations of the States East of the Jordan with the Mesopotamian Powers during the First Millennium BC,” in Studies in the History and Archaeology of Jordan III (ed. A. Hadidi; Amman: Department of Antiquities, 1987), 96–105; P. Bienkowski, “Transjordan and Assyria,” in The Archaeology of Jordan and Beyond: Essays in Honor of James A. Sauer (ed. L. Stager et al.; SAHL 1; Winona Lake, IN: Eisenbrauns, 2000), 44–58; N. Na‍ʾaman, “Province System and Settlement Pattern in Southern Syria and Palestine in the Neo-Assyrian Period,” in Ancient Israel and Its Neighbors: Interaction and Counteraction (Winona Lake, IN: Eisenbrauns, 2005), 220–237. 26  L. G. H err , “Shifts in Settlement Patterns of Late Bronze and Iron Age Ammon,” in Studies in the History and Archaeology of Jordan IV (ed. A. Hadidi; Amman: Department of Antiquities, 1992), 175–177; idem , “The Ammonites in the Late Iron Age and Persian Period,” in Ancient Ammon (ed. B. MacDonald and R. W. Younker; SHCANE 17; Leiden: Brill, 1999), 219–237; B. M ac D onald , “East of the Jordan”: Territories and Sites of the Hebrew Scriptures (ASOR Books 6; Boston: American Schools of Oriental Research, 2000), 160; L. G. Herr and M. Najjar, “The Iron Age,” in The Archaeology of Jordan (ed. B. MacDonald et al.; Levantine Archaeology 1; Sheffield: Sheffield Academic Press, 2001), 335–337; P. M. M. Daviau and P. E. Dion, “Independent and Well-Connected: The Ammonite Territorial Kingdom in Iron Age II,” in Crossing Jordan: North American Contributions to the Archaeology of Jordan (ed. T. E. Levy et al.; London: Equinox, 2007), 301–307. 27  R. Kletter, “The Rujm El-Malfuf Buildings and the Assyrian Vassal State of Ammon,” BASOR 284 (1991), 33–50. 28  H err , “Shifts in Settlement Patterns” (see n. 26), 175–177; idem , “The Iron Age II Period: Emerging Nations,” BA 60/3 (1997), 114–183, here 170; H err and N ajjar , “Iron Age” (see n. 26), 333. See MacDonald, East of the Jordan (see n. 26), 164–165, for a review of these and other proposals.

For the Border of the Ammonites Was . . . Where?

771

notwithstanding, we would know that Ammon is, generally speaking, north of the Arnon even if we did not have Deut 2. Second, even if Deut 2 provided us with the only piece of data available, it is much too sketchy to be useful to historical geographers in drawing borders. Being careful with literary texts involves not only attention to details of literary history but also appreciation for the character of the information we get so that we do not overinterpret and consequently misuse it. Miller notes that “we must be cautious not to extract from these passages [Num 21, Deut 2, and Judg 11] more precise information” than they can provide.29 He took the lack of detail to be indicative of lack of knowledge, but the issue is really literary goals and their influence on the use of geography. Limiting the Israelite promised land to Cisjordan and assigning Edom, Moab, and Ammon to the descendants of Esau and Lot did not require detailed geography; a rough sketch of the different territorial entities using Wadis Zered, Arnon, and Jabbok as rough boundaries sufficed to make the ideological point.30 We cannot take this rough sketch to mean that the scribe had no detailed knowledge of Transjordanian geography; we can conclude only that he did not give us a detailed description. Indeed, the scribe who revised Num 21 and Deut 2 in order to include the conquest episode revealed more detailed geographical knowledge because it was in his interest to do so: the poem in Num 21:27b–30 relates Sihonʼs conquest of territory from Moab before the Israelite conquest and thus “eliminate[s] any Moabite claim on the Mishor.”31 It would not have been necessary to include this poem unless there was information in the cognitive environment available to him and his initial audience that the plains and plateau just north of the Arnon had been Moabite at one point or another. In each case, the scribe gives his readers a point of access not to comprehensive geographical knowledge but to the geographical knowledge relevant for interpretation of his text. The assumption that he would have given us more detailed geography if he had known it presumes that was his literary goal in the first place – a misconstrual of the passageʼs literary goals and their influence on the use of geography. Failing to take account of how literary goals influence the character of geographical information in a text can lead one to make claims about the interpretation of archaeological remains that the texts cannot support. One example of this involves Iron II Jordan Valley sites like Nimrin, Tell al-Mazar, or Deir Alla, whose material culture includes some Ammonite material but is “not as  Miller, “Israelite Journey” (see n. 2), 588. Miller is concerned here with Moab, but the same would apply to Edom and Ammon, whose geography is also sketchy, if not problematic. 30  B. R outledge , Moab in the Iron Age: Hegemony, Polity, Archaeology (Archaeology, Culture, and Society; Philadelphia: University of Pennsylvania Press, 2004), 48, notes that the wadis Yarmuk, az-Zarqa (Jabbok), al-Mujib (Arnon), and al-Hasa (Zered) “have usually marked administrative subdivisions since Late Antiquity,” so it is possible that use of the rivers was enabled by a cognitive environment that involved this information. 31  Routledge, Moab in the Iron Age (see n. 30), 45–46, quote on 46. 29

772

Angela Roskop Erisman

homogenous” as the clearly Ammonite sites on the plain like Tell al-ʿUmayri.32 The definition of Ammonite territory in Judg 11:13 as extending from the Arnon to the Jabbok as far as the Jordan supports those who are inclined to view these Jordan Valley sites as Ammonite.33 Bruce Routledge identifies the problem with this interpretation archaeologically: “What is hugely problematic is that the scholars relating national/ethnic identities to the distribution of material culture provide no conceptual road map to explain how we get from one to the other.”34 This is a problem even with sites like Tell al-ʿUmayri, where the material culture is homogenous but its connection with group identity is undertheorized, but even more so at possible fringe sites where a variety of other factors could explain the mixed material culture. But there is also a literary problem. The territorial dispute between Jephthah and the Ammonite king in Judg 11 plays off of an ambiguity in the composite text of Deut 2 about whether territory north of the Arnon is Amorite (as the Sihon conquest indicates) or Ammonite (as Deut 2:16–23 leads one to expect). The Ammonite king claims that “when they came up from Egypt, Israel took my land, from the Arnon to the Jabbok and to the Jordan. Now, restore it peaceably” (Judg 11:13). His claim has its roots in Deut 2 as it would have looked before this territory was reconceptualized as Amorite in order to accommodate the conquest narrative. Jephthahʼs counterclaim plays the ambiguity to his advantage by citing Israelʼs legitimate conquest of Sihon the Amorite (Judg 11:19–23) instead.35 The definition of Ammonite territory extending from the Arnon to the Jabbok as far as the Jordan in Judg 11:13 is thus nothing more than a fleshed-out version of the same roughly sketched Transjordanian geography we find in Deut 2 and no more useful than Deut 2 for defining borders. A second example involves efforts to connect Late Bronze and early Iron Age settlement patterns in territory north of the Arnon with the various groups – Israelites, Ammonites, Moabites, and Amorites – that play a role in Num 21, which is set in a period prior to the emergence of an Israelite monarchy. The LB-Iron I transition is characterized by sedentarization, particularly evident at Tell al-ʿUmayri, which is similar in material culture to highland sites such as

 Herr and Najjar, “Iron Age” (see n. 26), 335–337, quote on 335.  E.g., E. Lipiński, On the Skirts of Canaan in the Iron Age: Historical and Topographical Researches (OLA 153; Leuven: Peeters, 2006), 299. 34  Routledge, Moab in the Iron Age (see n. 30), 47. 35  Jephthah also offers a historical retrospective that outlines the Israelitesʼ route through Transjordan from Yam Suf to Kadesh, around Edom and then Moab (Judg 11:16–18). This particular route comes not from Num 20–21 but from Num 33, which cleaned up the geographical mess that resulted from various additions and revisions to Num 21 and is the only other text to take the Israelites around both Edom and Moab (see Roskop, Wilderness Itineraries [see n. 8], 223–232, for discussion of Num 33). 32 33

For the Border of the Ammonites Was . . . Where?

773

Shechem and Izbet Sartah in Cisjordan but earlier, larger, and more developed.36 Øystein LaBianca and Randall Younker speak of the entire area as Ammon in this period.37 Herr is simultaneously more cautious and more confident: he notes that we “cannot be certain which tribal group settled at ʿUmayri”38 yet pushes for understanding it as a Reubenite site, given both its relationship with Israelite sites in Cisjordan and the fact that it was more prominent and settled earlier, which fits Reubenʼs status as firstborn among Israelite tribes.39 Herrʼs reliance on Frank Moore Crossʼs Reubenite hypothesis40 avoids biblical texts like Josh 13:15–28 and Num 32:34–38 that relate to conquest and occupation of this area – but on the basis of which one could make an equal claim that ʿUmayri is Reubenite or Gadite, because they differ in their understanding of which tribe settled where.41 But there is a bigger problem than this. Routledge argues that we can connect national or ethnic identity to material culture only where we see evidence in material culture of discursive practices that enact identity. An entity like Ammon is constructed and asserted through textual and cultural practices “within a space

36  L. G. H err , “Tell El-ʿUmayri and the Madaba Plains Region during the Late BronzeIron Age I Transition,” in Mediterranean Peoples in Transition: Thirteenth to Early Tenth Centuries BCE (ed. S. Gitin et al.; Jerusalem: Israel Exploration Society, 1998), 251–266; idem , “The Settlement and Fortification of Tell Al-ʿUmayri in Jordan during the LB/Iron I Transition,” in The Archaeology of Jordan and Beyond: Essays in Honor of James A. Sauer (ed. L. Stager et al.; SAHL 1; Winona Lake, IN: Eisenbrauns, 2000), 167–179; H err and Najjar, “Iron Age” (see n. 26); C.-H. Ji, “Tribes and Sedentarization in the Madaba Plains and Central Jordan during the Iron I and Ottoman Periods,” in The 1994 Season at Tall Al-ʿUmayri and Subsequent Studies (ed. L. G. Herr et al.; Madaba Plains Project 5; Berrien Springs, MI: Andrews University Press, 2002), 348–357. 37  Ø. S. LaBianca, “Intensification of the Food System in Central Transjordan during the Ammonite Period,” AUSS 27/3 (1989), 169–178, here 175; R. W. Y ounker , “Hesban: Its Geographical Setting,” in Hesban after 25 Years (ed. D. Merling and L. T. Geraty; Berrien Springs, MI: Institute of Archaeology, 1994), 55–63, here 60–63; idem , “The Emergence of Ammon: A View of the Rise of Iron Age Polities from the Other Side of the Jordan,” in The Near East in the Southwest: Essays in Honor of William G. Dever (ed. B. A. Nakhai; AASOR 58; Boston: American Schools of Oriental Research, 2003), 153–176. 38  Herr, “Settlement and Fortification” (see n. 36), 177. 39  L. G. Herr, “Tall Al-ʿUmayri and the Reubenite Hypothesis,” EI 26 (1999), 64*–77*; idem , “Tall Al-ʿUmayri and the Bible,” Journal of the Adventist Theological Society 9/1–2 (2000), 48–56; and L. G. H err and D. R. C lark , “Excavating the Tribe of Reuben,” BAR 27/2 (2001), 36–47, 64, 66, where they present the correlation with the Reubenite hypothesis as quite strong. Note that P. J. R ay Jr., Tell Hesban and Vicinity in the Iron Age (Hesban 6; Berrien Springs, MI: Andrews University Press, 2001), 79, makes a similar argument for Tell Hesbanʼs Iron I remains. 40  Most recently published in F. M. Cross, From Epic to Canon: History and Literature in Ancient Israel (Baltimore: Johns Hopkins University Press, 1998), 53–70. 41  As pointed out by Routledge, Moab in the Iron Age (see n. 30), 46.

774

Angela Roskop Erisman

where it is not the only possibility.”42 In other words, we cannot talk about Ammonites (or Reubenites or Gadites) in the archaeological record until we have evidence of such practices that can be pegged to a particular historical period and geographical area. Even though this has not yet been done for Iron II Ammon, it potentially could be, because we have a distinctive material culture and texts, both biblical and Assyrian. But we have no such texts for Iron I. Numbers 21 and Deut 2 come relatively late in the formation of the Pentateuch, and their conquest episodes are the latest revisions in them, highly driven by land ideology and only obliquely (at best) by historiographical concerns.43 These conquest episodes certainly constitute a discursive practice related to identity formation – but in a much later period.44 One should not mistake the setting of a piece of literature for the sociohistorical context in which it was produced and initially read. The limits on our ability to use biblical texts for historical geography are thus not imposed by the mere presence of ideology. They have to do with the quality and character of information we get after careful analysis. Some of these limits are of our own making, due to interpretive mistakes that can be fixed. Stronger literary analysis and attention to the way literary goals affect the use of geography can yield better information and enable us to appropriately calibrate our use of what limited information we may have. But other limits are due simply to lack of information available to us in our cognitive environment. Attention to literary history corrects previous misunderstandings about where Iye-abarim might be located and gives us a new potential lead, but we lack the information that would enable us to follow it. Without knowledge of Iye-abarimʼs location and settlement history, we can only guess at site identification.45 We know little about the location of Iye-abarim other than that it is likely in the southern Negev and that it would come after Oboth along a route. For the latter detail to be useful, we would need secure knowledge of where Oboth is, which we lack, and what direction of movement is envisioned by the itinerary notices in Num 21:10–11a,  Routledge, Moab in the Iron Age (see n. 30), 47–48.  As would be the texts related to settlement of this area, including Josh 13, Num 32, and Josh 22. 44  Exactly what period is rather difficult to say, as we can speak only in relative terms. I argue that the Priestly series of itinerary notices constitutes an effort to emplot the wilderness narrative as a march (back) to Zion (from Babylon) with the vessels of Yahweh, inspired by Isa 52 (Roskop, Wilderness Itineraries [see n. 8], 169–175). This makes the revisions to Num 21, which incorporate Deuteronomy and the Bilaam narrative into the wilderness narrative and which push for the inclusion of Transjordan in the idealized promised land, late exilic at the earliest and more likely postexilic, depending on how far apart in time they took place. 45  M iller , “Site Identification” (see n. 10), 119–129, and A. F. R ainey , “Historical Geography,” in Benchmarks in Time and Culture: Essays in Honor of Joseph A. Callaway (ed. J. F. Drinkard Jr. et al.; Archaeology and Biblical Studies 1; Atlanta: Scholars Press, 1988), 353–368. 42 43

For the Border of the Ammonites Was . . . Where?

775

which we also lack, given that these itinerary notices would have originally narrated an entry into Canaan from the south, now obscured by the revisions that take the Israelites around the south end of the Dead Sea to enter from the east.

Conclusions Numbers 21 turns out not to generate much that is useful for historical geography. When it comes to the border of Ammon, it does not tell us anything we do not already know, and know with more refinement, from other sources. When it comes to the location of Iye-abarim, a stronger understanding of literary history corrects a previous misunderstanding, but accident of time and preservation has not been in our favor when it comes to the information we would need in order to make headway in an effort to locate the site. In one sense, then, this article may be much ado about nothing. My goal here has actually been to offer a better understanding of why it offers us so little, and that is something, because it can help us become more sensitive readers of other texts. Numbers 21 offers so little for historical geography not because of the geographical muddle – which can be unknotted – or because the scribes were ignorant of Transjordanian geography. It is also not because Numbers 21 is a relatively late, ideologically motivated text, and late, ideologically motivated texts are in principle useless. More than anything, it has to do with the fact that, in this particular case, the character of the information is too general and cannot be confidently linked to a particular historical period. We should not assume that this will be the case with every text. A different scribeʼs literary goals may be different, the techniques he uses to achieve those goals may be different, and he may be less willing to sacrifice plausibility in service of them. We cannot generalize but must apply the ideas I have put forward here in a careful reading of each new pericope. Working with Num 21 also reveals how important it is to calibrate our arguments appropriately. Literary analyses are stronger or weaker depending on how much (or how little) they account for; to speak in these terms is to be honest about the fact that more than one interpretation of a text may be viable, and it gives us a sense of how heavily we can lean on the yield of one over another. Given the lack of clarity, even archaeologically, about where the borders of Ammon are in Iron II, it is important to make clear the warrants for one argument or another and offer appropriate qualifications. We can talk about sedentarization processes during the LB-Iron I transition, and it certainly seems possible that they could be related to the emergence of Israel and Ammon, but we have no solid hook on which to hang an argument about which ethnic entities might have been involved, and this is worth being transparent about. Finally, when faced with a roadblock like the lack of information we would need to locate Iye-abarim, it may be best simply to say, “I donʼt know.”

776

Angela Roskop Erisman

Finally, I have sought to illustrate the advantage of an interdisciplinary approach to biblical interpretation as well historical geography. Our knowledge of geography plays a role in biblical interpretation, and interpretation impacts what usable information is available to historical geographers. Thomas E. Levy and Thomas Higham associate literary criticism with the minimalist school and contrast it with the “historicist,” or maximalist, school, which values “command of the ancient languages, history, and archaeology” and with which they prefer to be associated.46 But there is nothing minimalist about the careful application of theory to literary analysis. Even as the details of my interpretation of Num 21 differ from Millerʼs, I mean to advocate for a similar point. Historians and archaeologists cannot properly understand and use biblical literature without engaging the methods appropriate to the study of literature;47 biblical scholars likewise need the data provided by archaeologists and historians to enrich the cognitive environment in which we read texts if we are to have the cultural repertoire needed to yield rich and productive interpretations. The often very complex and refracted historicity of biblical literature and its literary character are equally at stake.

46  T. E. Levy and T. Higham, “Introduction: Radiocarbon Dating and the Iron Age of the Southern Levant,” in The Bible and Radiocarbon Dating: Archaeology, Text, and Science (ed. T. E. Levy and T. Higham; London: Equinox, 2005), 3–14, here 6. 47  See W. K. V yhmeister , Historical Foundations: Studies of Literary References of Hesban and Vicinity (Hesban 3; Berrien Springs, MI: Andrews University Press, 1989), 1–23, for an example of what happens when attempts at writing history are made with little attention to the literary issues involved.

Part Eight Do the Pentateuchal Sources Extend into the Former Prophets?

Introduction Konrad Schmid Both redaction-critical and source-critical approaches usually accept that the Pentateuch is composed of sources.1 But are the sources, however scholars may determine and label them, reworked in the Pentateuch limited to the literary horizon of Genesis–Deuteronomy? Or do at least some of them extend into the Former Prophets? In nineteenth- and early twentieth-century scholarship, it was not uncommon to assume that the pentateuchal sources J, E, and P extended into Joshua, Judges, Samuel, and even Kings.2 For example, Carl Cornill, Karl 1  Cf. K. S chmid , “Has European Scholarship Abandoned the Documentary Hypothesis? Some Reminders on Its History and Remarks on Its Current Status,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 17–30. 2  See C. H. C ornill , “Ein elohistischer Bericht über die Entstehung des israelitischen Königthums in I Samuelis 1–15 aufgezeigt,” Zeitschrift für kirchliche Wissenschaft und kirchliches Leben 6 (1885), 113–141; idem , “Noch einmal Sauls Königswahl und Verwerfung,” ZAW 10 (1890): 96–109; idem , “Zur Quellenkritik der Bücher Samuelis,” Königsberger Studien 1 (1887), 25–89; K. B udde, Das Buch der Richter (KHC 7; Freiburg: Mohr, 1897), xii–xv; idem , Die Bücher Richter und Samuel: Ihre Quellen und ihr Aufbau (Gießen: Ricker, 1890), 165–166, 268–269; idem , Die Bücher Samuel (KHC 8; Tübingen: Mohr, 1902), xii–xxi; idem , Geschichte der althebräischen Litteratur: Apokryphen und Pseudepigraphen von Alfred Bertholet (Leipzig: Amelangs, 1909), 57–59; I. B enzinger , Jahvist und Elohist in den Königsbüchern (BWAT 2; Berlin: Kohlhammer, 1921); idem , Die Bücher der Könige (KHC 9; Freiburg: Mohr, 1899); W. S taerk , Die Entstehung des Alten Testaments (SG 272; Berlin, 1918), 11–16; R. S mend , “JE in den geschichtlichen Büchern des AT,” ZAW 39 (1921), 181–217; G. H ölscher , “Das Buch der Könige, seine Quellen und seine Redaktion,” in Eucharistérion, Studien zur Religion und Literatur des Alten und Neuen Testaments (ed. H. Schmidt; FRLANT 19; Göttingen: Vandenhoeck & Ruprecht, 1923) 158–213; idem , Geschichtsschreibung in Israel: Untersuchungen zum Jahvisten und Elohisten (Lund: Gleerup, 1952); idem , Geschichte der israelitischen und jüdischen Religion (Gießen: Töpelmann, 1922), 135, n. 1; O. E issfeldt , Die Quellen des Richterbuches in synoptischer Anordnung ins Deutsche übertragen samt einer in Einleitung und Noten gegebenen Begründung (Leipzig: Hinrichs, 1925); idem , Einleitung in das Alte Testament unter Einschluß der Apokryphen und Pseudepigraphen sowie der apokryphen und pseudepigraphenartigen Qumran-Schriften: Entstehungsgeschichte des Alten Testaments (3rd ed.; NTG; Tübingen: Mohr, 1964), 178–179, 771; C. A. S impson , Composition of the Book of Judges (Oxford: Blackwell, 1957); see the overviews provided by H ölscher , Geschichtsschreibung, 7–19; E. Jenni, “Zwei Jahrzehnte Forschung an den Büchern Josua bis

780

Konrad Schmid

Budde, Immanuel Benzinger, Willy Staerk, Rudolf Smend Sr., and Cuthbert A. Simpson all reached this conclusion. J, E, and P were also clearly present at several points in the book of Joshua according to Julius Wellhausen.3 It was only with Martin Nothʼs 1938 commentary on Joshua that it became popular to interpret the book of Joshua (and accordingly the following books) without relying on the Documentary Hypothesis and thus no longer to seek and find J, E, and P in the book of Joshua.4 Together with Gerhard von Radʼs study from the same year,5 Noth established an influential theory that the Pentateuch and Former Prophets are to be held apart in terms of their compositional history. In current scholarship, however, the separation between the Pentateuch and the Former Prophets has started to soften up, at times even to disappear. Evidence of the problems in the Noth-von Rad separation model began to appear in the 1970s. First were the books by John Van Seters,6 Hans Heinrich Schmid,7 and Rolf Rendtorff,8 all from the mid-seventies, which in various ways suggested a much closer relationship between the Deuteronomistic History and the Pentateuch than the one proposed by the Noth-von Rad compromise. Van Seters and Schmid dated the Yahwist very close to the Deuteronomist and also detected theological affinities between the two.9 Differing somewhat Könige,” ThR 27 (1961), 1–32, 97–146; G. Fohrer, Einleitung in das Alte Testament (10th ed.; Heidelberg: Quelle u. Meyer, 1965), 212–257. An early critical assessment of this assumption is provided by R. Kittel, “Die pentateuchischen Urkunden in den Büchern Richter und Samuel,” TSK 65 (1892), 44–71. See in more detail K. S chmid , “The Emergence and Disappearance of the Separation between the Pentateuch and the Deuteronomistic History in Biblical Studies,” in Pentateuch, Hexateuch, or Enneateuch: Identifying Literary Works in Genesis through Kings (ed. T. B. Dozeman et al.; SBL Ancient Israel and Its Literature 8; Atlanta: Society of Biblical Literature, 2011), 11–24. 3  J. Wellhausen, Die Composition des Hexateuch und der historischen Bücher des Alten Testaments (3rd ed.; Berlin: Reimer, 1899), 116–134. 4  M. N oth , Das Buch Josua (HAT 1/7; Tübingen: Mohr Siebeck, 1938), vii–viii. Noth refers to A. A lt , “Josua,” in Werden und Wesen des Alten Testaments (ed. P. Volz et al.; BZAW 66; Berlin: Töpelmann, 1936), 13–29, for that approach. 5  G. von Rad, “Das formgeschichtliche Problem des Hexateuchs (1938),” in Gesammelte Studien zum Alten Testament (TB 8; München: Kaiser, 1958), 9–86; English translation, “The Form-Critical Problem of the Hexateuch,” in The Problem of the Hexateuch and Other Essays (trans. E. W. Trueman Dickens; London: SCM Press, 1984 [1st ed., 1966]), 1–78. 6  J. Van S eters , Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975). 7  H. H. S c h m i d , Der sogenannte Jahwist: Beobachtungen und Fragen zur Pentateuchforschung (Zürich: Theologischer Verlag, 1976). 8  R. R endtorff , The Problem of the Process of Transmission in the Pentateuch (trans. J. J. Scullion; JSOTSup 89; Sheffield: Sheffield Academic Press, 1990); German original, Das überlieferungsgeschichtliche Problem des Pentateuch (BZAW 147; Berlin: de Gruyter, 1977). 9  To a certain extent, this was a return to Wellhausen, who had already found the Yehowist (that is, the combined JE) and the Deuteronomist to be kindred spirits (Geistesverwandtschaft; see Wellhausen, Composition [see n. 3], 94, n. 1).

Introduction

781

from Van Seters and Schmid, Rendtorff argued for a compositional model of the Pentateuch similar to that for the Deuteronomistic History proposed by Martin Noth. Rendtorff theorized that there were major text blocks not only in Deuteronomy–Kings but also in Genesis–Deuteronomy that were subsequently linked together by a Deuteronomistic redaction layer. Erhard Blum provided a significant breakthrough in 1984, with his book on the composition of the ancestorsʼ story in Gen 12–50, and again in his 1990 companion volume on Exodus–Deuteronomy.10 He extended and elaborated on Rendtorffʼs view that the Pentateuch is basically shaped by Deuteronomistic and Priestly composition layers. With regard to the Deuteronomistic texts in the Pentateuch, he developed the notion that they were composed within a literary horizon that overarches both the Pentateuch and the Deuteronomistic History.11 The influence of Blumʼs position, especially in its extended version from 1990, can be seen by the fact that many English-speaking introductions on the Old Testament assume D and P layers throughout the Pentateuch.12 In the wake of this overarching D perspective in Genesis–Kings, it has become customary to speak either of the so-called Primary History (as opposed to the Chroniclerʼs work as the Secondary History) 13 or of the  E. B lum , Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984); idem , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990). 11  B lum , Studien (see n. 10), 109. See his self-correction in idem , “Die literarische Verbindung von Erzvätern und Exodus: Ein Gespräch mit neueren Endredaktionshypothesen,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 119–156. 12  Cf., e.g., J. B lenkinsopp , The Pentateuch: An Introduction to the First Five Books of the Bible (New York: Doubleday, 2000). 13  E.g., D. N. F reedman , “The Law and the Prophets,” in Congress Volume: Bonn, 1962 (ed. G. W. Anderson et al.; VTSup 9; Leiden: Brill, 1963), 250–265, esp. 251, 254, 257; S. M andell and D. N. F reedman , The Relationship between Herodotusʼ History and Primary History (SFSHJ 60; Atlanta: Scholars Press, 1993), ix (see also 85); P. J. K issling , Reliable Characters in the Primary History: Profiles of Moses, Joshua, Elijah and Elisha (JSOTSup 224; Sheffield: Sheffield Academic Press, 1996); E. B en Z vi , “Looking at the Primary (Hi)story and the Prophetic Books as Literary/Theological Units within the Frame of the Early Second Temple: Some Considerations,” SJOT 12 (1998), 26–43 (see p. 26: “Primary Historical Narrative”); S. M andell , “Primary History as a Social Construct of a Privileged Class,” in Concepts of Class in Ancient Israel (ed. M. R. Sneed; Atlanta: Scholars Press, 1999), 21–35; A. A bela , “Is Genesis the Introduction of the Primary History?,” in Studies in the Book of Genesis: Literature, Redaction and History (ed. A. Wénin; BETL 155; Leuven: Leuven University Press, 2001), 397–406; A. G. A uld , “Counting Sheep, Sins and Sour Grapes: The Primacy of the Primary History?,” in Sense and Sensitivity: Essays on Reading the Bible in Memory of Robert Carroll (ed. A. Hunter and P. R. Davies; JSOTSup 348; Sheffield: Sheffield Academic Press, 2002), 63–72; D. N. F reedman and B. K elly , “Who Redacted the Primary History?,” in Sefer Moshe: The Moshe Weinfeld Jubilee Volume; Studies in the Bible and the Ancient Near East, Qumran, and Post-Biblical Judaism (ed. 10

782

Konrad Schmid

Enneateuch.14 At any rate, it is this very literary horizon that research on the Pentateuch needs to take into account as well when dealing with its alleged sources and redactions.15 There is a significant and important caesura between Deuteronomy and Joshua, but it is not an absolute one in terms of the compositional and redactional history of Genesis–Kings. The contributions in this part will therefore discuss and evaluate the question of possible extensions of literary sources within the Pentateuch into the realm of the Former Prophets.

C. Cohen et al.; Winona Lake, IN: Eisenbrauns, 2004), 39–47; J.-W. Wesselius , “The Functions of Lists in Primary History,” in “Basel und Bibel”: Collected Communications to the XVIIth Congress of the International Organization for the Study of the Old Testament, 2001 (ed. M. Augustin and H. M. Niemann; BEAT 51; Frankfurt: Lang, 2004), 83–89. 14  Cf., e.g., E. A urelius , Zukunft jenseits des Gerichts: Eine redaktionsgeschichtliche Studie zum Enneateuch (BZAW 319; Berlin: de Gruyter, 2003); the contributions in M. W itte et al. (eds.), Die deuteronomistischen Geschichtswerke: Redaktions- und religionsgeschichtliche Perspektiven zur “Deuteronomismus”-Diskussion in Tora und Vorderen Propheten (BZAW 365, Berlin: de Gruyter, 2006); H.-C. S chmitt , Theologie in Prophetie und Pentateuch: Gesammelte Schriften (BZAW 310, Berlin: de Gruyter, 2001), 255–276, 277–294, 295–308, 311–325; idem , “Das sogenannte jahwistische Privilegrecht in Ex 34,10–28 als Komposition der spätdeuteronomistischen Endredaktion des Pentateuch,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 157–171; idem , “Dtn 34 als Verbindungsstück zwischen Tetrateuch und Deuteronomistischen Geschichtswerk,” in Das Deuteronomium zwischen Pentateuch und Deuteronomistischem Geschichtswerk (ed. E. Otto and R. Achenbach; FRLANT 206; Göttingen: Vandenhoeck & Ruprecht, 2004), 180–192; idem , Arbeitsbuch zum Alten Testament: Grundzüge der Geschichte Israels und der alttestamentlichen Schriften (UTB 2146; Göttingen: Vandenhoeck & Ruprecht, 2005), 242–248; R. A chenbach , “Pentateuch, Hexateuch, und Enneateuch: Eine Verhältnisbestimmung,” ZABR 11 (2005), 122–154; T. B. D ozeman , “The Composition of Ex 32 within the Context of the Enneateuch,” in Auf dem Weg zur Endgestalt von Genesis bis II Regum: Festschrift Hans-Christoph Schmitt zum 65. Geburtstag (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 175–189; E. Blum , “Pentateuch – Hexateuch – Enneateuch? Oder: Woran erkennt man ein literarisches Werk in der hebräischen Bibel?” in Les dernières rédactions du Pentateuque, de lʼHexateuqe et de lʼEnnéateuque (ed. T. Römer and K. Schmid; BETL 203; Leuven: Peeters, 2007), 67–97; C. Levin , Re-Reading the Scriptures: Essays on the Literary History of the Old Testament (FAT 87; Tübingen: Mohr Siebeck, 2013), 115–141. 15  Cf. K. Schmid , “Buchtechnische und sachliche Prolegomena zur Enneateuchfrage,” in Auf dem Weg zur Endgestalt von Genesis bis II Regum: Festschrift Hans Christoph Schmitt (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gruyter, 2006), 1–14.

The Pentateuchal Sources and the Former Prophets A Neo-Documentarianʼs Perspective Baruch J. Schwartz Adherents of what has become known over the last few years as the Neo-Documentary Hypothesis regarding the composition of the Pentateuch1 have thus far confined their discussion to the study of the Pentateuch itself and have not yet begun to examine in detail the possible ramifications of the Documentary Hypothesis in its renewed and refined form for the study of the Former Prophets. In an initial attempt to begin to consider, cautiously and provisionally, some of the relevant issues, let us pose the following theoretical question: strictly from the vantage point of the modern version of the Documentary Hypothesis, is it possible that some, or all, of the four pentateuchal sources, as identified and understood by the Neo-Documentarians, continued their narratives into the periods following the death of Moses, and if so, is it possible that these continuations contributed to, and can still be detected in, Joshua, Judges, Samuel, and Kings?2 In other words, might J, E, P, or D continue into the Former Prophets? I begin with D. Following Menahem Haran,3 I define D as a collection of originally separate components arranged in a sequence that creates the illusion

 J. S. B aden , “The Re-Emergence of Source Criticism: The Neo-Documentary Hypothesis,” Bible and Interpretation (May 2012), http://www.bibleinterp.com/articles/ bad368008.shtml (accessed 5/15/14); idem , The Composition of the Pentateuch: Renewing the Documentary Hypothesis (New Haven, CT: Yale University Press, 2012); J. Stackert, A Prophet Like Moses: Prophecy, Law, and Israelite Religion (Oxford: Oxford University Press, 2014), 19–26. 2  For a convenient overview of the critical study of the Former Prophets, see R. W. Klein, 1 Samuel (WBC; 2nd ed.; New York: Bible Society International, 1983), xxviii–xxxii. A more recent survey may be found in B. Halpern and A. Lemaire, “The Composition of Kings,” in The Books of Kings: Sources, Composition, Historiography and Reception (ed. A. Lemaire and B. Halpern; VTSup 129; Leiden: Brill, 2010), 123–153. 3  M. Haran, The Biblical Collection: Its Consolidation to the End of the Second Temple Times and Changes of Form to the End of the Middle Ages (4 vols.; Jerusalem: Bialik Institute/ Magnes Press, 1996–2014) (Hebrew). Throughout this essay, I draw extensively on Haranʼs writings, not only because I find his approach both compelling and challenging but also because much of his work has never been translated into any Western language. 1

784

Baruch J. Schwartz

of continuity.4 These components are the writings of the literary school that acted in conjunction with those responsible for the cultic and legal measures undertaken in the reign of Josiah, and thus they are Deuteronomic. However, their collection into a single scroll or ‫ ֵס ֶפר‬is Deuteronomistic, since it was accomplished not in Josiahʼs time but rather after the destruction of 586, by ֵ at the later scribes who placed this collection, which they named ‫ּתֹורה‬ ָ ‫ס ֶפר ַה‬, head of their historical composition. There it served as the programmatic basis for everything to follow: the account of Israelʼs history we have before us in the books of the Former Prophets. Only much later, in postexilic Judah, when the Pentateuch was compiled, did D become one of the documents of the Pentateuch.5 The defining characteristic of the writings of the Deuteronomic authors is that they pertain exclusively to events surrounding the covenant that these scribes deemed to have been made in the land of Moab just prior to Mosesʼs death. The historical narrative that they contain is thus very brief. The Deuteronomic authors do of course make reference to events preceding the last days of Moses, both events that transpired in his lifetime and events that occurred long before. But all such references are incorporated in Mosesʼs own orations as recollections of the past, whereas the actual historical narrative of D begins by introducing Mosesʼs orations and ends with – or just before6 – his death. Since none of the separate literary components of D went beyond this point, D does not, in the strict sense, continue into the Former Prophets. However, since the compiled D was assembled in order to serve as the opening section of the Deuteronomistic History, one might say that D as a whole “continues” into DtrH as a whole. The Deuteronomistic History is inconceivable without the ‫ּתֹורה‬ ָ ‫ ֵס ֶפר ַה‬as its opening section, and in this sense D may be said to “continue” into the Former Prophets. This continuity, however, 4  It bears emphasizing that the D source is not identical to the canonical book of Deuteronomy. The latter (Deut 1:1–34:12) is one of the five ḥômāšīm into which the compiled Pentateuch was divided; the former, which concludes with Deut 32:45–47, is the “scroll of the Torah” that was created, in order to serve as the opening section of the Deuteronomistic History, by assembling and combining several originally independent scrolls, the contours of which correspond to (1) Deut 1:1–2, 4–4:44; (2) Deut 4:45–26:19; 28:1–69; (3) 27:1–26 (itself a composite); (4) Deut 31:9–13, 25–27, 29; 32:45–47. For full discussion, see Haran, Biblical Collection (see n. 3), 2:40–93, 195–219, 290–300; I diverge here from Haran on a few minor details. 5  Haran, Biblical Collection (see n. 3), 2:185–230, 290–299; B. J. Schwartz, “The Torah: Its Five Books and Four Documents,” in The Literature of the Hebrew Bible: Introductions and Studies (ed. Z. Talshir; 2 vols.; Ancient Literature of Eretz Israel and Its World 1; Jerusalem: Yad Ben-Zvi, 2011), 1:161–226; here 215–218 (Hebrew). 6  The question turns on whether any or all of the account of Mosesʼs death and the narratorʼs evaluation of his career (Deut 34:1–12) is to be assigned to D; for discussion, see Haran, Biblical Collection (see n. 3), 2:193, n. 8, 200, n. 19; more recently, P. Y. Y oo , “The Four Moses Death Accounts,” JBL 131 (2012), 423–441.

The Pentateuchal Sources and the Former Prophets

785

is one of redaction, not of authorship;7 insofar as the authentic Deuteronomic material is concerned, it concludes at the end of D. As for P, it is widely acknowledged that the more substantial narratives of Priestly character in the Former Prophets are essentially confined to the book of Joshua, in particular the account of the apportionment of the land among the tribes and the establishment of the tabernacle at Shiloh.8 It is further recognized that the few Priestly-looking passages that one encounters thereafter are brief, localized interpolations.9 As the latter are redactional in character, by definition they cannot be the continuation of P. But what of the narratives in Joshua? True, they do not display the uninterrupted flow and the completeness that characterize P in the Pentateuch, and no passage suggests itself as a particularly good candidate to serve as Pʼs original conclusion. But these passages do possess an impressive amount of internal continuity, and they presuppose, and relate exclusively to, the Priestly narrative in the Pentateuch.10 It is difficult to ignore the fact that developments and events to which P explicitly looks forward and which are of major importance to its program are not related in the Priestly source as preserved in the Pentateuch. This is especially perplexing in light of the highly formalized, periodized structure and consistency of theme and purpose that characterize the Priestly narrative. This would suggest that P did originally include an account of the land occupation and the arrival of the tabernacle at its destination. One might go so far as to say that even if there were no obviously Priestly texts relating these events to be found in the book of Joshua, we would still have good reason to presume that they once existed. Since we do have them, albeit not in their entirety, in Joshua, the inference that they were excerpted from what was originally the concluding portion of P is plausible.11  Cf. Haran, Biblical Collection (see n. 3), 2:27, n. 6.  Allowing for differences among critics on some of the particulars, there is general agreement that the Priestly narratives in Joshua that describe the distribution of the land and the designation of the Levitical cities and the cities of refuge as provided for by Moses according to P (Num 34–35) can be detected in Josh 13–22; for details, see Haran, Biblical Collection (see n. 3), 2:194–195, n. 11. For additional discussion, see A. Rofé, “Joshua Son of Nun in the History of Biblical Tradition,” Tarbiz 73 (2004), 333–364, esp. 343–345; J. Van Seters, “The Deuteronomist from Joshua to Samuel,” in Reconsidering Israel and Judah: Recent Studies on the Deuteronomistic History (ed. J. G. McConville and G. N. Knoppers; Winona Lake, IN: Eisenbrauns, 2000), 204–239; S. Boorer, “The Envisioning of the Land in the Priestly Material: Fulfilled Promise or Future Hope?,” in Pentateuch, Hexateuch, or Enneateuch: Identifying Literary Works in Genesis through Kings (ed. T. B. Dozeman et al.; Atlanta: Society of Biblical Literature, 2011), 99–125; see especially the summary and discussion of Lohfink (103, 105–113, 117–124). 9  See Haran, Biblical Collection (see n. 3), 2:195, n. 11, 298–299, n. 62. 10  For example, the command regarding the cities of refuge in Num 35:9–34 and the account of its implementation in Josh 20:7–9; Haran, Biblical Collection (see n. 3), 2:57–58. 11  The question of precisely how these Priestly excerpts found their way to the book of Joshua is a somewhat separate one. The conventional way of accounting for them is to 7 8

786

Baruch J. Schwartz

From the point of view of the modern documentarian, there is nothing surprising about this. We view the compilation of the Torah as a one-time event, the aim of which was to create, from existing sources, an amalgamated account of Israelʼs laws and the events leading up to their being commanded through Moses. The endpoint of the compiled Torah – Mosesʼs death – was determined by the compiler, to serve this purpose alone,12 but there is no reason to assume that any or all of the narrative sources of which the compiler made use ended at this point. The compiler of the Pentateuch was under no obligation to incorporate any more of each of his sources than suited his own ends. It is therefore just as reasonable to surmise that Pʼs narrative continued beyond the death of Moses as it is to acknowledge that D, and the JEPD composition as a whole, did not. The modern documentarians, unencumbered by the a priori assumption that J, E, and P were created in contact with one another, have come to appreciate how greatly they differ, having in common much less than was previously assumed. We therefore see no reason to presuppose that they were similar in scope and that they all covered precisely the same historical period. The Priestly source seems to have ended with the land occupation, D surely did not, and none of this has any bearing on J or E. With regard to the non-Priestly narratives that make up almost the entirety of the Former Prophets, it has become a commonplace that the Deuteronomistic writers, whether they be one or several, early or late, made use of a wide and very diverse selection of mostly unconnected literary materials and oral traditions, representing a broad spectrum of genre and type as well as date, origin, and provenance, and that they shaped them quite thoroughly to serve their own ends.13 On the basis of this consensus, documentarians of the last few generations too conceded that the materials used by the Deuteronomistic writers were not, as classical critics supposed, the continuations of the pentateuchal J postulate a “Priestly redaction” of DtrH (or some portion thereof) subsequent to its original composition, at which stage the Priestly redactors (who need to be posited for this purpose, since no other evidence exists of their existence) interpolated a few remnants of the earlier Priestly narrative; for this view, see M. N oth , Überlieferungsgeschichtliche Studien: Die sammelnden und bearbeiten Geschichtswerke im Alten Testament. (Tübingen: M. Niemeyer, 1957), 182–190. Although this possibility cannot be discounted, it is simpler, and therefore probably preferable, to suppose that the Deuteronomistic redactors themselves, when assembling the Deuteronomistic History in the first instance, incorporated in the historical account that they were compiling those portions of the Priestly narrative to which they had access and which were pertinent. The fact that these cases are confined to Joshua (see above) would be supporting evidence for the suggestion that the Priestly narrative did not extend beyond the time of Joshua. Haranʼs view is somewhat different; see Haran, Biblical Collection (see n. 3), 2:189, 298–299. 12  Haran, Biblical Collection (see n. 3), 2:210; Schwartz, “Torah” (see n. 5), 217. 13  For the succinct, definitive statement of this approach to DtrH, see N oth , Überlieferungsgeschichtliche Studien (see n. 11), 10–12.

The Pentateuchal Sources and the Former Prophets

787

or E. Of course this goes without saying for more recent critics who deny the existence of J and E altogether.14 Although a number of considerations certainly contribute to this still-widespread view, the one that would appear to outweigh all others is the impression that the non-Priestly narrative passages that populate the Former Prophets do not break down into identifiable literary sources at all, certainly not into two sources. If they cannot be separated and grouped into documentary sources, whether two or more, then they do not contain the continuations of any of the documents of the Pentateuch. It follows that neither J nor E continues into the Former Prophets. Yet there may be grounds for reexamining this. First, a methodological consideration should be raised. When scholars speak of the highly diverse character of the preexisting materials used to construct DtrH, the diversity to which they refer is one of genre. This recognition of a plurality of types of narrative is significant for many reasons, but it has no bearing on the question of literary source. A literary source may indeed contain more than one literary type. Each one of the pentateuchal sources as documentarians understand them (and not only today but in the past as well) contains several types of narratives; all or most of them also contain several types of legal material; two of them – J and E – contain more than one type of poetry;15 and all contain passages that might be formally classified as lists, as itineraries, as “brief notices,” and so forth. Variety of form may actually be regarded as a normal feature of the literary source.16 What is true about types is just as true of traditions. Early tradition critics fully acknowledged the aggregate of differing traditions underlying each of the pentateuchal sources, as do documentarians today.17 One might go so far as to say that the diversity of traditions in each source is what makes it a source. In any case, it certainly does not disqualify it from being one. The tendency to equate form-critical and/or tradition-critical distinctions with literary-critical ones is, to my mind, one of the most serious methodological flaws in the contemporary non-documentary approach to the Pentateuch.18 And  See Halpern and Lemaire, “Composition of Kings” (see n. 2), 125–140.  Jʼs poetic passages include such diverse genres as oracles (e.g., Gen 49:1b–27), hymns (e.g., Exod 15:1–19a), and blessings (e.g., Num 23:7–10, 18–24; 24:3–9, 15–24; Deut 33:1– 29); E contains fewer poetic passages, yet the generic diversity is evident in the difference between, e.g., the taunting Song of the Bards (Num 21:27–30) and Mosesʼ farewell song (Deut 32:1–43). Of course I make no claim that the poetic passages were authored by the authors of the narratives, only that the literary sources, J and E, respectively, incorporated more than one poetic genre. 16  Haran, Biblical Collection (see n. 3), 2:218–219. 17  See G. N. K noppers , introduction to Reconsidering Israel and Judah: Recent Studies on the Deuteronomistic History (ed. J. G. McConville and G. N. Knoppers; Winona Lake, IN: Eisenbrauns, 2000), 1–18, here 15–16. 18  See Baden, Composition (see n. 1), 45–67. 14 15

788

Baruch J. Schwartz

it is no less problematic in this approach to the Former Prophets, where it may actually have originated. The features of one text that identify it as belonging to the same source as another are the literary ones: internal references and overall continuity of narration, narratological and historical assumptions, indications of geographical and historical provenance, and, of course, commonalities of language, diction, and style. Keeping this in mind, it is hard to see how the typological diversity from Joshua to the end of Kings can shed any light on the question of literary sources. Another one of the underpinnings of what seems to have become the dominant view of DtrH is the extent of its Deuteronomistic character. The unique and identifiably Deuteronomistic theology underlying DtrHʼs presentation of Israelite and Judean history has come to be thought of as all-pervasive. The tendency is to interpret every passage, every section, every stratum, and every internal division of the work as somehow expressive of one or more Deuteronomistic ideas or messages. Of course, if all, or almost all, of the Former Prophets is thoroughly Deuteronomistic, none of it can be the continuation of the non-Deuteronomic sources of the Pentateuch. Here, too, there are perhaps grounds for reconsideration. The tangible, objective evidence for the Deuteronomistic character of the historical work extending from Joshua through Kings, in addition to the very fact of its existence, is the presence of Deuteronomistic passages – words, phrases, verses, and larger sections – in the text itself. These are recognizable – even if some details remain in dispute – not only by style and content but also by the fact that they are very obviously redactional; their secondary, interpolated character speaks for itself. This being the case, what remains after the Deuteronomistic redaction is peeled away is identifiable as well. Much of the remainder resembles one or the other of what documentarians view as the non-Deuteronomic sources in the Pentateuch, namely, J and E: in its narratological assumptions and geographical orientations, in the explicitly and unmistakably non-Deuteronomic beliefs expressed, in its language and style and, to some extent, in the historical background reflected.19 When the painstaking but indispensable task of distinguishing source from redaction in DtrH is undertaken with care, the Deuteronomistic character of the whole that seemed to preclude the very possibility that J and E might continue into the Former Prophets turns out not to be quite so ubiquitous after all. We come next to connectivity and continuity. Even after the Deuteronomistic interpolations and revisions are identified and disregarded, one cannot easily detect two recognizably continuous narratives intertwined from Joshua to the end of Kings. And yet, examples of narrative connectivity, and at least partial continuity, within the source material are not totally absent either. One has  For example, the prophetic stories (most of them included in 1 Kgs 17–2 Kgs 10). See M. Haran, “Problems in the Composition of the Former Prophets,” Tarbiz 37 (1967), 1–14, esp. 4 (Hebrew). 19

The Pentateuchal Sources and the Former Prophets

789

only to think about the Ark Narrative, its interspersed pieces obviously connected;20 the Succession Narrative (however one defines its precise parameters), not only connected but highly continuous;21 or the Ephraimite (or Efrathite) History identified by Alexander Rofé.22 In all of these cases, the links and the  See P. D. Miller and J. J. M. Roberts, The Hand of the Lord: A Reassessment of the “Ark Narrative” of 1 Samuel (Baltimore: Johns Hopkins University Press, 1977). 21  L. R ost , The Succession to the Throne of David (trans. M. D. Rutter and D. Gunn; Sheffield: Almond Press, 1982). 22  A. Rofé, “Ephraimite versus Deuteronomistic History,” in Storia e Tradizioni di lsraele: Scritti in onore di J. A. Soggin (ed. D. Garrone and F. Israel; Brescia: Paideia, 1991), 221–235 (reprinted in Reconsidering Israel and Judah: Recent Studies on the Deuteronomistic History [ed. J. G. McConville and G. N. Knoppers; Winona Lake, IN: Eisenbrauns, 2000], 462–474). As Rofé relates, C. F. Burney asserted long ago that Josh 24–1 Sam 12 was written by a historiographical school that he named “Late-Elohistic,” “Ephraimite,” or “E2.” Burney characterized this text as (1) prophetic in orientation and (2) northern Israelite in provenance, (3) viewing the Northern Kingdom as “naturally the principal repository of the Joshua tradition.” See C. F. Burney, The Book of Judges (2nd ed.; London: Rivingtons, 1920), 507, for the full list of passages attributed to the “Ephraimite document.” Rofé demonstrated that the Ephraimite History is also (4) antimonarchist and (5) views Israel as passive in all of its military encounters, depicting YHWH as intervening to deliver his people. He also defined (6) the distinct role of the prophet in the Ephraimite History as that of castigating the people for past transgressions or presenting them with an alternative and urging them to choose the right path. Finally, Rofé showed that the Ephraimite History (7) incorporates its dogmatic statements in the context of dialogues between the leader and the people or YHWH. From the Neo-Documentarian point of view, all but one of the characteristics of the Ephraimite History identified by Rofé are also true of E, which is a (1) prophetic, (2) northern-oriented source and which (3) makes prominent mention of the figure of Joshua alongside Moses at every significant moment. E describes the war against Amalek (Exod 17:8–13) as (5) won miraculously by YHWH himself, and it is in E that we find the idea of the ‫( ִצ ְר ָעה‬traditionally: the “hornet”) unleashed by YHWH in order to decimate the inhabitants of Canaan in advance of the Israelitesʼ arrival (the ‫ ִצ ְר ָעה‬appears in E in Exod 23:28; then in D, drawing here as elsewhere upon E, in Deut: 7:20; and finally in Josh 24:28; see also B. J. S chwartz , “Reexamining the Fate of the ‘Canaanites’ in the Torah Traditions,” in Sefer Moshe: The Moshe Weinfeld Jubilee Volume [ed. in C. Cohen et al., Winona Lake, IN: Eisenbrauns, 2004], 151–170, esp. 158). As but one example of the (6) characteristically Elohistic view of the prophetic role we may cite Eʼs depiction of Mosesʼs words of farewell and the accompanying song of reproof (Deut 31:1–8, 14–24 [reading ‫ ַה ִּׁש ָירה‬for MT ‫ּתֹורה‬ ָ ‫ַה‬ in 24], 25*, 28, 30; 32:1–33). For a detailed discussion, and challenging reevaluation, of Eʼs prophetic orientation, see Stackert, Prophet Like Moses (see n. 1), 70–125. Finally, Eʼs account of Aaron and Miriamʼs challenge to Moses and of YHWHʼs response to it (Num 12:1–8) may serve to illustrate the way in which (7) Eʼs programmatic and dogmatic statements are made through dialogue between YHWH and the protagonists. The only characteristic of the Ephraimite History lacking in E is the antimonarchist tendency, and this may simply be a result of the fact that Eʼs pentateuchal narrative did not have occasion to make reference to the topic of kingship. For further general discussion, see A. W. Jenks, The Elohist and North Israelite Traditions (Missoula, MT: Scholars Press, 1977). 20

790

Baruch J. Schwartz

signs of original continuity are present in the source material, not in the Deuteronomistic redaction. To adduce some of the further examples mentioned by Haran, the curse of Jericho in Josh 6:26 and the notice of how it was carried out in 1 Kgs 16:34 are mutually aware of and point to each other, but they are not Deuteronomistic, nor do they appear to be redactional at all. The same goes for of the tale of the man of God who foretold the demise of the house of Eli in 1 Sam 2:27–36 and the account of the expulsion of Abiathar in 1 Kgs 2:27. The tale of the man of God who came to Bethel in 1 Kgs 13 and the non-Deuteronomistic verses in 2 Kgs 23 (vv. 8 and 16–20) are similarly mutually referential, and they do not belong to any redactional stratum either.23 Continuity on a more localized level is detectable in each of the two tales of Saulʼs rejection and Davidʼs election,24 where the Deuteronomistic redaction, if there is any at all, is very sparse and yet the essentially continuous contours of each of the two partially parallel, intertwined stories is recognizable in the unredacted material, that is, the sources. These examples should at least give us pause to question whether the narratives in Joshua through Kings are quite as unconnected and discontinuous as is generally thought. Moreover, the internal continuity of individual narrative passages themselves, narratives neither fragmentary nor particularly brief, at least suggests that for all the typological diversity, this sort of writing, namely, fully developed, extensive prose narratives, may indeed be the distinctive, unifying characteristic of the sources of DtrH no less than of the sources of the Pentateuch. It must be admitted that to recognize more connection and continuity than has typically been acknowledged does not necessarily imply the existence of only two non-Priestly, non-Deuteronomistic narrative sources in DtrH. But this possibility should not be dismissed out of hand either. In the critical study of the Pentateuch, the first intimation of separate narrative sources was the discovery of competing accounts of the same event, either woven together or presented sequentially. These are certainly much rarer in the Former Prophets than in the Torah, but some do exist, and it would be a fatal error to allow the numerous cases of source-plus-redaction (i.e., source-plus-Dtr) to obscure from view the admittedly rarer cases of source-plus-source. I have already mentioned the Saul-David narrative; one should also note Simeon Chavelʼs fascinating study of 2 Sam 21.25 And whenever there are competing non-Priestly and non-Deuter23  H aran , Biblical Collection (see n. 3), 2:211–212; see also 138, n. 57, 188–189, 190, n. 6, 197. 24  The accounts in 1 Sam 13 and 15 are competing versions of how and why Saul was deposed as king. The story of YHWHʼs directing Samuel to Jesse the Bethlehemite in order to anoint David and of Davidʼs subsequent rise to popularity in Saulʼs court (1 Sam 16) is the sequel to the latter and is unconnected to the former. 25  S. Chavel, “Compositry and Creativity in 2 Samuel 21:1–14,” JBL 122 (2003), 23–52.

The Pentateuchal Sources and the Former Prophets

791

onomistic accounts of the same event, whether sequential or interwoven, their number, to the best of my knowledge, is usually two. So the possibility that two non-Priestly sources served as the basis for the Deuteronomistic composition is at least deserving of another look.26 Entertaining this possibility would necessitate keeping in mind at least two differences between DtrH and the Torah. The first relates to the nature of the sources. The narrative documents of the Torah, for all the differences between them, are still essentially parallel versions of the “same” story. They therefore contain differing accounts of the “same” episode fairly often. Furthermore, since there are three such sources, J, E, and P, intersections occur whenever any two of the three have an episode in common, thus increasing the frequency of such occurrences. In contrast, if one were to postulate two narrative sources relating the Israelʼs history in Canaan, one northern in provenance and one southern, one should not really expect them to contain a large number of parallel accounts. Presumably each one would have told the tale of the region, later the kingdom, where it originated. Interwoven, they would often look like two vastly different histories that alternate only on a chronological basis, sharing very little content in common. This would perhaps be rather unlike the Pentateuch, but it would not be too unlike the synchronic history in Kings. Paucity of common content does not make the existence of sources any less likely. The second difference relates to the nature of the redaction. Neo-Documentarians are convinced that the redaction of the Torah was highly conservative, the redactor intervening as little as possible and quite committed to keeping the sources intact. Yet we have no reason to imagine that the Deuteronomistic redaction of the Former Prophets, which took place a century or more before the redaction of the Torah, in a different context and for different reasons, conformed to this hands-off policy as well. The Deuteronomistic redaction is indeed more active, adding to and touching up the sources, most notably in the quoted speech of key figures but also in the narration of events. It is at least imaginable, perhaps even likely, that the Deuteronomistic redaction was similarly freer than the pentateuchal redaction with regard to omitting portions of its sources. As a possible way of accounting for gaps in the sources, this is no less likely than the assertion that more-or-less continuous narratives never existed.27 If indeed it could be shown that the Deuteronomistic redactors were working with two sources, what would be required in order to view them as the contin-

 Even if such a suggestion is not outside of the realm of theoretical possibility, it of course pertains only to the penultimate stage – the literary sources as employed by Dtr – and not to the literary history of each of the sources themselves, which one can, and should, examine on its own. 27  See H aran , “Problems in the Composition of the Former Prophets” (see n. 19), 3–5; idem , Biblical Collection (see n. 3), esp. 2: 206–220. 26

792

Baruch J. Schwartz

uations of J and E?28 First and most obviously, one would be literarily similar to, and have literary connections with, J in the Torah, and the other with E – as the two are identified by the Neo-Documentarians. We would also expect to find J and E in the Torah anticipating at least some of what comes after Mosesʼs death and to find at least some of what each one anticipated either fulfilled or frustrated in what follows in the same source – much as we do find in P. And we would expect to find in each source in the Former Prophets indications that the J and E narratives in the Torah, respectively, are presupposed or at least not ignored – this last being perhaps the most important and the most difficult. Classical critics made some significant strides in this direction.29 For the present, I am provisionally suggesting that there is no compelling reason for abandoning this line of inquiry.30 Second, we would need to determine precisely how much of the Deuteronomistic compilation relied on, and incorporated, J and E. Did the Deuteronomistic redactor use them exclusively, or did he make use of other sources as well? Third, we would need a historical theory. We would have to explain how it can be that J and E in the Pentateuch seem to reflect mid-monarchic times, if their narratives somehow reach down to the seventh or even early sixth century. Would we reevaluate our impression of the historical period reflected in J and E in the Torah? Would we instead adopt Haranʼs redefinition of the “sources” as the products of literary schools,31 so that each one is not an author, or even a group of authors, at all but rather a sequence of scribes belonging to a single guild that endured for a few generations? Would we perhaps find a cut-off point for J and E in the Former Prophets, presumably at some point in the book of Kings? Or would some other alternative suggest itself?  Noth found no literary affinity between the pentateuchal sources and the various materials he believed to have been employed by the Deuteronomist in Joshua. And yet, although he was of the opinion that the continuation of the pentateuchal sources into the Former Prophets was both unsubstantiated and improbable, he nevertheless felt obliged to admit that it cannot be excluded altogether. See Noth, Überlieferungsgeschichtliche Studien (see n. 11), 180–182, 211. 29  See Cf. Haran, Biblical Collection (see n. 3), 2:27, n. 6. To be sure, one shortcoming of classical criticsʼ search for the pentateuchal sources in the Former Prophets was their expectation that the sources, once recovered, would look as complete and continuous as they do in the Pentateuch. The absence of such clear continuity is one of the factors that led eventually to the conclusion that the sources did not exist. The realization that the two redactions differ from each other effectively removes this obstacle from the attempt to recover the source material. 30  It is of course possible that one, but not both, of the non-Priestly, non-Deuteronomic pentateuchal sources identified by the documentarians continues into DtrH. For a comprehensive presentation of the theory that J can be traced throughout the Pentateuch and Former Prophets, see R. E. Friedman, The Hidden Book in the Bible: The Discovery of the First Prose Masterpiece (San Francisco: HarperCollins, 1998). 31  Haran, Biblical Collection (see n. 3), 2:27, n. 6; idem, Temples and Temple-Service in Ancient Israel (Oxford: Clarendon, 1978), 146–147. 28

The Pentateuchal Sources and the Former Prophets

793

I do not claim to have answers to these questions, nor do I claim to have demonstrated the presence of J and E in the Deuteronomistic composition or even to have reached a conclusion about it myself. I wish only to suggest that considering it does not necessitate rejecting the well-established scholarly consensus regarding the existence, scope, structure, aims, and date (or dates) of DtrH. If it does turn out to be a viable theory, it will actually enable us to retain all of the various conceptions of Genesis–Kings that have been proposed in scholarship, each in its own appropriate context. If, as Haran suggests, J and E extend throughout DtrH,32 the suggestion of an Enneateuch will have been borne out, but only with regard to J and E. The notion of Hexateuch too will be retained, but only with regard to P. It will still be possible to speak of a Pentateuch, but only when speaking of the composite JEPD work and not of any of its four sources. Even the Tetrateuch will remain what it has in fact been all along: a term for what the Torah would have looked like if the redactor had not inserted D. With all due caution, therefore, I would advocate returning the careful consideration of this possibility to the scholarly agenda for the next stage of documentarian research.33

 Haran, Biblical Collection (see n. 3), 2:185–242.  Sincere thanks to Guy Darshan for his assistance in the preparation of this study, to Dror Jacob and David Knoll for their expert help in preparing it for publication, and to the Israel Science Foundation for its continued generous support of my pentateuchal research. This essay is the result of research conducted during my stay as a fellow at the Israel Institute for Advanced Study in Jerusalem (2012–2013). It is a pleasure to extend my sincere thanks to the institute and its staff and to the esteemed group of fellows with whom I was privileged to spend this time. 32 33

Do the Pentateuchal Sources Extend into the Former Prophets? Joshua 1 and the Relation of the Former Prophets to the Pentateuch Cynthia Edenburg Do the pentateuchal sources extend into the former prophets? This question presumes, of course, that there is such a thing as pentateuchal sources. I do not question the value of this premise, since it provides the basis for all critical inquiry into the composition of the Pentateuch. But what constitutes a pentateuchal source? For the sake of my argument here, I will take the narrative strands of the Documentary Hypothesis as a convenient point of departure. Toward the end, I will return to this question in order to show how my findings might relate to the question of what constitutes a pentateuchal source.

1 Premises of the Hexateuchal Thesis and Methodological Considerations The idea that the Former Prophets continue the literary strands of the Pentateuch is considerably older than the so-called New Documentary Hypothesis. By the seventeenth century, Benedict Spinoza had already marked the narrative and thematic continuity running from Genesis to the end of Kings and thus viewed the Pentateuch along with the Former Prophets as a continuous historical work composed or compiled by a single author on the basis of diverse sources.1 In the first quarter of the nineteenth century, Friedrich Bleek argued that Joshua is the natural continuation of the history that begins in Genesis, since Genesis anticipates fulfillment of the promise of the land to Abrahamʼs descendants.2  See B. de Spinoza, Theological-Political Treatise (trans. M. Silverthorne and J. Israel; Cambridge: Cambridge University Press, 2007), 120–129. 2  F. Bleek, “Einige aphoristische Beiträge zu den Untersuchungen über den Pentateuch,” in Biblisch-exegetisches Repertorium oder die neuesten Fortschritte in der Erklärung der heiligen Schrift (ed. E. F. K. Kosenmüller and G. H. Kosenmüller; 2 vols.; Leipzig: Baumgärtner, 1822), 1:1–79, here 44–48. 1

796

Cynthia Edenburg

The end of the Pentateuch might mark an epoch in the history, but it does not supply closure. This premise persists today among many scholars and provides the basis for Hexateuch and Enneateuch models and even for the idea of a Deuteronomistic History. The Deuteronomistic account of the conquest not only continues the narration provided by Deuteronomyʼs framework and illustrates how the Deuteronomic war prescriptions were carried out, but it also purportedly supplanted the earlier versions of J and E regarding the possession of the land.3 Earlier scholars not only were certain that J and E must have continued to relate to possession of the land, they also thought they could identify threads of these sources in Joshua.4 Today most agree that JE (or J and E) cannot be uncovered in Joshua by peeling away the sections displaying Deuteronomistic and Priestly influence.5 Even so, many continue to adhere to the thesis that the non-Priestly story in Exodus and Numbers dealing with the exodus and the promise of the land must have continued in a story relating the conquest of the land and its possession as represented by a postulated pre-Deuteronomistic layer in Joshua.6 This thesis is based on further premises. Firstly, given the prevalence of the promise and fulfillment theme, a narrative that stops short of fulfillment is unsatisfactory, if not unthinkable. Secondly, that narrative overlapping indicates 3  See, e.g., M. Noth, The Chroniclerʼs History (trans. H. G. M. Williamson; JSOTSup 50; Sheffield: JSOT Press, 1987), 139–141, and Rendtorffʼs critique in R. R endtorff , The Problem of the Process of Transmission in the Pentateuch (trans. J. J. Scullion; JSOTSup 89; Sheffield: Sheffield Academic, 1990), 39–42. 4  See, e.g., S. R. Driver, An Introduction to the Literature of the Old Testament (Gloucester, MA: P. Smith, 1972), 104–116; J. W ellhausen , Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (3rd ed.; Berlin: de Gruyter, 1963), 61, 118–134; G. A. C ooke , The Book of Joshua (Cambridge Bible for Schools and Colleges; Cambridge: Cambridge University Press, 1918), xiv–xix; O. Eissfeldt, The Old Testament: An Introduction (New York: Harper & Row, 1976), 247–255, as well as surveys of earlier literature by J. Nentel, Trägerschaft und Intentionen des deuteronomistischen Geschichtswerks: Untersuchungen zu den Reflexionsreden Jos 1;23;24, 1 Sam 12 und 1 Kön 8 (BZAW 297; Berlin: de Gruyter, 2000), 13, and J. J. Krause, Exodus und Eisodus: Komposition und Theologie von Josua 1–5 (VTSup 161; Leiden: Brill, 2014), 71–72, with further references there. 5  See Noth, Chroniclerʼs History (see n. 3), 174–175, and T. B. Dozeman, Joshua 1–12: A New Translation with Introduction and Commentary (AB; New Haven, CT: Yale University Press, 2015), 6–8. A somewhat different approach is proposed by Van Seters. Though he agrees that pentateuchal sources are absent from Joshua, he argues that the Yahwist and the Priestly writer viewed Joshua as a necessary component of the promise-fulfillment scheme and composed their work as an extended prologue to the Deuteronomistic History; see, e.g., J. Van Seters, In Search of History (New Haven, CT: Yale University Press, 1983), 322–323. 6  See, e.g., R. G. K ratz , The Composition of the Narrative Books of the Old Testament (trans. J. Bowden; London: T&T Clark, 2005), 303–306; C. Nihan, “The Literary Relationship between Deuteronomy and Joshua: A Reassessment,” in Deuteronomy in the Pentateuch, Hexateuch, and the Deuteronomistic History (ed. R. Person and K. Schmid; FAT 2/56; Tübingen: Mohr Siebeck, 2012), 79–114, here 108–109.

Do the Pentateuchal Sources Extend into the Former Prophets?

797

a primary continuity, such as camping at Shittim in Num 25:1 and sending spies out from Shittim in Josh 2:1.7 But do narrative and thematic continuity provide an adequate basis for concluding that the books of the Former Prophets share or are based upon the same sources found in the Pentateuch? Narrative and thematic continuity are just as likely the product of a late or final stage of editing. Thematic continuity can be imposed upon texts as they are brought together, in order to mask their original lack of thematic correspondence. In my opinion, the pursuit of pentateuchal sources in Joshua or other books of the Former Prophets is on firmer ground when based upon linguistic and stylistic considerations. For example, both D and P employ distinctive language that is related to distinctive ideology. Here, too, it is necessary to state reservations. For example, Deuteronomistic idiom is long lived, and is found in works such as Ezekiel, Nehemiah, Chronicles, and different compositions from Qumran (e.g., the Temple Scroll and Dibrei Moshe), but the imprint of Deuteronomistic language upon these works does not provide a basis for attributing them to a common source or redaction or even a common circle of scribes. More important is the question of how a particular expression is used. Is it employed in stereotyped fashion and does it convey the ideology and concepts we associate with the source, D or P? Or does its use in another work deviate from the conventions of the source? If so, then there are good grounds to suspect that it is being employed in a derivative fashion, only partly influenced by the previous source. In my opinion, this is all the more true when we find a mélange of idioms reflecting the influence of different sources. A further complication derives from redaction-critical models that demonstrate the editorial complexity of Deuteronomy and the Priestly literature. Thus, if we find echoes of D or P in the Former Prophets, are they evoking the pentateuchal sources or one of their redactional layers?8

7  See, e.g., K ratz , Composition (see n. 6), 217, and, for critiques, E. B lum , “Pentateuch-Hexateuch-Enneateuch? Or: How Can One Recognize a Literary Work in the Hebrew Bible?,” in Pentateuch, Hexateuch, or Enneateuch? Identifying Literary Works in Genesis through Kings (ed. T. B. Dozeman, et al.; AIL 8; Atlanta: Society of Biblical Literature, 2011), 43–71, here 54–57; D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011), 112–114. Compare C. Levin, “On the Cohesion and Separation of Books within the Enneateuch,” in Pentateuch, Hexateuch, or Enneateuch? Identifying Literary Works in Genesis through Kings (ed. T. B. Dozeman et al.; AIL 8; Atlanta: Society of Biblical Literature, 2011), 127–154, here 150–151. 8  For similar caveats see Nihan, “Literary Relationship” (see n. 6), 87.

798

Cynthia Edenburg

2 Joshua 1 in Relation to Pentateuchal Sources If indeed there was an original narrative continuity in which the story of the exodus and desert sojourn concluded with the conquest under the leadership of Joshua, we then should expect to find a key link at the beginning of the book of Joshua. After all, the Pentateuch in its final form prepares us for the role that Joshua will play after Mosesʼs death. Throughout the Pentateuch, Joshua is second only to Eleazar in the number of mentions by name among the characters who remain alive after Mosesʼs demise.9 Whereas Eleazar is defined by his priestly role, Joshua is Mosesʼs servant (‫משרת משה‬, Num 11:28; ‫משרתו‬, Exod 24:13; 33:11), who leads the battle against Amalek in Exod 17:8–13 and who is designated as Mosesʼs successor not only in Deuteronomy (1:38; 3:21, 28; 31:2–8, 14, 23; 34:9) but also according to Num 27:18–23.10 Admittedly, many recent scholars would now attribute much of this pentateuchal material to scribes working much later than the period of the initial composition of the conquest narrative.11 However, this need not trouble us at this point, since the question under investigation here is whether pentateuchal sources as understood by the Documentary Hypothesis find a continuation in the Former Prophets. Martin Noth famously rejected the idea that pentateuchal sources continued into Joshua and the rest of the Former Prophets but viewed the larger part of the Former Prophets as a comprehensive historical work that was composed along with the narrative frame of Deuteronomy. For Noth, Josh 1 was one in a series of speeches that were placed at turning points in the historical narrative and served to impose structure upon the whole from Deuteronomy to Kings. Thus, according to Nothʼs thesis, Josh 1 plays a key role in presenting the program not only for crossing the Jordan and possessing the land but also for the transition from the period of Mosesʼs leadership to Joshuaʼs.12

 Joshua is mentioned by name twenty-seven times (seven in Exodus, eleven in Numbers, nine in Deuteronomy); Eleazar thirty times (two in Exodus, two in Leviticus, twenty-five in Numbers, one in Deuteronomy); Caleb thirteen times (two in Exodus, nine in Numbers, two in Deuteronomy); and Phinehas four times (one in Exodus, three in Numbers). 10  See T. B. Dozeman, “Joshua in the Book of Joshua,” in Raising Up a Faithful Exegete: Essays in Honor of Richard D. Nelson (ed. K. L. Noll and B. Schramm; Winona Lake, IN: Eisenbrauns, 2010), 103–116, here 109–110. 11  To illustrate recent trends in dating non-Priestly materials in Exodus and Numbers, see, e.g., T. B. Dozeman, Exodus (Grand Rapids, MI: Eerdmans, 2009), 356–357; T. Römer, “Israelʼs Sojourn in the Wilderness and the Construction of the Book of Numbers,” in Reflection and Refraction, Studies in Biblical Historiography in Honour of A. Graeme Auld (VTSup 113; ed. R. Rezetko et al.; Leiden: Brill, 2007), 419–445, here 442–443. 12  M. N oth , The Deuteronomistic History (JSOTSup 15; 2nd ed.; Sheffield: Sheffield Academic, 1991), 18–20. 9

Do the Pentateuchal Sources Extend into the Former Prophets?

799

In order to test the validity of the theses with regard to Josh 1 – both the documentary Hexateuch thesis and the Deuteronomistic thesis – I will examine the phraseology of the chapter in order to see whether it reflects the style, typical usage, and ideology of one of the pentateuchal sources or whether it is likely to represent derivative use. Analysis of the intertextual relations between Josh 1 and different pentateuchal texts can also help determine whether they indicate source continuity or later reuse, interreference or echoing. On this score, it is crucial to bear in mind that unique or striking parallel formulations or other intertextual references do not necessarily derive from the same compositional or editorial layer. In fact, authors frequently emulate previous texts in order to evoke their authority, and formulations picked up from an earlier text may be echoed in order to impose a semblance of unity and continuity and thus mask disjunction. As Erhard Blum succinctly put it, “Not every literary connection, nor every parallel or cross-reference, no matter how uncontroversial, is evidence for a compositional connection within a single work.”13 Accordingly, it is important to evaluate the possible diachronic dimension of intertextual references before drawing conclusions regarding compositional context and purpose.

3 Literary-Critical, Stylistic, and Intertextual Analysis of Joshua 1 The chapter opens with a brief exposition in v. 1, which places the events “after the death of Moses.” This notice presumably refers back to the report of Mosesʼs death in Deut 34:5, although anyone familiar with the tradition of the legendary figure of Moses would be able to place the event as sometime after he had died. The rest of the chapter falls into three major parts: a lengthy divine speech addressed to Joshua (vv. 2–9), Joshuaʼs instructions to the officers to prepare for crossing the Jordan (vv. 10–11), and a lengthy passage dealing with the tribes that received their allotments in the Transjordan (vv. 12–18, comprising two speech acts: Joshua in vv. 12–15 and the tribesʼ answer in vv. 16–18). To be sure, there is a certain logic to the text as it stands. A new leader is required after Mosesʼs death, so YHWH designates Joshua to lead the people into the land and spurs him to action with words of encouragement designed to fortify him for the task ahead. The divine pep talk is effective since, in lieu of reply, Joshua immediately sets to work and issues his first orders, which are directed to the officers in charge of readying and mobilizing the people for crossing the Jordan. The preparation for crossing the Jordan prompts Joshua to 13

 Blum, “Pentateuch-Hexateuch-Enneateuch?” (see n. 7), 48.

800

Cynthia Edenburg

remind the east Jordan tribes of their commitment to help their brethren take possession of the land west of the Jordan and only afterward to return and take up settlement in the Transjordan. The eastern tribes then reassure Joshua that they recognize his authority as Mosesʼs successor and that any who defy his orders will be put to death. Thus the chapter as a whole follows a linear progression in thought. At the same time, it also closes on a note that echoes its opening. The divine designation of Joshua to succeed Moses and YHWHʼs assurance that he will be with Joshua just as he was with Moses (v. 5) are complemented by the eastern tribesʼ response: “We will obey you just as we obeyed Moses, just let YHWH be with you as he was with Moses” (v. 17). However, a closer examination of the chapter reveals narrative breaks, gaps, repetitions, and uneven use of language, all of which are telltale signs that betray literary intervention by later scribes. Occasionally, text-critical evidence further supports literary-critical analyses. Thus, the basic inner logic of the whole might not indicate compositional unity but rather the interests of successive scribes. These later scribes sought to reshape the earlier material according to their logic, but slips and irregularities in the finer details betray their presence behind the final form of the work. YHWHʼs speech in vv. 2–9 is particularly marked by repetition and change in perspective regarding topic and addressees, as well as second- and third-person variation. Since YHWH is addressing Joshua, we should expect the second-person singular to refer to Joshua and the third-person plural to refer to people, as in v. 2: ‫“( אתה וכל העם הזה אל הארץ אשר אנכי נתן להם‬you and all this people to the land which I give to them”). Verses 3–4, however, suddenly switch to the second-person plural when speaking of the promise of land (v. 3: ‫רגלכם‬, ‫;לכם‬ v. 4: ‫)גבולכם‬. The second-person plural is out of place here,14 but it is at home in a nearly identical section in Deut 11:24–25: “Every place where your (pl.) feet step will be yours (pl.), your (pl.) border will extend from the wilderness to the Lebanon and from the river, the Euphrates, to the last sea. No one will stand up to you (pl.).”15 Deuteronomy 11:24–25 ‫כל המקום אשר תדרך כף רגלכם בו לכם יהיה מן המדבר והלבנון מן הנהר נהר פרת ועד הים‬ ‫האחרון יהיה גבלכם‬ ‫לא יתיצב איש בפניכם‬

 See also Nentel, Trägerschaft (see n. 4), 15, 22–23; Dozeman, “Joshua in the Book of Joshua” (see n. 10), 107–108, with further literature there. 15  On the intertextual relations between Josh 1:1–9 and Deuteronomy, see, e.g., N. Lohfink, Theology of the Pentateuch: Themes of the Priestly Narrative and Deuteronomy (Minneapolis: Fortress, 1994), 241; J. R. P orter , “The Succession of Joshua,” in Proclamation and Presence: Old Testament Essays in Honour of Gwynne Henton Davies (ed. J. I. Durham and J. R. Porter; London: SCM, 1970), 102–131, here 103–105; R. D. N elson , Joshua (OTL; Louisville: Westminster John Knox, 1997), 33. 14

Do the Pentateuchal Sources Extend into the Former Prophets?

801

Joshua 1:3–5 ‫כל מקום אשר תדרך כף רגלכם בו לכם נתתיו כאשר דברתי אל משה מהמדבר והלבנון הזה ועד‬ ‫הנהר הגדול נהר פרת כל ארץ החתים ועד הים הגדול מבוא השמש יהיה גבולכם‬ ‫לא יתיצב איש לפניך כל ימי חייך כאשר הייתי עם משה אהיה עמך‬

This section of Josh 1 also interacts with another text in Deut 1:7–8: “Get up and journey and come to [. . .] the land of the Canaanites and the Lebanon up to the great river, the Euphrates. Here, I give before you the land that YHWH promised to your fathers, to Abraham, Isaac and Jacob, to give to them” (Deut 1:7–8; italics indicates shared formulae). This time, the parallel extends to Josh 1:6. Deuteronomy 1:7–8

‫] ארץ הכנעני והלבנון עד הנהר הגדל נהר פרת‬. . .[ ‫פנו וסעו לכם ובאו‬ ‫ראה נתתי לפניכם את הארץ אשר נשבע יהוה לאבתיכם לאברהם ליצחק וליעקב לתת להם‬

Joshua 1:4–6 ‫מהמדבר והלבנון הזה ועד הנהר הגדול נהר פרת כל ארץ החתים ועד הים הגדול מבוא השמש‬ ]. . .[ ‫יהיה גבולכם‬ ‫אתה תנחיל את העם הזה את הארץ אשר נשבעתי לאבותם לתת להם‬

A third intertext comes into play in Josh 1:5–6, which shares formulations with Deut 31:6–8, 23, in which Moses appoints Joshua as his successor: “YHWH your God goes before you, he will not abandon you nor leave you. [. . .] Be strong and brave, for you will bring this people to the land, which YHWH swore to their fathers to give to them, and you shall cause them to inherit it. YHWH [. . .] will be with you, he will not abandon you nor leave you” (Deut 31:6–8; shared formulae translated). Deuteronomy 31:6–8

‫יהוה אלהיך הוא ההלך עמך לא ירפך ולא יעזבך‬ ‫ויקרא משה ליהושע ויאמר אליו לעיני כל ישראל חזק ואמץ כי אתה תבוא את העם הזה אל הארץ‬ ‫אשר נשבע יהוה לאבתם לתת להם ואתה תנחילנה אותם‬ ‫ויהוה הוא ההלך לפניך הוא יהיה עמך לא ירפך ולא יעזבך‬

Deuteronomy 31:23 ‫ויצו את יהושע בן נון ויאמר חזק ואמץ כי אתה תביא את בני ישראל אל הארץ אשר נשבעתי להם‬ ‫ואנכי אהיה עמך‬ Joshua 1:5–6

‫כאשר הייתי עם משה אהיה עמך לא ארפך ולא אעזבך‬ ‫חזק ואמץ כי אתה תנחיל את העם הזה את הארץ אשר נשבעתי לאבותם לתת להם‬

I hold that, as a general rule, when one text (here Josh 1:3–6) separately interacts with two others or more (here Deut 1:7–8; 11:24–25; 31:6–8, 23), it is more likely that a late scribe has drawn upon a body of authoritative literature and less feasible that the single narrative exerted its influence upon several different texts. Admittedly, it is not impossible that a single author would repeat his stock of formulae, but the intertextual mosaic before us in Josh 1:3–6 does not appear

802

Cynthia Edenburg

to have been freely composed and instead seems to be the product of scribal study of previous texts.16 Hence it would appear that the section in Josh 1:3–6 was drafted by a scribe who was familiar with the scroll of Deuteronomy in its close-to-final form. The way this material has been reused without fully adapting it to the context in Josh 1 betrays its origin as a secondary expansion to the base narrative.17 Of course, one cannot dismiss out of hand the alternate possibility that Josh 1 exerted influence upon the text of Deuteronomy. Indeed, secondary assimilation is a known literary phenomenon, whereby an initial similarity inherent to two texts or created by the author of one of the texts triggers later scribes to further develop the similarity by inserting into the source text additional intertextual links lifted from the text it influenced.18 However, the theme of Josh 1:3–6 is that YHWH will be with Joshua just as he was with Moses. It would be odd if a text that places Joshua on par with Moses should exert influence on intertexts in Deuteronomy, but it is not at all surprising that the section in Josh 1 would invoke the authority of the Deuteronomy intertexts. Verses 7–8 are marked by a change in focus as well as repetition. Verse 7 opens by picking up the exhortation to be strong and brave (‫ )חזק ואמץ‬from v. 6 but qualifying it with a disjunctive ‫רק‬. Afterwards, vv. 7 and 8 continue the theme of Torah observance, thereby reinterpreting the military thrust of the call to be strong and brave (‫ ;חזק ואמץ‬cf. Deut 31:6; Josh 10:25; 2 Chr 32:7), by applying the display of fortitude to Torah observance. Nevertheless, there are some indications that vv. 7 and 8 are not from the same hand. Verse 7 is a pastiche of Deuteronomistic expressions (cf. Deut 5:32; 29:8; Josh 23:6).19 By contrast, though v. 8 seems to reflect Deuteronomistic notions, its language is not typically Deuteronomistic. For example, the verb ‫“( מוש‬to depart”) is notably  In a similar vein, see Nelson, Joshua (see n. 15), 33; Dozeman, “Joshua in the Book of Joshua” (see n. 10), 112–114. Krause, Exodus (see n. 4), 75–76, explains the second-person plural in Josh 1:3–4 in light of the citation formula at the end of v. 3: “as I spoke to Moses” (‫)כאשר דברתי אל משה‬. In other words, Josh 1:3–4 self-consciously labels itself as a quotation in which the second-person plural was blindly retained, although other changes were made to accommodate the intertext to its new context (e.g., ‫ לכם יהיה‬from Deut 11:24 becomes ‫לכם‬ ‫ נתתיו‬in Josh 1:3). However, Krause prefers to attribute all of Josh 1:3–6 (with the exception of a few glosses) to a single Deuteronomistic author, who is also responsible for the Deuteronomistic historical frame of Deuteronomy (132–133). 17  By contrast, E. A. Knauf, Josua (ZBK 6; Zürich: Theologischer Verlag, 2008) 39, 42, includes vv. 3–6 in the base narrative. 18  See Y. Z akovitch , “Assimilation in Biblical Narratives,” in Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 175–196; C. Edenburg, Dismembering the Whole (Judg 19–21): Composition and Purpose of the Story of the Outrage at Gibeah (AIL 24; Atlanta: Society of Biblical Literature, 2016), 212–214, 219–220. 19  Only in Josh 1:7; 23:6; and Deut 5:32 does the phrase )‫ שמר לעשות (כ)אשר (צוה‬occur along with ‫לא תסור (ממנו) ימין ושמאל‬. Furthermore, the phrase )‫ למען תשכיל(ו‬occurs only in Josh 1:7; Deut 29:8; and 1 Kgs 2:3. 16

Do the Pentateuchal Sources Extend into the Former Prophets?

803

absent from Deuteronomy, although it occurs elsewhere in the Pentateuch (Exod 13:22; 33:11; Num 14:44), whereas the phrase ‫לא ימוש ספר התורה הזה מפיך‬ (“do not let this law book depart from your mouth”) is uniquely paralleled by Isa 59:21: ‫] לא ימושו מפיך‬. . .[ ‫“( ודברי‬my words . . . will not depart from your mouth”).20 Verse 8 further picks up on several of the lexemes and expressions from v. 7 while reversing the syntax, which is a sign of reinterpretation. Joshua 1:7 ‫רק חזק ואמץ מאד לשמר לעשות ככל התורה אשר צוך משה עבדי אל תסור ממנו ימין ושמאול‬ ‫למען תשכיל בכל אשר תלך‬ Joshua 1:8 ‫לא ימוש ספר התורה הזה מפיך והגית בו יומם ולילה למען תשמר לעשות ככל הכתוב בו כי אז‬ ‫תצליח את דרכך ואז תשכיל‬

Thus, whereas v. 7 speaks of the instruction (‫ )תורה‬that Moses commanded,21 v. 8 deals with a book of the Torah (‫)ספר התורה הזה‬, and observance applies to what is written (‫)הכתוב‬. Finally, v. 8 shares the sentiment and a few of the lexical values in Deut 17:18–20. Joshua 1:7–8

‫רק חזק ואמץ מאד לשמר לעשות ככל התורה אשר צוך משה עבדי‬ ‫אל תסור ממנו ימין ושמאול למען תשכיל בכל אשר תלך‬ ‫לא ימוש ספר התורה הזה מפיך והגית בו יומם ולילה למען‬ ‫תשמר לעשות ככל הכתוב בו‬ ‫כי אז תצליח את דרכך ואז תשכיל‬

20  Other atypical expressions include the verb ‫“( הגה‬to utter, think upon”), which does not recur in the Pentateuch or the historical books, whereas the phrase ‫ והגית בו יומם ולילה‬has a close parallel in Ps 1:2: ‫“( ובתורתו יהגה יומם ולילה‬he shall think upon his Torah day and night”). Though the collocation ‫ועשה‬/‫ שמר לעשות‬and preposition ‫ למען‬frequently occur separately in Deuteronomy, the phrase ‫( למען תשמר לעשות‬or even ‫ למען שמר‬or ‫ )למען עשה‬does not recur in Deuteronomy or the historical books. The phrase )‫“( (כ)כל הכתוב בו (בספר התורה‬all that is written in it/the book”) is also not found in Deuteronomy, even though it might be expected in Deut 31:12 (‫ ;)את כל דברי התורה הזאת‬otherwise it recurs in Josh 8:34; 23:6; and 2 Kgs 22:13 (// 2 Chr 34:21); cf. Jer 25:13. Finally, the expression ‫אז תצליח את דרכך( צלח דרך‬, “then your way will be successful”) recurs but once in Deut 28:29, whereas the other instances are clearly postexilic (Gen 24:21, 40, 42; Judg 18:5; Isa 48:15; Ps 37:7). 21  The words ‫“( ככל התורה‬all the Torah”) in v. 7 are a plus in MT. Tov holds that MT represents an expanded edition of Joshua that was similar to the Vorlage of LXX; see E. Tov, “The Growth of the Book of Joshua in Light of the Evidence of the Septuagint,” in The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999), 385–396, here 394–96; idem , Textual Criticism of the Hebrew Bible (3rd ed.; Minneapolis: Fortress, 2012), 253. For critique and arguments for the priority of several of the MT pluses, see M. N. van der Meer, “Textual Criticism and Literary Criticism in Joshua 1:7 (MT and LXX),” in Tenth Congress of the International Organization for Septuagint and Cognate Studies: Oslo, 1998 (ed. B. A. Taylor; Society of Biblical Literature Septuagint and Cognate Studies 51; Atlanta: Society of Biblical Literature, 2001), 355–371, here 362–371.

804 Deuteronomy 17:18–20

Cynthia Edenburg

‫וכתב לו את משנה התורה הזאת על ספר מלפני הכהנים הלוים‬ ‫והיתה עמו וקרא בו כל ימי חייו למען‬ ‫ילמד ליראה את יהוה אלהיו‬ ‫לשמר את כל דברי התורה הזאת ואת החקים האלה לעשתם‬ ‫לבלתי רום לבבו מאחיו‬ ‫ולבלתי סור מן המצוה ימין ושמאול למען‬ ‫יאריך ימים על ממלכתו הוא ובניו בקרב ישראל‬

This suggests that one of the aims of the reinterpretation of v. 7 was to bring its exhortation in line with the law of the king, which devalues royal and military leadership and instead dictates that the true king and military leader is he who devotes himself to Torah.22 All the lines of evidence thereby indicate that, though v. 7 might be a nomistic Fortschreibung, or overwriting, v. 8 is a very late scribal reinterpretation that seems to derive from a time in which its intertexts from Psalms and Trito-Isaiah were already considered authoritative. According to this analysis, there are strong literary-critical grounds for viewing vv. 3–8 as the product of a series of secondary accretions. Once this material is bracketed out, it is possible to reconstruct a continuous narrative running from the command to cross the Jordan in v. 2 to the divine words of encouragement in v. 9 and Joshuaʼs commands to the officers in vv. 10–11.23 The most significant break in the narrative occurs following v. 11. There Joshua instructs the officers to prepare the people for crossing the Jordan in another three days (1:11). This thread is resumed only in Josh 3:2, which reports, “three days later, the officers went through the camp.” Although repetitive resumption, or Wiederaufnahme, can be a stylistic feature of a composition,24 in 22  On vv. 7–8 see, e.g., Noth, Deuteronomistic History (see n. 12), 62, n. 1; A. Rofé, “The Piety of the Torah-Disciples at the Winding-Up of the Hebrew Bible: Josh 1:8; Ps 1:2; Isa 59:21,” in Bibel in jüdischer und christlicher Tradition: Festschrift für Johann Maier zum 60. Geburtstag (ed. H. Merklein et al.; BBB 88; Frankfurt: Anton Hain, 1993), 78–85, here 78– 79; Lohfink, Theology of the Pentateuch (see n. 15), 239–241; T. Römer, “Josué, lecteur de la Torah (Jos 1,8),” in “Lasset uns Brücken bauen . . .”: Collected Communications to the XVth Congress of the International Organization for the Study of the Old Testament, Cambridge, 1995 (ed. K.-D. Schunck and M. Augustin; BEATAJ 42; Frankfurt: Lang, 1998), 117–124. Some also attribute v. 9 to the expansion in vv. 7–8; see R. Smend, “The Law and the Nations; A Contribution to Deuteronomistic Tradition History,” in Reconsidering Israel and Judah: Recent Studies on the Deuteronomistic History (ed. G. N. Knoppers and J. G. McConville; Winona Lake, IN: Eisenbrauns, 2000), 95–110, here 96–98; Nentel, Träger­schaft (see n. 4), 24–27. By contrast, Krause maintains that vv. 7–8 are original to YHWHʼs speech but without engaging any of the idiomatic and intertextual evidence presented here; see J. J. Krause, “The Book of the Torah in Joshua 1 and 23 and in the Deuteronomistic History,” ZAW 127 (2015): 412–428, here 417–420. 23  See also T. Römer, The So-Called Deuteronomistic History: A Sociological, Historical and Literary Introduction (London: T&T Clark, 2005), 117, n. 19; 175. 24  See S. Talmon , Literary Studies in the Hebrew Bible – Form and Content: Collected Studies (Jerusalem: Magnes, 1993), 112–133, esp. 118–122; I. L. Seeligmann, Gesammelte

Do the Pentateuchal Sources Extend into the Former Prophets?

805

this case it is employed as an editorial device to bridge the break in the narrative that occurred when secondary accretions were inserted – namely, the passage about the Transjordanian tribes (1:12–18) and the spy story in chapter 2.25 This suspicion is further supported by analysis of the language in 1:12–18, as well as its intertextual relations with other passages that deal with the inheritance of the Transjordan tribes. Though this section is massively dependent upon Deut 3:8–20, it appears that Num 32 also exerted influence upon the text of both Deut 3 and Josh 1 at some stage in the process of transmission.26 For example, the formula used in v. 12 and elsewhere in Joshua when speaking of the two and a half tribes differs from the use in Deuteronomy, which would seem to indicate separate transmission, if not separate composition.27 Though the phrase ‫( נתן את הארץ הזאת‬v. 13, “give this land”) recurs in both Num 32:5 and Deut 3:18 with regard to the inheritance of the Transjordan tribes, this precise wording of the land-grant formula is particularly frequent in the patriarchal promises (Gen 12:7; 15:7, 18; 24:7; 48:4), whereas in Deuteronomy it recurs only in the first-fruits passage (Deut 26:9). Not surprisingly, all these passages are assigned by recent scholars to the latest layers in the composition of the Pentateuch.28 Verse 14, ‫נשיכם טפכם ומקניכם ישבו בארץ אשר נתן לכם משה‬ (“your women, children, and cattle will dwell in the land that Moses gives you”), shares with Deut 3:19 the unique formula ‫נשיכם וטפכם ומקנכם‬,29 but at the same time it also shares with Num 32:7, 9 the phrase ‫הארץ אשר נתן להם‬, which is absent from Deut 3. The elements from the two sources – Num 32 and Deut 3 – are combined by the scribe of Josh 1:14 in a peculiar fashion that creates an alternate description, according to which the eastern tribesʼ women, children, and cattle will dwell in the land (rather than in the cities, as in Num 32:17, 26; Deut 3:19) that Moses gave them (rather than YHWH, as in Num 32:7, 9). The end of Josh 1:14 also combines elements from both sources: “you shall pass arrayed for war” shares the passive participle ‫חמשים‬ ֻ with Num 32:17 (LXX; Studien zur Hebräischen Bibel (ed. E. Blum; FAT 41; Tübingen: Mohr Siebeck, 2004), 119–138, esp. 128–129, 135. 25  In a similar vein, see J. Van S eters , “Joshuaʼs Campaign of Canaan and Near Eastern Historiography,” in Israelʼs Past in Present Research: Essays on Ancient Israelite Historiography (ed. V. P. Long; Winona Lake, IN: Eisenbrauns, 1999), 170–189, here 172; Nelson, Joshua (see n. 15), 41; Knauf, Josua (see n. 17), 46–47; Blum, “Pentateuch-Hexateuch-Enneateuch?” (see n. 7), 55–57; Krause, Exodus (see n. 4), 132–133. 26  On the redaction history of Num 32, see L ohfink , Theology of the Pentateuch (see n. 15), 216–218. 27  Josh 1:12 and 12:6 read ‫ראובני גדי וחצי שבט המנשה‬, whereas in Deuteronomy, when the three are mentioned together, they all are inflected as gentilics: ‫ראובני גדי וחצי ששבט המנשי‬ (Deut 4:43; 29:7). The inflected form is notably absent from the book of Joshua but is paralleled elsewhere by 2 Kgs 10:33. 28  The same formulation also recurs in Jer 32:22; see also Exod 32:13 and Num 14:8, where the syntax differs. 29  Num 32:26 has a different formulation: ‫טפנו נשינו מקננו וכל בהמתנו‬.

806

Cynthia Edenburg

MT ‫)חשים‬, ֻ and “before your brothers, all the brave men” (‫ )כל גבורי החיל‬shares the term ‫ חיל‬with Deut 3:18 (‫)כל בני חיל‬. Verses 17–18 also display some peculiarities that indicate that its diction is derivative rather than that of a pentateuchal source. The eastern tribesʼ reply, “Just as we obeyed Moses so we shall obey you” (v. 17, ‫ככל אשר שמענו אל משה‬ ‫)כן נשמע אליך‬, echoes both Exod 24:7 (‫כל אשר דבר יהוה נעשה ונשמע‬, “we shall verily obey all that YHWH has said”) and Deut 5:27 (‫כל אשר ידבר יהוה אלהינו‬ ‫אליך ושמענו ועשינו‬, “all that YHWH our god said to you we shall verily obey”). In the Pentateuch, the affirmation of obedience applies to divine dicta, whereas in v. 17 it affirms the authority of human leadership. The affirmation of Joshuaʼs authority in v. 17 also evokes the language of Deut 34:9, which is widely thought to stem from the final redaction of Deuteronomy or even of the Pentateuch as a whole:30 Joshua 1:17

‫נשמע אליך רק יהיה יהוה אלהיך עמך כאשר היה עם משה‬

Deuteronomy 34:9 ‫] וישמעו אליו בני ישראל ויעשו כאשר צוה יהוה את משה‬. . .[ ‫ויהושע בן נון‬

In v. 18, the tribes further affirm Joshuaʼs authority with the sanction, “whoever disobeys your pronouncements (‫[ )ימרה את פיך‬. . .] shall be executed.” Elsewhere the idiom ‫ להמרות פה‬is applied only to YHWHʼs pronouncements (Num 20:24; 27:14; Deut 1:26, 43; 9:23; 1 Sam 12:14–15; 1 Kgs 13:21, 26), whereas here it applies to a human agent, Joshua.31 Additionally, the final note of empowerment at the end of v. 18, ‫חזק ואמץ‬, is voiced by the eastern tribes, when previously it was employed to designate YHWHʼs empowerment of Joshua. Throughout this section, there is a marked tendency to appropriate language used in relation to YHWHʼs authority and to reapply it to the human sphere; this is done with an eye to bolstering the authority of Joshua beyond the comparisons with Moses that figure throughout the chapter (vv. 3, 5, 7, 17). This type of appropriation seems to be a clear sign of distance from the composition of the pentateuchal sources themselves. Finally, although the language of the chapter is dominated by Deuteronomistic idioms,32 it also contains various locutions that are not characteristic of

 See, e.g., T. Römer and M. Z. Brettler, “Deuteronomy 34 and the Case for a Persian Hexateuch,” JBL 119 (2000), 401–419. 31  The only comparable text that specifies death for defying a human authority is Deut 17:12 (priests and judges), but here the idiom ‫ להמרות פה‬is lacking. See also the law of the prophet in Deut 18:19, which employs the common expression ‫ לא שמע‬instead of ‫להמרות פה‬, which leaves the penalty to divine sanction. 32  See, e.g., Noth, Deuteronomistic History (see n. 12), 61; Van Seters, In Search (see n. 5), 324; and, most recently, Krause, Exodus (see n. 4), 132–133. 30

Do the Pentateuchal Sources Extend into the Former Prophets?

807

Deuteronomy but rather recur in the Pentateuch in Priestly or non-P contexts.33 These include the characterization of Joshua as Mosesʼs servant (‫ משרת משה‬in v. 1),34 the expression “Moses my servant” (‫ משה עבדי‬in vv. 2, 7),35 “the great sea” (‫ הים הגדול‬in v. 4),36 the verb ‫“( מו״ש‬to withdraw, be removed” in v. 8),37 the term ‫חמשים‬ ֻ (“arrayed for war” in v. 14),38 and the phrase “YHWH was/will be with him/them” (‫את‬/‫ היה יהוה עם‬in v. 17).39 Some of the expressions, which are also found in Deuteronomy, are limited to very late Deuteronomistic (or postDtr) redaction layers, such as the title “servant of YHWH” applied to Moses (‫ עבד יהוה‬in vv. 1, 13, 15; cf. v. 2, ‫)משה עבדי‬,40 ‫ התיצב‬in the sense of “stand up to, oppose” (v. 5),41 and the word pair ‫ עזב‬// ‫( רפה‬v. 5).42 There are also unusual usages, such as the expression “the land of the Hittites” (‫ ארץ החתים‬in v. 4),  See T. B. Dozeman, “Joshua 1,1–9: The Beginning of a Book or a Literary Bridge?,” in The Book of Joshua (ed. E. Noort; BETL 250; Leuven: Peeters, 2012), 159–182, here 169–171. 34  See R ömer , So-Called Deuteronomistic History (see n. 23), 52. The construct phrase recurs elsewhere only in Num 11:28, which belongs to a late post-Dtr supplement; see, e.g., H. Seebass, Numeri (BKAT; Neukirchen-Vluyn: Neukirchener Verlag, 1993), 34–35; Römer, “Israelʼs Sojourn” (see n. 11), 438–439, 442–443. Elsewhere it occurs with the genitive suffix (‫ ;)משרתו‬see Exod 24:13; 33:11. 35  See R ömer , So-Called Deuteronomistic History (see n. 23), 52. The same word order recurs only in Mal 3:22, whereas only the reverse order occurs in the Pentateuch: ‫עבדי משה‬ (Num 12:7–8; see also 2 Kgs 21:8). Compare ‫( אברהם עבדי‬Gen 26:24); a single instance of ‫( יעקב עבדי‬Isa 44:1) compared to eight instances of ‫עבדי יעקב‬, all in Deutero-Isaiah, Jeremiah, and Ezekiel; and ‫( עבדי כלב‬Num 14:24). 36  The phrase ‫ הים הגדול‬occurs in the Pentateuch only in Num 34:6–7; it recurs elsewhere only in Josh 9:1; 15:47; and 23:4 (Former Prophets) and Ezek 47:15, 19–20; 48:28 (boundary lists). 37  The verb ‫ מו״ש‬is lacking in Deuteronomy but otherwise occurs in the Pentateuch in Exod 13:22; 33:11; and Num 14:44; it recurs in the Former Prophets only in Judg 6:18 and elsewhere in Isa 22:25; 46:7; 54:10 (twice); 59:21; Jer 17:8; 31:36; Zech 14:4; Ps 55:12; Prov 17:13; and Job 23:12. 38  The passive participle ‫חמשים‬ ֻ occurs only twice in the Pentateuch – Exod 13:18; Num 32:17 (LXX) – and elsewhere only in Josh 4:12 and Judg 7:11. 39  The phrase recurs fifteen times with the same syntax; in the Pentateuch: Gen 26:28; 39:2, 21; Num 14:43; in the Former Prophets: Josh 6:27; Judg 1:19; 2:18; 1 Sam 18:12 (David); 20:13; 1 Kgs 1:37; 8:57; 2 Kgs 18:7; in Chr Sondergut: 1 Chr 22:11, 16; 2 Chr 17:3. The phrase also occurs three more times with the subject preceding the verb: 1 Sam 3:19; 17:37; and 2 Sam 14:17. 40  Deut 34:5; Josh 8:31, 33; 11:12; 12:6; 13:8; 14:7; 18:7; 22:2, 4–5; 2 Kgs 18:12. ‫עבד‬ ‫ ה׳‬is an MT plus in Josh 1:1, 15, but there is no warrant for excising the expression since LXX does attest to its use in v. 13. For use of the title, see G. Braulik, “Die deuteronomistische Landeroberungserzählung aus der Joschijazeit in Deuteronomium und Josua,” in Das deuteronomistische Geschichtswerk (ed. H.-J. Stipp; ÖBS 39; Frankfurt: Lang, 2011), 89–150, here 103. However, Braulik seems to ignore the implications of the fact that the term appears in verses that are widely considered late insertions into the initial Deuteronomistic composition. 41  Deut 7:24; 9:2; 11:25; cf. Ps 2:2; 2 Chr 20:6. 42  Deut 31:6, 8; cf. Ps 37:8; 1 Chr 28:20. 33

808

Cynthia Edenburg

which recurs only in Judg 1:26, whereas the parallel context in Deut 1:7 reads ‫;ארץ הכנעני‬43 the verb ‫הגה‬, which does not recur in any other narrative context (v. 8);44 and the application of the idiom ‫להמרות פה‬, “to disobey” (v. 18) to a human object when elsewhere it always refers to disobeying YHWH. To sum up, close analysis of the language of the chapter shows that it is far from being classically Deuteronomistic, since it displays many affinities with the diction of P and late non-P in the Pentateuch, as well as late Deuteronomistic redaction, late prophetic literature, and the Psalms. Furthermore, large sections of the chapter were shaped by means of creative reutilization and recombination of intertexts from various pentateuchal sources, particularly Deuteronomy in its near-to-final form. Additionally, literary analysis has shown that the chapter has grown in at least three stages: the base narrative is present in vv. 1–2, 9–11; the second stage added vv. 3–6 and probably vv. 12–17; and the third stage added vv. 7–8, although v. 8 might be a later reinterpretation of v. 7. These stylistic and intertextual findings might be explained in different ways. The fact that the material in the chapter has grown in stages could indicate that an earlier pre-Deuteronomistic or Deuteronomistic source has been massively overwritten. Nonetheless, the alternative must also be considered, namely that even the basic narrative stems from a late Deuteronomistic or even post-Deuteronomistic scribe, who wished to show how Joshua faithfully carried out the program laid out by Moses in Deuteronomy. In any event, in its final form, the chapter seems to be a Pentateuch-oriented composition that links Joshua to both Deuteronomy and Exodus and Numbers.45

43  The expression “the land of the Hittites” is an MT plus and glosses the detailed geographic designation of the extent of the land. Nevertheless, I think it is an explicatory gloss that was deleted from the LXX or its Vorlage when the term was no longer understood. The expression is undoubtedly a reflection of the Akkadian formula māt Ḫatti, which indicates the geographical extent of land west of the Euphrates, from Asia Minor to Egypt. Though the term māt Ḫatti originated in the second millennium BCE, it became prevalent in the Neo-Assyrian period as a designation for the land west of the Euphrates. Afterwards, it remained a fossil in the scribal curriculum, as attested by chronographic literature from the late seventh century BCE down to the Seleucid period; see J.-J. G lassner , Mesopotamian Chronicles (SBLWAW 19; Atlanta: Society of Biblical Literature, 2004), texts 24 (obv. l. 16, rev. l. 9), 52 (l. 14), 53 (iv ll. 50′–57′). 44  The verb occurs elsewhere in Is 8:19; 16:7; 31:4; 33:18; 38:14; 59:3, 11, 13; Jer 48:31; Pss 1:2; 2:1; 35:28; 37:30; 38:13; 63:7; 71:24; 77:13; 115:7; 143:5; Prov 8:7; 15:28; 24:2; and Job 27:4. 45  This view is broadly in agreement with Dozemanʼs, although we differ on points of the analysis; see Dozeman, “Joshua in the Book of Joshua” (see n. 10); idem, “Joshua 1,1–9” (see n. 33); idem, Joshua 1–12 (see n. 5), 204–206.

Do the Pentateuchal Sources Extend into the Former Prophets?

809

4 Conclusions and Implications Josh 1 achieved its form through a lengthy process of accretion, or overwriting, similar to the process that can be observed in the framework of Deuteronomy. These accretions do not necessarily represent systematic redactions that run throughout the book, and only in a few cases (such as Josh 1:12–18; 13:7–8; 22:1–8) would I venture to relate them to a common editorial layer. Joshua 1 interacts with many pentateuchal materials and formulations that derive not only from Deuteronomy (particularly from late layers of the framework and parenesis) but also from Priestly material and late non- or post-Priestly material, as can be seen in various idioms. These intertexts have been combined and reworked in a derivative fashion that indicates the influence of the pentateuchal source texts and the authority they held – an authority that the scribes wished to evoke while reworking the Joshua scroll. Thus, Josh 1 was composed by scribes who were heavily influenced by Deuteronomistic style and concerns but who also seem to have been familiar with other streams of pentateuchal literary tradition. The scribes of Josh 1 utilize pentateuchal sources as prooftexts, but Josh 1 does not belong to the same literary framework as the pentateuchal sources themselves. In fact, the way Josh 1 evokes pentateuchal authority suggests its distance from the pentateuchal sources themselves. Joshua 1 thus indeed introduces the narrative sequel to the Pentateuch, which now focuses on fulfillment of divine promises and observance (or infraction) of divine strictures.46 This chapter, perhaps more than any other, provides the glue necessary to produce a continuous narrative running from both the end of Numbers and the end of Deuteronomy to the account of the possession of the land in Joshua. From this we can infer its purpose – namely, to facilitate reading the Joshua scroll as a direct continuation of the scrolls that dealt with the period of Moses.47 Indeed, it is important to consider the role of the scroll as the material medium for the composition of biblical texts. Literary-critical examination of the biblical books shows that they grew outward from a central core by means of accreted blocks at the both ends. This is best explained by the nature of the scroll medium, since it was simplest to revise a composition by redirecting its purpose through new material added to the end sheets of the scroll. In theory, the interpolation of revisions that overwrite the text could occur at any stage in which a new copy of a scroll was prepared, but I surmise that most overwriting (or Fortschreibung) occurred after the scope of a scroll had attained its basic shape. Contrary to Nothʼs view that the Deuteronomistic History was subsequently divided into 46  Knaufʼs succinct remark on Josh 1:7–8 could well apply to the chapter as a whole: “1,7–8 machen aus dieser Einleitung zur Josua-Geschichte eine Einleitung in das gesamte Propheten-Corpus des hebräischen Kanons”; see Knauf, Josua (see n. 17), 40. 47  For a canon-oriented reading of Josh 1, see Lohfink, Theology of the Pentateuch (see n. 15), 234–247.

810

Cynthia Edenburg

separate books, it would seem that the comprehensive history work was the product of bringing together the separate scrolls, each of which was Deuteronomistic in its own fashion.48 This insight holds significant implications for understanding the shape of Joshua (and various other biblical books), as well as the question whether pentateuchal sources extended into the Former Prophets. I agree with those who propose that the story of the conquest was composed during Josiahʼs reign, as a means to justify the annexation of areas of southern Samaria. This proposal best explains the close relations between the conquest story and the Neo-Assyrian ideology of war and its practice.49 However, there is no reason to assume that the composition at this time opened with the crossing of the Jordan. The idea of entering the land from outside would make no sense to a late seventh-century Judean readership. Indeed, Ezek 33:24 seems to reflect an early tradition that attributes possession of the land to an autochthonous Abraham.50 The conquest story that was composed during Josiahʼs reign dealt with the conquest of Jericho, Ai, and the Shephelah (Josh 6–8*, 10*)51 and concluded with the fulfillment summary in Josh 21:43–45.52 I suggest that the idea of entering the land from the east was added during the Babylonian period. This has weighty implications not only for Josh 1 and 3–5 but also for the literary fiction of Mosesʼs speech on the plains of Moab as carried out in the narrative framework of Deuteronomy (e.g., Deut 1:1–5; 2:2–9; 28:69; 31:7–8, 23) and in the conditions of several of its laws (cf. Deut 12:10; 17:14; 18:9; 26:1).53 The eisodus theme represents the fulfillment of the promises to the fathers, which is 48  C. Edenburg, “Rewriting, Overwriting, and Overriding: Techniques of Editorial Revision in the Deuteronomistic History,” in Words, Ideas, Worlds in the Hebrew Bible: Essays in Honour of Yairah Amit (ed. F. Polak et al.; Sheffield: Sheffield Phoenix, 2012), 54–69, here 64–66. See also Römerʼs view of a Deuteronomistic anthology or library in Römer, So-Called Deuteronomistic History (see n. 23), 71–72, 104–106. 49  See, e.g., Lohfink, Theology of the Pentateuch (see n. 15), 187–188; Braulik, “Die deuteronomistische Landeroberungserzählung” (see n. 40); Römer, So-Called Deuteronomistic History (see n. 23), 82–90; Nihan, “Literary Relationship” (see n. 6), 109–114. 50  I. Finkelstein and T. Römer “Comments on the Historical Background of the Abraham Narrative: between ‘Realia’ and ‘Exegetica,’ ” HBAI 3 (2014): 3–23, here 10–11. 51  See Nihan, “Literary Relationship” (see n. 6), 106–107, with further references there. 52  Nihan, “Literary Relationship” (see n. 6), 85–93, argues at length that Josh 21:43–45 is a secondary conclusion, since it interacts with Deut 12:8–12 and 1 Kgs 8. However, as Nihan himself notes, cross-referencing and recurring phraseology do not necessarily mean that they are the product of one and the same scribe (87). Hence, there is no reason not to suppose that other related scrolls picked up themes from the conclusion of the Joshua scroll in order to create a thematic unity. 53  Otto suggests that the addressees of Mosesʼs speech within the story world – namely, the second generation of the desert wanderings – represent the second generation of exile in the authorʼs world; see E. Otto, Deuteronomium 1–11, Erster Teilband: 1,1–4,43 (HThKAT; Freiburg: Herder, 2012), 238, 243–244, 247–248.

Do the Pentateuchal Sources Extend into the Former Prophets?

811

later revoked in light of the continued infractions of YHWHʼs treaty and thus provides the theological justification of the exile. At the same time, the entry theme (particularly from the east) provided a basis for fostering hopes for a return, when YHWH would once more lead his people into the land he gave them (cf. Deut 30:3–5). Thus, it appears that the original opening of the story of the conquest was heavily overwritten or replaced by the materials relating to crossing the Jordan in Josh 1* and 3:1–5:12, all of which work to connect the Joshua scroll with the scrolls of the Pentateuch. Many of the various accretions at the beginning of the Joshua scroll presume the authority of preexisting pentateuchal texts, so they do not continue pentateuchal sources but are secondary actualizations or reinterpretations of authoritative sources.54 Thus, to my way of thinking, there is no true continuation of the pentateuchal sources in Josh 1–5. How, then, should we understand the relationship between the Former Prophets and the pentateuchal sources, and what constitutes a pentateuchal source? For adherents of the Documentary Hypothesis, the sources are independent and continuous literary works that were combined together in different stages of redaction. If this is what is meant by JE, P, and D, then they have no direct continuation in the Former Prophets. In contrast, if pentateuchal sources are thought to be blocks of tradition that evolved separately and merged together at a late stage, then we would probably agree that there are traces of pentateuchal sources in the Former Prophets. In any event, we must seriously consider the likelihood that literarily proficient scribes freely revised the scrolls they copied up until a very late stage. This consideration critically hampers the certainty of our attempts to reconstruct sources and retrace the stages of their revision and redaction, which carries consequences for understanding the concept of a Hexateuch as well. The Hexateuch (as well as the Enneateuch) is neither a composition nor a division recognized in reception history but is rather a reading that is cued by the redactional work which created the literary links between the separate scrolls.55 Finally, is it really inconceivable that Judean scribes might compose literary works about the divine promise of land and conclude them without relating the fulfillment of the divine promises? The answer depends on how we are disposed to view the historical context within which these works were composed, how we identify their target audiences, and how we understand their original message. Within the context of the Deuteronomistic History, the complete fulfillment of the divine promises is countered by the continued unfaithfulness of YHWHʼs people, which in turn leads YHWH to revoke that which he promised. Without 54  For a similar view, see Dozeman, “Josh 1,1–9” (see n. 33); idem, Joshua 1–12 (see n. 5), 24–31, 200; cf. Rendtorff, Problem (see n. 3), 188. 55  Cf. C. Frevel, “Die Wiederkehr der Hexateuchperspektive: Eine Herausforderung für die These vom deuteronomistischen Geschichtswerk,” in Das deuteronomistische Geschichtswerk (ed. H.-J. Stipp; ÖBS 39; Frankfurt: Lang, 2011), 13–53, here 23–27, 44–45.

‫‪Cynthia Edenburg‬‬

‫‪812‬‬

‫‪the fulfillment of the divine promises, the lesson of the history work is lost,‬‬ ‫‪and Deuteronomy remains little more than a program for cult centralization.‬‬ ‫‪However, a Tetrateuch or Pentateuch that stops short of fulfillment of the divine‬‬ ‫‪promises leaves room to hope for their eventual fulfillment – a message that‬‬ ‫‪must have been meaningful not only to a diaspora community but also to those‬‬ ‫‪living in Judah and Samaria under foreign rule.56‬‬ ‫‪Compositional Analysis of Joshua 1:1–18‬‬ ‫)‪ (1‬ויהי אחרי מות משה עבד יהוה ויאמר יהוה אל יהושע בן נון משרת משה לאמר )‪ (2‬משה עבדי‬ ‫מת ועתה קום עבר את הירדן הזה אתה וכל העם הזה אל הארץ אשר אנכי נתן להם לבני ישראל‬ ‫)‪ (3‬כל מקום אשר תדרך כף רגלכם בו לכם נתתיו כאשר דברתי אל משה )‪ (4‬מהמדבר והלבנון‬ ‫הזה ועד הנהר הגדול נהר פרת כל ארץ החתים ועד הים הגדול מבוא השמש יהיה גבולכם‬ ‫)‪ (5‬לא יתיצב איש לפניך כל ימי חייך כאשר הייתי עם משה אהיה עמך לא ארפך ולא אעזבך‬ ‫)‪ (6‬חזק ואמץ כי אתה תנחיל את העם הזה את הארץ אשר נשבעתי לאבותם לתת להם‬ ‫)‪ (7‬רק חזק ואמץ מאד לשמר לעשות ככל התורה אשר צוך משה עבדי אל תסור ממנו ימין‬ ‫ושמאול למען תשכיל בכל אשר תלך‬ ‫)‪ (8‬לא ימוש ספר התורה הזה מפיך והגית בו יומם ולילה למען תשמר לעשות ככל‬ ‫הכתוב בו כי אז תצליח את דרכך ואז תשכיל‬ ‫)‪ (9‬הלוא צויתיך חזק ואמץ אל תערץ ואל תחת כי עמך יהוה אלהיך בכל אשר תלך )‪ (10‬ויצו‬ ‫יהושע את שטרי העם לאמר )‪ (11‬עברו בקרב המחנה וצוו את העם לאמר הכינו לכם צידה כי בעוד‬ ‫שלשת ימים אתם עברים את הירדן הזה לבוא לרשת את הארץ אשר יהוה אלהיכם נתן לכם לרשתה‬ ‫)‪ (12‬ולראובני ולגדי ולחצי שבט המנשה אמר יהושע לאמר )‪ (13‬זכור את הדבר אשר צוה‬ ‫אתכם משה עבד יהוה לאמר יהוה אלהיכם מניח לכם ונתן לכם את הארץ הזאת )‪ (14‬נשיכם‬ ‫טפכם ומקניכם ישבו בארץ אשר נתן לכם משה בעבר הירדן ואתם תעברו חמשים לפני אחיכם‬ ‫כל גבורי החיל ועזרתם אותם )‪ (15‬עד אשר יניח יהוה לאחיכם ככם וירשו גם המה את הארץ‬ ‫אשר יהוה אלהיכם נתן להם ושבתם לארץ ירשתכם וירשתם אותה אשר נתן לכם משה עבד‬ ‫יהוה בעבר הירדן מזרח השמש‬ ‫)‪ (16‬ויענו את יהושע לאמר כל אשר צויתנו נעשה ואל כל אשר תשלחנו נלך )‪ (17‬ככל אשר‬ ‫שמענו אל משה כן נשמע אליך רק יהיה יהוה אלהיך עמך כאשר היה עם משה )‪ (18‬כל איש‬ ‫אשר ימרה את פיך ולא ישמע את דבריך לכל אשר תצונו יומת רק חזק ואמץ‬ ‫ויהי מקצה שלשת ימים ויעברו השטרים בקרב המחנה‬

‫‪Joshua 3:2‬‬

‫­‪ See C. L evin , “Nach siebzig Jahren: Martin Noths Überlieferungsgeschichtliche Stu‬‬ ‫‪dien,” ZAW 125 (2013), 72–92, here 90–91. Compare also Dozemanʼs approach to the northern‬‬ ‫‪orientation of Joshua in his commentary: Dozeman, Joshua 1–12 (see n. 5), 31.‬‬ ‫‪56‬‬

The Problem of the Hexateuch Thomas Römer Pentateuch, Hexateuch, Enneateuch1 If one reads the Hebrew Bible from a canonical perspective, one may of course consider that the death of Moses as reported in Deut 34 is a major conclusion or, to the Pentateuch, an absolute ending. The last verses of the Pentateuch clearly indicate the idea of a chronological and theological caesura: Deuteronomy 34:10–12 Since that time no prophet has risen in Israel like Moses, whom YHWH knew face to face, for all the signs and wonders which YHWH sent him to perform in the land of Egypt against Pharaoh, all his servants, and all his land, and for all the mighty power and for all the great terror which Moses performed in the sight of all Israel.

This concluding remark suggests that the narration has come to an end, because it reports the death and a final appreciation of Moses, the main figure of the preceding story. If one looks, however, at the verses that precede this conclusion, one gets the impression that the story will continue through Joshua, introduced as Mosesʼs successor: Deuteronomy 34:8–9 So the sons of Israel wept for Moses in the plains of Moab thirty days; then the days of weeping and mourning for Moses came to an end. Now Joshua the son of Nun was filled with the spirit of wisdom, for Moses had laid his hands on him; and the sons of Israel listened to him and did as YHWH had commanded Moses.

These verses suggest that the story will be followed by the conquest of the land related in the book of Joshua. Indeed, the divine promise of the land, which is repeated throughout the books of the Torah, apparently still awaits its fulfillment, which comes, in the current form of the text, with the conquest narrative in the book of Joshua. It is, therefore, tempting to replace the Pentateuch with the Hexateuch. This theory is almost as old as historical-critical research on

1  For this chapter, see T. Römer, “How Many Books (teuchs): Pentateuch, Hexateuch, Deuteronomistic History, or Enneateuch?,” in Pentateuch, Hexateuch, or Enneateuch? Identifying Literary Works in Genesis through Kings (ed. T. B. Dozeman et al.; AIL 8; Atlanta: Society of Biblical Literature, 2011), 25–42.

814

Thomas Römer

the Pentateuch, at least as old as the Documentary Hypothesis as elaborated by Kuenen and Wellhausen.2 An alternative to the Hexateuch as the major textual unit is the Enneateuch, running from Gen 1 to 2 Kgs 25. This idea had already been formulated by Spinoza, who claimed that “all these books [. . .] were all written by a single author, who wished to relate the antiquities of the Jews from their first beginning down to the first destruction of the city.”3 Spinoza used this observation in order to argue that the Pentateuch could not have been written before all of the events found in the books of Genesis–Kings had taken place. He considered the Torah as well as the Former Prophets to have been compiled in the Persian period. Many scholars still think that there existed “a consecutive history from creation to exile,”4 an Enneateuch running from the book of Genesis to the books of Kings, from “Paradise lost to the loss of Jerusalem.”5 R. Clements suggested that the Former Prophets together with the Pentateuch should be seen as constituting the first corpus of Scripture in nascent Judaism.6 This idea comes close to that of a great Deuteronomistic History, composed during the Babylonian exile and running from Gen 2:4b to 2 Kgs 25, as advocated by E. Zenger.7 Similarly, H.-C. Schmitt thinks that the Enneateuch preceded the Pentateuch. According to him, one can recover in Genesis–Kings the hand of a late Deuteronomistic redactor who revised the older sources in the Pentateuch in order to combine them with the so-called Deuteronomistic History.8 K. Schmid is also sympathetic to the idea of an Enneateuch, but he is more skeptical about the idea that it would have existed without the Latter Prophets.9

2  For the rise of the idea of a Hexateuch, see T. Römer, “ ‘Higher Criticism’: The Historical and Literary-critical Approach – with Special Reference to the Pentateuch,” in Hebrew Bible/ Old Testament: The History of Its Interpretation (ed. M. Sæbø; Göttingen: Vandenhoeck & Ruprecht, 2013), 3.1:393–423, here 406–407. 3  B. de S pinoza , A Theologico-Political Treatise and A Political Treatise (New York: Dover, 1951), 128. 4  J. B lenkinsopp , The Pentateuch: An Introduction to the First Five Books of the Bible (ABRL; New York: Doubleday, 1992), 34. 5  B. Gosse, “Lʼinclusion de lʼensemble Genèse–II Rois, entre la perte du jardin dʼEden et celle de Jérusalem,” ZAW 114 (2002), 189–211. 6  R. E. Clements, Prophecy and Tradition (Growing Points in Theology; Oxford: Blackwell, 1975), 55. 7  See, e.g., E. Z enger et al., Einleitung in das Alte Testament (ed. C. Frevel; 8th ed.; Studienbücher Theologie 1/1; Stuttgart: Kohlhammer, 2012), 120–129. 8  H.-C. Schmitt, “Das spätdeuteronomistische Geschichtswerk Gen I–2 Regum XXV und seine theologische Intention,” in Theologie in Prophetie und Pentateuch: Gesammelte Aufsätze (BZAW 310; Berlin: de Gruyter, 2001), 277–294. 9  K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 359–373; English translation, idem, Genesis and

The Problem of the Hexateuch

815

Space does not allow a discussion of the Enneateuch theory in this context. Suffice it to say that the theory relies on the presence of a narrative coherence covering (at least parts of) the Pentateuch and the Former Prophets. But the different theories about the existence of an Enneateuch do not claim that the so-called sources or documents that can be found in the Torah also occur in the books of Kings. That has, in fact, never been a serious option in the history of research. The few attempts to trace the pentateuchal sources J and E through the beginning of Kings have never convinced anyone other than those who had this idea, namely, K. Budde, who thought that J and E ended in 1 Kgs 2, and G. Hölscher, who placed their end in 1 Kgs 12*.10 Unlike the Enneateuch theory, the hypothesis of a Hexateuch is quite different, since it is intrinsically linked to the development of the Documentary Hypothesis.

The Disappearance and Return of the Hexateuch The idea that the sources of the Pentateuch continue in the book of Joshua was a widely shared assumption of the Documentary Hypothesis. As Wellhausen had already admitted, however, there was also a kind of agreement that many, if not a majority, of texts in the book of Joshua could not be attributed to J/E or P.11 That led A. Alt, M. Nothʼs teacher, to propose that the conquest narratives in the first part of Joshua are an independent tradition that does not belong to J or E.12 Noth used this observation when he came up with his theory about the Deuteronomistic History.13 The book of Deuteronomy became the opening of a narrative unit that covered the books of Deuteronomy to 2 Kings. Several observations can indeed support Nothʼs idea of such a literary unit. First, if one looks at the openings of the books that constitute the Pentateuch, one realizes that Exodus, Leviticus, and Numbers are each closely related to the preceding book by a consecutive waw – ‫וְ ֵא ֶּלה ְשֹׁמות‬, ‫וַ ּיִ ְק ָרא‬, and ‫ – וַ יְ ַד ֵּבר‬whereas ֵ This may the book of Deuteronomy opens in an “absolute way” – ‫א ֶּלה ַה ְּד ָב ִרים‬.

the Moses Story: Israelʼs Dual Origins in the Hebrew Bible (trans. J. D. Nogalski; Siphrut 3; Winona Lake, IN: Eisenbrauns, 2010), 334–353. 10  For bibliography and details, see O. Kaiser, Einleitung in das Alte Testament (5th ed.; Gütersloh: Gerd Mohn, 1984), 94. 11  J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (Berlin: de Gruyter, 1963), 116–117. 12  A. Alt, Kleine Schriften zur Geschichte des Volkes Israel (3 vols; Munich: Beck, 1953), 1:176–192. 13  M. N oth , Überlieferungsgeschichtliche Studien: Die sammelnden und bearbeitenden Geschichts­werke im Alten Testament (Darmstadt: Wissenschaftliche Buchgesellschaft, 1967); English translation, idem, The Deuteronomistic History (trans. J. Doull et al.; 2nd ed.; JSOTSup 15; Sheffield: Sheffield Academic Press, 1991).

816

Thomas Römer

suggest that the book of Deuteronomy should be understood as the beginning of a new compositional unit. Second, the book of Deuteronomy contains pretty clear allusions to the Former Prophets: the Jordan crossing is repeatedly announced as the first step of the conquest of the land in Mosesʼs speech in Deuteronomy. This crossing is reported in Josh 3–4. Deuteronomy also contains many allusions to the conquest itself, recounted in Josh 2–12. But the previews in Deuteronomy are not limited to the book of Joshua: The separation of the people from YHWH, described at the beginning of Judges, is already mentioned in Deuteronomy (cf. Deut 6:12–15 and Judg 2:12–14). The law of the king in Deut 17:14–20 is essentially a preview of the history of the monarchy found in the books of Samuel and Kings.14 The desire to appoint a king, as is the practice among other nations (17:14), points toward 1 Sam 8. The warning against too many women, who would confuse the kingʼs mind and heart, hints at the narrative of Solomon (1 Kgs 11). The commandment to read the Torah can be understood as alluding to the devout Josiah and his discovery of the book (2 Kgs 22–23). Finally, the loss of the land and the Babylonian exile described toward the end of the two books of Kings already stands behind the curses of Deut 28. Third, the book of Deuteronomy shares some important vocabulary with the Former Prophets. It contains an important number of words and expressions that do not or only rarely occur in the Tetrateuch, as for instance the theme of the Israelites following other gods. The expression ‫ אלהים אחרים‬appears often in Deuteronomy–Kings but is found only twice in the book of Exodus. The same can be said of the root ‫( שמד‬to destroy), frequently attested in Deuteronomy and the Prophets but rare in the Tetrateuch. One may also mention the expression “to do what is evil in the eyes of YHWH,” which occurs often in all books from Deuteronomy to Kings (twenty-eight times) but only once before (in Num  The core of Deut 17:14–20 is dated by most scholars to either the seventh or the sixth century BCE; see, e.g., F. G arcía L ópez , “Le roi dʼIsraël: Dt 17,14–20,” in Das Deuteronomium: Entstehung, Gestalt und Botschaft (ed. N. Lohfink; BETL 68; Leuven: Peeters, 1985), 277–297; B. M. Levinson, “The Reconceptualization of Kingship in Deuteronomy and the Deuteronomistic Historyʼs Transformation of the Torah,” VT 51 (2001), 511–534; E. W. Nicholson, “ ‘Do Not Dare to Set a Foreigner Over You’: The King in Deuteronomy and ‘The Great King,’ ” ZAW 118 (2006), 46–61; R. Albertz, “A Possible terminus ad quem for the Deuteronomic Legislation? A Fresh Look at Deut. 17:16,” in Homeland and Exile: Biblical and Ancient Near Eastern Studies in Honour of Bustenay Oded (ed. G. Galil et al.; VTSup 130; Leiden: Brill, 2009), 271–296. There are, however, indications that the whole text is a post-Dtr insert that wants to introduce into the Pentateuch the possibility of the restoration of a Judean monarchy; see R. Achenbach, “Das sogenannte Königsgesetz in Deuteronomium 17,14–20,” ZABR 15 (2009), 216–233; T. Römer, “La loi du roi en Deutéronome 17 et ses fonctions,” in Loi et justice dans la Littérature du Proche-Orient ancien (ed. O. Artus; BZABR 20; Wiesbaden: Harrassowitz, 2013), 99–111. In this case, Deut 17:14–20 would not belong to the Deuteronomistic History. 14

The Problem of the Hexateuch

817

32:13). Equally, the root ‫( כעס‬hiphil, “to offend”) is attested in Deuteronomy and the Prophets but not in the Tetrateuch. For more than two decades, Nothʼs theory has come under heavy attack, especially in European scholarship, and this for different reasons, which I will not present in this contribution.15 Instead of a Deuteronomistic History, several scholars have returned to the idea of a Hexateuch.16 Yet, the question remains what kind of Hexateuch those scholars have in mind.

Hexateuchal Themes in the Pentateuch Some passages in the books of Genesis–Numbers clearly try to create a Hexateuch. For instance, Gen 50:25 and Exod 13:19 deal with the transportation of Josephʼs bones from Egypt to Israel. These verses do not make much sense in the context of the Pentateuch, but they do function well to prepare the reader for Josh 24. Joshua 24:32 thus marks the end of a narrative trajectory that starts in Gen 50:2517 (or even 33:19).18 Genesis 50:25 Then Joseph made the sons of Israel swear, saying, “God will surely take care of you, and you shall carry my bones up from here.” Exodus 13:19 Moses took the bones of Joseph with him, for he had made the sons of Israel solemnly swear, saying, “God will surely take care of you, and you shall carry my bones from here with you.”

 For a presentation and evaluation of the Deuteronomistic controversy, see T. R ömer , “The Current Discussion on the So-Called Deuteronomistic History: Literary Criticism and Theological Consequences,” Humanities 46 (2015), 43–66. 16  E. O tto , Das Deuteronomium im Pentateuch und Hexateuch: Studien zur Literatur­ geschichte von Pentateuch und Hexateuch im Lichte des Deuteronomiumsrahmen (FAT 30; Tübingen: Mohr Siebeck, 2000); R. G. K ratz , “Der vor- und nachpriesterschriftliche Hexateuch,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Dis­kussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 295–323; C. Frevel, “Die Wiederkehr der Hexateuchperspektive: Eine Herausforderung für die These vom deutero­ nomis­tischen Geschichtswerk,” in Das deuteronomistische Geschichtswerk (ed. H.-J. Stipp; ÖBS 39; Frankfurt am Main: Lang, 2011), 13–53. 17  In fact, the suggestion that Joseph was buried in Shechem creates a specific link with the beginning of the Joseph story, as noted by the medieval Jewish commentator Rashi (Rabbi Solomon son of Isaac, 1040–1105): “They [Josephʼs brothers] stole him from Shechem (see Gen 37:13), and they [Joshuaʼs generation] returned him to Shechem”; see Rashi on Josh 24, translated by Marc Brettler, in T. Römer and M. Z. Brettler, “Deuteronomy 34 and the Case for a Persian Hexateuch,” JBL 119 (2000), 401–419, here 410. 18  E. B lum , Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 44–45. 15

818

Thomas Römer

Joshua 24:32 Now they buried the bones of Joseph, which the sons of Israel brought up from Egypt, at Shechem, in the piece of ground which Jacob had bought from the sons of Hamor the father of Shechem for one hundred pieces of kesita; and they became the inheritance of Josephʼs sons.

The idea that these passages belong to the same literary level could garner broad agreement. The question is whether they belong to any old sources. Genesis 50:25 and Exod 13:19 have traditionally been attributed to E, which would mean that Josh 24:32 also belongs to E. But, as is often observed, it is difficult to reconstruct an E account of the passage through the sea. Furthermore, Gen 50:25 may not belong to the original Joseph story, which was probably conceived as an independent narrative before it was incorporated into the end of the book ָ ‫ ְק ִשׂ‬means “money,” as advocated in most commentaries of Genesis. If ‫יטה‬ and translations, then the text of Josh 24 could hardly be older than the fifth century, because it was only at that time that people started to use coins in Palestine.19 Therefore, it is not astonishing that European scholars like E. Blum and J. C. Gertz tend to ascribe these verses to late redactors.20 A further Hexateuchal theme is the promise of land to Caleb in Num 13–14. The promise that he will enter the land together with Joshua and receive a territory is related to Deut 1:35–36 and Josh 14:13–14, where the territory around Hebron is allotted to him. Numbers 14:23–24 “They shall by no means see the land, which I swore to their fathers, nor shall any of those who spurned me see it. But my servant Caleb, because he has had a different spirit and has followed me fully, I will bring into the land which he entered, and his descendants shall take possession of it.” Deuteronomy 1:35–36 “Not one of these men, this evil generation, shall see the good land which I swore to give your fathers, except Caleb the son of Jephunneh; he shall see it, and to him and to his sons I will give the land on which he has set foot, because he has followed YHWH fully.” Joshua 14:13–14 So Joshua blessed him and gave Hebron to Caleb the son of Jephunneh for an inheritance. Therefore, Hebron became the inheritance of Caleb the son of Jephunneh the Kenizzite until this day, because he followed YHWH, the God of Israel, fully.  E. A. Knauf, Josua (ZBK 6; Zürich: Theologischer Verlag, 2008), 200.  E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1990), 363–365, and J. C. G ertz , Tradition und Redaktion in der Exodus­erzählung: Unter­ suchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 1999), 380, 389. See also M. W itte , “Die Gebeine Josefs,” in Auf dem Weg zur End­gestalt von Genesis bis II Regum: Festschrift für Hans-Christoph Schmitt zu seinem 65. Ge­burts­tag (ed. M. Beck and U. Schorn; BZAW 370; Berlin: de Gryuter, 2006), 139–156, who argues that these very late texts reflect the transport of Alexanderʼs corpse. 19

20

819

The Problem of the Hexateuch

In contrast to the theme of Josephʼs bones, it is more difficult to decide whether these verses all belong to the same literary level. The decision hinges on how one understands the relationship between the parallel texts in Numbers and in Deuteronomy. Joshua 14 may have been added by a redactor already aware of the tradition about Caleb in Numbers and Deuteronomy. Indeed, many commentators agree that Josh 14:1–15* does not belong to the same literary level as either Num 14 or Deut 1.21 There also exists a strong connection between Exod 16 and Josh 5. Exodus 16:35 relates the beginning of the divine gift of manna. This narrative opens a period that ends only after the entry into the land, as stated in Josh 5:12: “The manna ceased the day they ate the produce of the land.” Exodus 16 is at the earliest a P text,22 so the note in Josh 5 cannot be earlier. Interestingly, the whole of Joshua 5 takes up themes from the book of Exodus. Joshuaʼs vision in Josh 5:13–1523 creates a parallel with Mosesʼs call in Exod 3:5: Joshua 5:15 ‫ַשל־נַ ַע ְלָך ֵמ ַעל ַרגְ ֶלָך‬ ‫ִכי ַה ָמקֹום ֲא ֶשר ַא ָתה ע ֵֹמד ָע ָליו ק ֶֹדׁש הּוא‬ Take off your sandal from your foot. Indeed, the place where you are standing is holy.

Exodus 3:5 ‫ַשל־נְ ָע ֶליָך ֵמ ַעל ַרגְ ֶליָך‬ ‫עֹומד ָע ָליו ַא ְד ַמת־ק ֶֹדׁש הּוא‬ ֵ ‫ִכי ַה ָמקֹום ֲא ֶשר ַא ָתה‬ Take off your sandals from your feet. Indeed, the place where you are standing is holy ground.

It is not clear which text depends on which, but it is clear that this intertextuality makes Joshua appear as a new Moses. The note about the circumcision of the children of the exodus generation reads like a midrash on Exod 12:43–50. If one considers Josh 3–4 to be related to the crossing of the sea (Exod 14), then one may even construct a sort of chiasm, exposing how the redactors of the first chapters of Joshua wanted to correlate the conquest account with the exodus story:24  V. Fritz, Das Buch Josua (HAT 1/7; Tübingen: Mohr Siebeck, 1994), 152–154, attributes v. 13 to a “D-redactor” (later than the Dtr edition of Joshua) and v. 14 to a later addition. According to Otto, Das Deuteronomium (see n. 16), 79–84, and Knauf, Josua (see n. 19), 20–21, 138–140, the passage belongs to a hexateuchal redaction. 22  For J. S. Baden, “The Original Place of the Priestly Manna Story in Exodus 16,” ZAW 122 (2010), 491–504, Exod 16 as a whole belongs to the P document. L. S chmidt , “Die Priesterschrift in Exodus 16,” ZAW 119 (2007), 483–498, considers only vv. 1aß, 2f, 9–14bα, 15, 21, 11aαb, 23, 24a, 25f, and 35a as belonging to P; similarly, R. Albertz, Exodus 1–18 (ZBAT 2/1; Zürich: Theologischer Verlag, 2012), 256–278. 23  For this passage, see T. Römer, “Joshuaʼs Encounter with the Commander of Yhwhʼs Army (Josh 5:13–15): Literary Construction or Reflection of a Royal Ritual?,” in Warfare, Ritual, and Symbol in Biblical and Modern Contexts (ed. B. Kelle et al.; AIL 18; Atlanta: Society of Biblical Literature, 2014), 49–63, especially 52–55. 24  See also K. Bieberstein, Josua-Jordan-Jericho: Archäologie, Geschichte und Theolo­ gie der Landnahmeerzählungen Josua 1–6 (OBO 143; Göttingen: Vandenhoeck & Ruprecht, 1995), 418. 21

820

Thomas Römer

A Divine revelation to Moses (Exod 3)  B Passover (Exod 12:1–28)   C Circumcision for the Passover (Exod 12:43–50; see also 4:24–26)    D Crossing of the Sea (Exod 14)      Sinai and wilderness    D′ Crossing of the Jordan (Josh 3–4)   C′ Circumcision before the Passover (Josh 5:2–9)  B′ Passover (Josh 5:10–12) A′ Divine revelation to Joshua (Josh 5:13–15) It is possible that the episodes in Josh 5 belong to a hexateuchal redaction.25 In this context, the appearance of the divine warrior in 5:13 can also be understood as fulfilling the promise made in Exod 23:20: “I am going to send an angel in front of you, to guard you on the way and to bring you to the place that I have prepared.” Its aim is to connect the book of Joshua as narrowly as possible with the preceding Pentateuch.26 It does not, therefore, seem very plausible to find old pentateuchal sources in Josh 5. Rather, we witness here the activity of redactors who indeed wanted to construct a Hexateuch. So far, it seems difficult to reconstruct older sources in the book of Joshua. Before we pronounce a final judgment on the question, however, let us examine the question of P in Joshua as well as the question of the different endings of the book.

The End of P in Joshua? Although Wellhausen was convinced of the existence of a Hexateuch, he had some difficulty reconstructing the original P (or Q) source in the book of Joshua. He tentatively considered 18:1 to belong to P. Its original place would have been before Josh 14.27 The reorganization of the text of Joshua by Wellhausen did not meet with much success, but the idea of 18:1 as a possible end for P did. In the current discussion concerning the end of P (which cannot be summarized

25  Knauf, Josua (see n. 19), 63–67, attributes vv. 10–12 to P (see below) and vv. 1–10 and 13–15 to a late hexateuchal redaction that, because of the idea of a second circumcision cannot be older than the second century BCE (p. 64). For a hexateuchal perspective on Josh 5, see also E. Blum, “Beschneidung und Passa in Kanaan: Beobachtungen und Mutmaßungen zu Jos 5,” in Freiheit und Recht: Festschrift für Frank Crüsemann zum 65. Geburtstag (ed. C. Hardmeier et al.; Gütersloh: Kaiser, 2003), 292–322. 26  H. A usloos , “The Book of Joshua, Exodus 23 and the Hexateuch,” in The Book of Joshua (ed. E. Noort; BETL 250; Leuven: Peeters, 2012), 259–266. 27  Wellhausen, Composition (see n. 11), 127–129.

The Problem of the Hexateuch

821

in this context),28 some scholars advocate a modified Wellhausenian solution. E. A. Knauf,29 closely following N. Lohfink,30 reconstructs P in Joshua as follows: Joshua 4:19* The people came up from the Jordan on the tenth of the first month. Joshua 5:10–12 The sons of Israel camped at Gilgal and they observed the Passover on the evening of the fourteenth day of the month on the desert plains of Jericho. On the day after the Passover, on that very day, they ate some of the produce of the land, unleavened cakes and parched grain. The manna ceased on the day after they had eaten some of the produce of the land, so that the sons of Israel no longer had manna, but they ate some of the yield of the land of Canaan during that year. (Joshua 14:1–2* Now these are the territories which the sons of Israel inherited in the land of Canaan, which Eleazar the priest, and Joshua the son of Nun, and the heads of the households of the tribes of the sons of Israel apportioned to them for an inheritance, by the lot of their inheritance.) Joshua 18:1 Then the whole congregation of the sons of Israel assembled themselves at Shiloh and set up the tent of meeting there; and the land was subdued before them. (Joshua 19:51 These are the inheritances that Eleazar the priest, and Joshua the son of Nun, and the heads of the households of the tribes of the sons of Israel distributed by lot in Shiloh before the LORD at the doorway of the tent of meeting. So they finished dividing the land.)31 (Joshua 24:29b Joshua the son of Nun, the servant of YHWH, died, being one hundred and ten years old.)32

This reconstruction confronts us with a very short P account, quite different from what we can find in the books of Genesis and Exodus. Two main arguments support the idea of P ending in Joshua: First, since YHWH promises the gift of the land in Exod 6:8 (P), this gift must also be narrated. Second, only Josh 18 brings the program of Gen 1 to an end; the verb ‫“( כבש‬to subdue”) appears in Josh 18:1 and in Gen 1:28, creating an inclusion that signifies the fulfillment of Godʼs 28  See the discussion in C. N ihan , From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 20–68. 29  Knauf, Josua (see n. 19), 20. 30  N. L ohfink , “The Priestly Narrative and History,” in Theology of the Pentateuch: Themes of the Priestly Narrative and Deuteronomy (trans. L. M. Mahoney; Edinburgh: T&T Clark, 1994), 136–172. 31  Josh 14:1–2* and 19:51 are attributed to P by Lohfink but not by Knauf. 32  This half verse is attributed to P only by Knauf, Josua (see n. 19), 20, although even he wonders whether this verse belongs to P or to a later Priestly redaction.

822

Thomas Römer

original order after the Israelites enter the land. This view is not convincing, however. The command in Gen 1:28 is addressed to all humankind and defines its role in creation, whereas Josh 18:1 is about Israel and its land. Genesis 1 depicts an ideal creation, not the world in which humankind lives. This world is established after the flood, where the command to subdue the earth is no longer part of the divine order given to Noah (Gen 9:1–5). The mention of Shiloh also does not fit with the idea that 18:1 should be the ending of P.33 Shiloh is mentioned here for the first time and will appear often in the following books of the Former Prophets, especially in Samuel. This first mention of Shiloh does not make sense if Josh 18:1 is supposed to be a conclusion. Josh 18:1 should, therefore, not be considered as the conclusion of a P document starting in Gen 1. More likely, this verse belongs to late Priestly additions to the so-called Deuteronomistic History, which may reach as far as the books of Kings.34 Joshua 18:1 takes up Num 32:22–29 (note the use of the ‫ כבש‬root in vv. 22 and 29)35 and prepares for 2 Sam 8:11, where the root ‫כבש‬ reappears in the statement that all nations were subdued by David.

Different Endings in the Book of Joshua36 It has sometimes been observed that the book of Joshua contains several passages or verses that sound like a conclusion. Can we detect in one of those conclusions the end of an old Hexateuchal narrative? (a) Joshua 10:42: Joshua captured all these kings and their land at one time because YHWH, the god of Israel, fought for Israel.

According to Knauf, the passage 10:40–42* concludes the oldest conquest account, which was part of an “exodus and conquest narrative,” with a possible beginning in Exod 2.37 The statement “YHWH fought for Israel” does indeed 33  As R. D. N elson , Joshua: A Commentary (OTL; Louisville: Westminster John Knox, 1997), 209, rightly points out, the emphasis on Shiloh is not a specific Priestly feature. In the so-called Deuteronomistic History, “Shiloh was viewed as the legitimate forerunner to Jerusalem.” 34  R. Achenbach, “Der Pentateuch, seine theokratischen Bearbeitungen und Josua–2 Kö­ nige,” in Les dernières rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (ed. T. Römer and K. Schmid; BETL 203; Leuven: Peeters, 2007), 225–253. 35  Fritz, Das Buch Josua (see n. 21), 179–180. 36  This chapter is a summarized and somewhat modified version of T. Römer, “Book-Endings in Joshua and the Question of the So-Called Deuteronomistic History,” in Raising Up a Faithful Exegete: Essays in Honor of Richard D. Nelson (ed. K. L. Noll and B. Schramm; Winona Lake, IN: Eisenbrauns, 2010), 85–99. 37  Knauf, Josua (see n. 19), 17.

The Problem of the Hexateuch

823

have a parallel in Exod 14:14, but is this enough to postulate a narrative reaching from Moses to Joshua? The ideology of YHWH fighting for Israel and the delimitation of the conquered land from Kadesh-Barnea to Gibeon in Benjamin38 may well fit a seventh-century BCE setting for the passage, probably under the rule of Josiah. But Josh 10:40–42 may also be considered the “summation of southern conquests”39 related in Josh 10:28–42 and thus not correspond to the conquest account in Josh 3–10*. As a matter of fact, the mention of the kings fits better with 10:28–39 than with the preceding stories, which are not centered on foreign monarchs. Furthermore, 10:42 does not speak of the conquest of the whole land (contrary to 11:23 and 21:43), only of the land controlled by the kings mentioned in chapter 10. (b) Joshua 11:23: Joshua took the entire land according to all that YHWH had spoken to Moses; and Joshua gave it as an inheritance to Israel according to their tribal allotments by their tribes. And the land was at rest from war.

This verse does indeed sound like a conclusion. It refers to YHWHʼs initial speech to Joshua in 1:1–9*, with references to the conquest of the land (1:2), the distribution of the land as a ‫נחלה‬, and the former promises made to Moses (1:3). These links indicate that we have here a Deuteronomistic text, which suggests that the distribution of the land has already taken place, although the repartition of the land to the tribes is related only in chapters 13–19. It is therefore possible that the concluding remark in 11:23 reflects a stage of the formation of Joshua in which the list material contained in 13–19 did not yet exist.40 U. Becker suggested that 11:23 was the original conclusion of the book and was followed by the report of Joshuaʼs death in Judg 2:8–9 (or Josh 24:29–30).41 The language of 11:23 is clearly Deuteronomistic, although the last phrase, “and the land had rest from war,” does not appear often in the Deuteronomistic History. In the Former Prophets, the verb ‫ שקט‬with ‫ ארץ‬as subject occurs only in the framing remarks of the book of Judges: 3:11, 30; 5:31; 8:28.42 In contrast to Judg 2:20, Josh 11:23 expresses the idea that Joshua conquered the whole (‫ )כל‬land. The same idea occurs in Josh 21:43–45. Of course, one cannot exclude the possibility that the passage in 11:16–23* preserves traces of an ending of a pre-Deuteronomistic 38  The mentions of Gaza and the unidentifiable “land of Goshen” in 10:41 may be later additions, as argued by most commentators. 39  Nelson, Joshua (see n. 33), 138. 40  Nelson, Joshua (see n. 33), 164. 41  U. Becker, “Endredaktionelle Kontextvernetzungen des Josua-Buches,” in Die deutero­ nomisti­schen Geschichtswerke: Redaktions- und religionsgeschichtliche Perspektiven zur ‘Deu­tero­nomismus’-Diskussion in Tora und Vorderen Propheten (ed. M. Witte et al.; BZAW 365; Berlin: de Gruyter, 2006), 139–161, here 151–152. 42  With other subjects in Judg 18:7, 27 (‫ )עם‬and 2 Kgs 11:20 (‫)עיר‬. Its main occurrences are in Isaiah, Jeremiah, and Chronicles.

824

Thomas Römer

story heavily reworked at a later time. Be that as it may, to reconstruct this older story is a difficult, if not impossible, task. (c) Joshua 21:43–45 and Josh 23. Joshua 21:43–45 provides a clear Deuteronomistic conclusion, with the idea that everything is fulfilled. These verses introduce the original version of Josh 23, which shares the same idea. As I have tried to demonstrate elsewhere, the original account of Josh 23 (verses 1–3*, 9, 11, 14b–16a) has been reworked by a later Deuteronomistic redactor.43 The revision of Josh 23, which took place at the end of the sixth or the beginning of the fifth century BCE, introduces the idea that YHWH did not expel all of Israelʼs enemies and thus prepares the transition to the book of Judges (Josh 23 and Judg 2:6–22*). As Wellhausen already observed, it is likely that Josh 23 is a Dtr composition.44 (d) Joshua 24. One may also agree that Joshuaʼs second farewell speech in chapter 24 is not a Deuteronomistic text. If so, the question arises whether Josh 24 is an older text and would therefore be the ending of a pre-Dtr source, as Wellhausen stated. He was inclined to ascribe Josh 24* to E.45 In recent European scholarship, there is a trend, following the older works of Van Seters and Anbar,46 to consider Joshuaʼs speech in 24:3–14 a post-Dtr creation that apparently presupposes P.47 The closest parallel for Joshuaʼs speech occurs in Neh 9, a text from the middle of the Persian period at the earliest, 48 with the following elements in common: 43  T. R ömer , “Das doppelte Ende des Josuabuches: Einige Anmerkungen zur aktuellen Diskussion um ‘deuteronomistisches Geschichtswerk’ und ‘Hexateuch,’ ” ZAW 118 (2006), 523–548. 44  Wellhausen, Composition (see n. 11), 133. 45  Wellhausen, Composition (see n. 11), 133–134. 46  J. Van Seters, “Joshua 24 and the Problem of Tradition in the Old Testament,” in In the Shelter of Elyon: Essays on Ancient Palestinian Life and Literature in Honor of G. W. Ahlström (ed. W. B. Barrick and J. R. Spencer; JSOTSup 31; Trowbridge: JSOT Press, 1984), 139–158; M. A nbar , Josué et lʼalliance de Sichem (Josué 24:1–28) (BBET 25; Frankfurt am Main: Lang, 1992). 47  Cf. Exod 14:9 (P) and Josh 24:6; Exod 14:10 (P); and Josh 24:7; and the expression ‫ל־א ֶרץ ְּכנַ ַען‬ ֶ ‫ ָּכ‬in Gen 17:8 (P) and Josh 24:3. For a post-Priestly date of the speech, see, among many others, E. A urelius, “Zur Entstehung von Josua 23–24,” in Houses Full of All Good Things: Essays in Memory of Timo Veijola (ed. J. Pakkala and M. Nissinen; Publications of the Finnish Exegetical Society 95; Göttingen: Vandenhoeck & Ruprecht, 2008), 95–114. Carrʼs theory that Josh 24 was a pre-P text that underwent “priestly washing” is a circular argument; see D. M. C arr , “Strong and Weak Cases and Criteria for Establishing the Post-Priestly Character of Hexateuchal Material,” in The Post-Priestly Pentateuch: New Perspectives on Its Redactional Development and Theological Profiles (ed. F. Giuntoli and K. Schmid; FAT 101; Tübingen: Mohr Siebeck, 2015), 19–34. 48  M. Oeming, “ ‘See, We Are Serving Today’ (Nehemiah 9:36): Nehemiah 9 as a Theological Interpretation of the Persian Period,” in Judah and the Judeans in the Persian Period (ed. O. Lipschits and M. Oeming; Winona Lake, IN: Eisenbrauns, 2006), 571–588.

825

The Problem of the Hexateuch

Abraham Fathers – Egypt– Cry – Sea Miracle at the sea Wilderness Enemies “given in the hand” Enemies as “king” List of the “goods of the land” Sondergut

Joshua 24 3 (+ Isaac, Esau, Jacob) vv. 6–7 v. 7 v. 7 v. 8 v. 12 v. 12 Balaam vv. 9–10 Crossing of the Jordan v. 11

Nehemiah 9 vv. 7–8 v. 9 v. 11 v. 19 v. 24 vv. 22, 24 v. 25 Creation v. 6 Sinai and rebellion vv. 13–20

Therefore, it seems difficult to find traces of older sources in Joshuaʼs speech. The same holds true for the following ceremony. It has often been observed that chapter 24 depicts Joshua as a second Moses:49 like Moses, he concludes a covenant; like Moses, he enacts laws and decrees (v. 25); like Moses, he raises a stone; and, like Moses, he writes a scroll (v. 26: “and Joshua wrote all the words in the scroll of the law of God”). The rare expression ‫ספר תורת אלהים‬ (see also Neh 8:18) was possibly coined as an alternative to the term ‫תורת משה‬, which in the Persian period became a name for the nascent Pentateuch. The fact that Moses is not mentioned in the original form of Joshuaʼs historical recapitulation50 may be explained by the emphasis that the author wants to put on Joshua. The same holds true concerning the absence of the Sinai lawgiving, since Joshua promulgates the law at the end of chapter 24. The statement that Joshua establishes a statute and ordinance for the people (‫ )חק ומשפט‬creates a parallel between Joshua and Ezra (see Ezra 7:10). It offers an alternative to the giving of the law on Mount Sinai.51 Joshua 24 thus reflects a discussion in the Persian period about whether the Torah52 should end with the book of Deuteronomy or should also include the book of Joshua. The very different end of this chapter in the LXX,53 apparently also known in Qumran (even if the text was later reworked),54 perhaps reflects the original ending of the Hexateuch, since it emphasizes the role of Joshua even more.55 The location of this chapter in Shechem and the pan-Israelite perspec For the following, see Römer and Brettler, “Deuteronomy 34” (see n. 17), 401–419.  Verse 5a is a gloss. 51  Schmid, Erzväter (see n. 9), 228. 52  For an overview of the current understanding of the promulgation of the Torah, see G. N. K noppers and B. M. L evinson (eds.), The Pentateuch as Torah: New Models for Understanding Its Promulgation and Acceptance (Winona Lake, IN: Eisenbrauns, 2007). 53  A. Rofé, “The End of the Book of Joshua according to the Septuagint,” Hen 4 (1982), 17–36. 54  B. Lucassen, “Josua, Richter und CD,” RevQ 18 (1998), 373–396. 55  See 24:31 (LXX), where Joshua is buried together with “the stone swords by which he circumcised the sons of Israel in Galgal, when he (!) led them out from Egypt,” quoted from the English translation of A. G. Auld, Joshua: Jesus Son Of Naué in Codex Vaticanus (Septuagint 49 50

826

Thomas Römer

tive of Josh 24 (v. 1 mentions “all tribes”) can be understood as an attempt to integrate the Samaritan Yahwists.56 Nevertheless, the alternatives of Pentateuch or Hexateuch reflect different perspectives on Israelʼs identity. In a Hexateuch, the Torah is linked with the conquest of the land, whereas in the Pentateuch, Moses dies without entering the land. The Pentateuch better fits a diaspora perspective. It states explicitly that the land is part of the promise but that it is not necessary to listen to and to live according to the Torah. For that reason, the final cut was made after the book of Deuteronomy and Joshua became a sort of deuterocanonical book (see Josh 1:8). In sum, Josh 24 indeed creates a Hexateuch, but only in the middle of the Persian period as an alternative to the Torah. The last chapters of the book of Joshua, therefore, do not allow the reconstruction of older pentateuchal sources.

Exodus and Eisodus G. von Rad, who defended the existence of a Yahwistic Hexateuch from the time of Solomon, never tried to demonstrate precisely which texts in Joshua could be attributed to J, E, or P.57 His argument was based on the so-called historical credo in Deut 26:5–9, in which the exodus and the possession of the land cannot be separated. In fact, several scholars defend the idea that the oldest exodus narrative must have ended with an account of the entrance or the conquest of the land. This might be the case, but we also have to ask ourselves whether we can still reconstruct this oldest exodus-conquest account. In a way, M. Noth proposed a way out when he postulated that the conquest stories of the older sources were deleted when P, which according to him had no conquest account, became the framework of the Pentateuch.58 This solution has always been criticized. Yet, we have to admit that the idea that all older sources were religiously kept by the later redactors is anachronistic. It presupposes the idea of a canonical or sacred text, which did not exist before the work of the Masoretes.

Commentary Series; Leiden: Brill, 2005), 83. For the complex history of transmission of the LXX conclusion of Joshua, see M. Rösel, “The Septuagint-Version of the Book of Joshua,” SJOT 16 (2002), 5–23. 56  C. N ihan , “The Torah between Samaria and Judah: Shechem and Gerizim in Deuteronomy and Joshua,” in The Pentateuch as Torah: New Models for Understanding Its Promulgation and Acceptance (ed. G. N. Knoppers and B. M. Levinson; Winona Lake, IN: Eisenbrauns, 2007), 187–223, here 197–199. 57  G. von Rad, Gesammelte Studien zum Alten Testament (TB 8; Munich: Kaiser, 1971), 9–86; English translation, idem, The Problem of the Hexateuch and Other Essays (London: SCM, 1984), 1–78. 58  Noth, Überlieferungsgeschichtliche Studien (see n. 13), 210–211.

The Problem of the Hexateuch

827

Conclusion Although there is narrative continuity between the books of the Pentateuch and Joshua (Hexateuch) and even across the entire Former Prophets (Enneateuch), there is hardly any evidence of the classic pentateuchal sources in the Former Prophets. The Former Prophets are closely related to the book of Deuteronomy and may therefore still be considered Deuteronomistic. On the literary level, it is impossible to find older sources in the Former Prophets. The links between the Pentateuch and Joshua, as well as with the following books, do not reflect sources that begin somewhere in the Pentateuch but are the work of different redactors who, with different agendas, tried to correlate the pentateuchal scrolls with those of the Former Prophets.

Part Nine Rethinking the Relationship between the Law and the Prophets

Introduction Dalit Rom-Shiloni The relationship between the Law and the Prophets suggests outside, or indirect, perspectives on some of the major issues that occupy pentateuchal studies. This part of the book brings together twelve articles, written by scholars of prophetic literature who have specialized in various intertextual relationships between the two corpora, the Pentateuch and the prophetic literature (or specific compositions within them). Furthermore, their experience has allowed these scholars to reflect on the implications of their findings for the broader issues at stake. As expected, this variety of studies illustrates different points of departure on both the Pentateuch and its evolution and the prophetic literature and its literary growth, and inevitably the articles arrive at very different conclusions. The common denominator we seem to share is the understanding that both ends were not created in isolation from each other; rather, interaction is the name of the game. And yet, the discussions below reflect our scholarly struggles over the actual descriptions of the literary connections between this or that tradition or text, over the directions of influence, and over the relative datings of the invoked traditions or texts. This part opens with a general discussion of “the Law and the Prophets” or vice versa and then moves on to papers that give closer attention to Jeremiah and the Twelve, addressing Hosea, Amos, Micah, Zech 9–14, and Malachi in particular. The last four papers are the product of a workshop on Lev 26 and Ezekiel (and see Kesslerʼs paper for the relationship between Lev 26 and Amos). Each of the articles (and, not least, the collection as a whole) addresses, and oftentimes challenges, major scholarly assumptions on both sides of the question, from the standpoint of the prophetic literature and from that of the Pentateuch, in hopes of shaking longstanding scholarly assumptions. Inescapable in the study of innerbiblical allusion and exegesis, the problem of direction of influence is at the forefront of several of the papers. To introduce this part, I would like to address some methodological issues that all of its authors have either implicitly based their studies on or explicitly considered. Additional comments may clarify the background of the discussion, the directions currently taken, and the major questions still open on the relationship and interaction between the Pentateuch (or pentateuchal materials) and the Prophets.

832

Dalit Rom-Shiloni

1 “The Law and the Prophets,” or Vice Versa? The greatest challenge of all, the relationship between “the Law and the Prophets” – a relationship that was reversed from its canonical order in Hebrew Bible scholarship beginning in the nineteenth century – still seems to be unsolved.1 In his essay, Konrad Schmid provides a detailed history of research, outlining the status of these scholarly debates in biblical scholarship and suggesting a careful solution of possible mutual influence. In contrast to diachronic perspectives that see a linear literary evolution of the two large corpora, examining them side by side opens up new avenues of understanding. Indeed, the literary evolution of both the pentateuchal materials and the prophetic literature took place in fairly similar places (Israel, Judah, Babylon, and then Persian Yehud) and in parallel periods (from the eighth century to the fifth/fourth centuries BCE). Yet, the long and complex literary evolution and transmission of the pentateuchal literature leaves basic questions about the relationship between pentateuchal materials and prophetic literature: Could the two literary circles (and their many authors) that nurtured these literary compositions have been totally isolated or unfamiliar to each other? Or were the authors of the pentateuchal literature aware of prophetic proclamations and writings, and, vice versa, were pentateuchal traditions at the disposal of prophets? What parts of the pentateuchal traditions were known and when? Were they known orally or in writing, or should David Carrʼs option of “writing-supported transmission of texts through memory” ease these difficult questions?2 Finally, looking again at the corpora as a whole, is there any correlation between the growth of the Pentateuch and its formation and the prophetic literature and its no-less complicated evolution?

2 The Dichotomy between Visionaries and Authors (Interpreters) The distinction between visionaries (speakers) and authors (interpreters) is another of those scholarly axioms about prophetic literature introduced in the  For the sake of this discussion, it seems that the two components of this title deserve clarification, as their different understandings may be in the background of the varied scholarly perspectives. “The Law” may stand for the entire edited Pentateuch but also for earlier stages of its literary evolution and thus refer to smaller-scale compositions. “The Prophets” may stand for the prophetic literature of the Latter Prophets (though traditionally “the Prophets” may include the Former Prophets as well, and possibly the entire Hebrew Bible beyond “the Law”). For this much-contested discussion in scholarship, see J. B arton , Oracles of God: Perceptions of Ancient Prophecy in Israel after the Exile (London: Darton, Longman and Todd, 1986), 35–95. 2  D. M. C arr , The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011), 22–34, 90–98. 1

Introduction

833

nineteenth century, mainly by G. H. A. Ewald and followed by A. Kuenen and J. Wellhausen, that still influence scholars.3 To quote Ewald in but one of his references to these later prophets, It is the new race of prophets which arose suddenly towards the end of the Babylonian captivity who were the first to weave many passages of older prophets into their own writing, and artistically reproduce them; but they had not laboured publicly as the earlier prophets had done, and had, from the very first, more of the character of prophetic writers. However, the prophets from this period generally never rose again to the independent vigour of the earlier ones.4

Thus was a divide established between the classical prophets down to Jeremiah, in the late seventh and early sixth centuries BCE, and the prophets starting with Ezekiel in the early sixth century and continuing with Deutero-Isaiah, Haggai, and Zechariah in the late sixth and fifth centuries BCE.5 The differentiation was connected to the presumed decline of prophetic spirit (or inspiration), culminating in the complete loss of prophecy in Second Temple Judaism, a decline that left its mark in the diverse reuses of earlier prophetic traditions and conceptions by later prophets, who were deemed mere epigonis – imitators, interpreters, authors, or scribes.6 3  G. H. A. E wald , Prophets of the Old Testament (trans. J. F. Smith; 5 vols.; Edinburgh: Williams and Norgate, 1875–1881). Ewaldʼs Die Propheten des alten Bundes was first published in 1840–1841 (the second edition of 1867–1868 was translated into English between 1875 and 1881) and had significant influence on scholarship, particularly on J. Wellhausen, Prolegomena to the History of Israel (trans. J. S. Black and A. Menzies; Edinburgh: A&C Black, 1994), 392–404, as also on A. Kuenen (see below), both of whom were among Ewaldʼs many distinguished students. 4  Ewald, Prophets of the Old Testament (see n. 3), 1:82–83. 5  Compare Ewaldʼs words of praise to Jeremiah (Prophets of the Old Testament [see n. 3], 3:59–65) and, in contrast, his ambivalent approach to Ezekiel, whom he found to be “the last outburst of the prophetic activity in its ancient manner” (4:9) but the first to labor more as author (than publicly). Ezekiel, according to Ewald (4:7–17), gradually lost his spiritual freedom to utilize passages of the Pentateuch, and his household limitations affected his imagination and message. This decline developed more broadly, and by the Persian period prophets gave way to “the ancient law” (5:35–36). Compare A. K uenen , Ezekiel (London: W. Speaight, 1884), 1–25, who admires Ezekiel as “the first designer and [. . .] the father of Judaism” and of “Judaic legalism” (25), who established a school from which, in the matter of a century, sprang the priest and scribe Ezra. 6  W ellhausen , Prolegomena (see n. 3), 402–404, argued that these later writers were aware of their position outside of the prophetic tradition, and he based his assumption on Lam 2:9; Isa 43:11; and Ps 74:11, concerning lack of prophecy during the exilic period. See, for example, Wellhausenʼs words on Ezekiel: “We may call Jeremiah the last of the prophets. Those who came after him were prophets only in name. Ezekiel had swallowed a book (iii, 1 – 3), and gave it out again. He also, like Zechariah, calls the pre-exilic prophets the old prophets, conscious that he himself belongs to the epigoni: he meditates on their words like Daniel and comments on them in his own prophecy (xxxviii, 17; xxxix, 8)” (403).

834

Dalit Rom-Shiloni

Furthermore, this decline in prophecy was bluntly illustrated by the supposed growing prophetic interest, from Ezekiel onward, in legalistic and cultic issues.7 The scholarly assumption was (and to an extent still is) that the late prophets had gradually become involved in the presumed priestly and Levitical activity of the Persian period, activity that included interpretation of pentateuchal (mostly priestly) traditions and that was, as J. Lindblom observed, increasingly interested in issues of purity, holiness, the spirit of God, the temple, sacrifices, fasting, the Sabbath, and the cult and law in general.8 Ezekiel was of course considered a major contributor to this tendency, and references to pentateuchal legal materials (throughout the prophetic literature) were thus taken as evidence of a late, Persian period (and Deuteronomistic) conception of the prophetic role as “torah-teaching.”9 Over the last four decades, the study of the Latter Prophets – Jeremiah, Ezekiel, Isaiah, and the Twelve in general, and specifically the last prophets – has opened new venues. What was formerly considered a great failing of Haggai and Zechariah, for instance, has turned out to be one of their strengths, thanks to the work of P. Ackroyd, R. Mason, J. Tollington, R. Nurmela, J. Kessler, M. J. Boda, M. Stead, and J. Wöhrle, among others.10 These scholars paid attention to the  Following Wellhausen, this has become a common scholarly observation; see, for instance, G. von R ad , Old Testament Theology (trans. D. M. G. Stalker; 2 vols.; Edinburgh: Harper Collins, 1965), 2:278–300, who seems to have struggled with this distinction in spirituality among prophets. Nevertheless, von Rad did draw very clear distinctions between Jeremiah, Ezekiel, and Deutero-Isaiah, whom he admired, and the Persian period prophets, whom he treated as those who had done their best under rough social circumstances but who had not inspired great prophecy (278–282). When discussing Haggai and Zechariah, von Rad further accentuated “the different spiritual condition of the people to whom the later prophets were sent” (281), a condition in which the prophets could only wrestle with the people to broaden their religious beliefs, which they did using their predecessorsʼ great legacy (282, 292–297). See R. A. Mason, “The Prophets of the Restoration,” in Israelʼs Prophetic Tradition: Essays in Honour of Peter Ackroyd (ed. R. J. Coggins et al.; Cambridge: Cambridge University Press, 1982), 137–154. 8  J. Lindblom, Prophecy in Ancient Israel (Oxford: Blackwell, 1962), 403–422, esp. 404, 409–410, 421. Lindblom follows Wellhausen (see above) when he concludes that “[i]t must, however, be recognized that in spite of new elements in their teaching the post-exilic prophets may not unfairly be described as epigoni” (421). 9  See C. M aier , Jeremia als Lehrer der Tora: Soziale Gebote des Deuteronomiums in Fortschreibungen des Jeremiabuches (FRLANT 196; Göttingen: Vandenhoeck & Ruprecht, 2002), 353–372; and consult R. Achenbachʼs article in this volume. 10  P. R. A ckroyd , Exile and Restoration: A Study of Hebrew Thought of the Sixth Century B.C. (Philadelphia: Westminster, 1968), 138–217; R. A. M ason , The Books of Haggai, Zechariah and Malachi (CBC; Cambridge: Cambridge University Press, 1977); J. E. Tollington, Tradition and Innovation in Haggai and Zechariah (JSOTSup 150; Sheffield: Sheffield Academic Press, 1993); R. N urmela , Prophets in Dialogue: Inner-Biblical Allusions in Zechariah 1–8 and 9–14 (Åbo: Åbo Akademis Förlag, 1996); M. J. B oda , “Zechariah: Master Mason or Penitential Prophet,” in Yahwism after the Exile: Perspectives 7

Introduction

835

intertextual and, even more, the allusive and interpretive qualities of these prophetic collections, refuting longstanding nineteenth- and twentieth-century notions about the decline in prophetic quality. Moreover, they saw the allusive-interpretive dimensions of prophecy as characteristic of prophetic writing, not limited to the arena of prophetic traditions and vast in their utilizations of biblical literature, particularly of pentateuchal materials. But several questions come to mind and still seem to await attention concerning this dichotomy between visionaries and authors. The first addresses the genuine concept of prophecy that this dichotomy reveals: Is writing (interpreting) a substitute for vision? Are there valid reasons to explain the differences between preexilic prophets (and prophecy) and exilic/postexilic ones according to this paradigm? A second question concerns time: Is writing (interpreting), specifically of pentateuchal literary and legal traditions, a late substitute for (or addition to) vision, such that it entered prophetic literature only in the Persian period? What more should be said of this assumed transition from the last of the visionaries to the first of the authors, that is, Jeremiah and Ezekiel, and their utilizations of pentateuchal materials? Finally, a particularly bothersome question concerns the function of allusion and exegesis in prophetic literature. Once the phenomenon is recognized as shared by prophetic writings at least as of Jeremiah, what do those allusions to pentateuchal materials contribute to the prophetic message? And vice versa, assuming the direction of influence is the other way around, how do allusions to prophetic literature serve in their pentateuchal contexts? These questions bring us face to face with the very essence of prophecy and prophetic writings, and they are especially intriguing once scholars allow themselves to consider interconnections between the prophetic literature and pentateuchal materials.

on Israelite Religion in the Persian Era (ed. R. Albertz and B. Becking; Studies in Theology and Religion 5; Assen: Van Gorcum, 2003), 49–69; idem , Haggai, Zechariah (NIV Application Commentary Series; Grand Rapids, MI: Zondervan, 2004); J. K essler, The Book of Haggai: Prophecy and Society in Early Persian Yehud (Leiden: Brill, 2002); M. R. Stead, The Intertextuality of Zechariah 1–8 (LHBOTS 506; New York: T&T Clark, 2009); J. Wöhrle, “So Many Cross-References! Methodological Reflections on the Problem of Intertextual Relationships and Their Significance for Redaction Critical Analysis,” in Perspectives on the Formation of the Book of the Twelve: Methodological Foundations, Redactional Processes, Historical Insights (ed. R. Albertz et al.; Berlin: de Gruyter, 2012), 3–20. To mark the significant change, M ason , “Prophets of the Restoration,” (see n. 7), 137–154, accentuated that allusion and exegesis served as prominent literary tools within late prophecy, tools that show “originality and inventiveness” (142).

836

Dalit Rom-Shiloni

3 Interaction In the framework of literary analysis, interaction denotes any literary exchange, explicit or implicit, between sources within the Pentateuch, as well as between pentateuchal literature and other compositions of the Hebrew Bible outside the Pentateuch. The notion of interaction, therefore, suggests external or indirect approaches to some of the major issues that preoccupy pentateuchal studies. In fact, a supposition of interaction between prophetic literature and pentateuchal materials sheds light on three major issues within pentateuchal studies: (a) the existence of, or acquaintance with, pentateuchal literature, to which different authors within biblical literature might have been exposed; (b) the compositional growth of pentateuchal literature; and (c) issues of canonization, or rather standardization, of pentateuchal materials. On each of these three topics, recent scholarship on the prophetic literature, and the papers in this collection in particular, may suggest contributions to pentateuchal studies. a. The existence of, or acquaintance with, pentateuchal literature within biblical literature. Literary interaction points to two basic modes of influence. The first is the recognized effect of one text upon another; the second is the possible mutual influence of the interacting compositions upon one another. The recognized literary effects could be of all kinds: approval of pentateuchal legal and literary materials by way of analogy; adaptation through various modes of transformation; or expansion and even reversal of pentateuchal traditions. The phenomenon of mutual exchange may be thought of as synchronic, going both ways simultaneously, or it may be approached diachronically, as a matter of allusion and exegesis. In reference to the latter, the directions of borrowing and influence are crucial and seem to lead to definitive results.11 Hence, interpretive relationships between prophetic passages and pentateuchal materials may illustrate an acquaintance of prophetic circles with specific legal and literary traditions at a variety of possible literary stages of the pentateuchal materialsʼ development, from oral traditions to texts. Conversely, proving literary dependence of pentateuchal texts on the prophets may likewise enrich our knowledge of the evolution and accumulation of the Pentateuch.

 The diachronic value of allusion and exegesis has been extensively investigated; see M. Fishbane, “Types of Biblical Intertextuality,” in Congress Volume: Oslo, 1998 (VTSup 80; Leiden: Brill, 2000), 39–44; B. D. S ommer , A Prophet Reads Scripture: Allusion in Isaiah 40–66 (Contraversions; Stanford: Stanford University Press, 1998), 6–31; and see an exemplary discussions in B. M. L evinson , Legal Revision and Religious Renewal in Ancient Israel (Cambridge: Cambridge University Press, 2008), 57–89, and M. A. Lyons, From Law to Prophecy: Ezekielʼs Use of the Holiness Code (LHBOTS 507; New York: T&T Clark, 2009), 47–75. 11

Introduction

837

b. The compositional growth of the pentateuchal literature. Bringing together the notions of interaction and formation of the Pentateuch highlights the need for additional paradigms to explain the compositional growth of pentateuchal literature. The presumed processes of gradual growth in the prophetic literature (as with historiographical compositions) suggest various paradigms for compositional growth that may apply to the pentateuchal literature as well. Looking to those other biblical compositions may suggest quite different processes than are suggested by the Documentary (or Neo-Documentary) Hypothesis. Such investigations may rather support the Fragmentary Hypothesis or the Supplementary Hypothesis. c. Issues of canonization, or rather standardization, of pentateuchal materials. Interaction between pentateuchal traditions and other biblical materials and compositions may serve as a key to reconstructing the long process of authorization (or protocanonization, to use Fishbaneʼs term) that pentateuchal traditions gradually underwent, even prior to their coalescing into a unified corpus.12 This long process might have begun much earlier than is usually thought; it might have developed slowly, even within preexilic Judah, and on a larger scale by the early postexilic era and after (that is, the Persian period). The fundamental characteristic of the basic paradigm of interaction is that it implies a contrasting model to paradigms of isolation, according to which pentateuchal sources are oftentimes said to have been composed and crystalized independently within isolated scribal schools.13 Instead, interaction signifies cultural and literary contact and, to a certain extent, negotiation between poets, historiographers, prophets, sages, priests, etc., including the keepers of pentateuchal traditions (oral or written; Jer 2:8). Interaction opens the horizon to the thought that these socioliterary circles were the intellectual cradles for ancient authors, who nurtured their own compositions in lively deliberation with other authors of different social and literary circles. Moreover, such contacts may at times be shown to have occurred prior to the final literary compilation of both the pentateuchal and the nonpentateuchal collections.14

 Fishbane, “Types of Biblical Intertextuality” (see n. 11), 39–44.  “Isolation” characterizes the Neo-Documentarian approach; see B. J. S chwartz , “Does Recent Scholarshipʼs Critique of the Documentary Hypothesis Constitute Grounds for Its Rejection?,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; Tübingen: Mohr Siebeck, 2011), 3–16; idem, “How the Compiler of the Pentateuch Worked: The Composition of Genesis 37,” in The Book of Genesis: Composition, Reception, and Interpretation (ed. C. A. Evans et al.; Leiden: Brill, 2012), 263–278; J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (New Haven, CT: Yale University Press, 2012), 13–33, 214–229, 246–249. 14  This distinction between “pentateuchal literature” and the Pentateuch was suggested by Y. K aufmann , The Religion of Israel (4 vols.; Jerusalem: Bialik, 1976 [1st ed., 1955]), 1:185–220, esp. 212 (Hebrew). I would suggest that most of the examples in the book of 12

13

838

Dalit Rom-Shiloni

Though many scholars would be willing to recognize interaction as a useful paradigm to explain innerbiblical relationships in general, these methodologies of interaction (whether intertextuality or allusion and exegesis) seem to have influenced the discussion about the actual formation of the Pentateuch only to a limited extent. To be more specific, research has developed (and continues to develop) to the point that we could ask what parts of the pentateuchal traditions were known to other biblical sources. Were they known orally or in writing, or should “writing-supported transmission of texts through memory” be the solution to the differences between the base texts and their invokers?15 Yet again, these extrapentateuchal references in prophetic literature do not (yet) play a major enough role in the study of the formation of the Pentateuch itself. I hope that this collection of essays will afford a better understanding of the opportunities that methodologies of interaction may bring to both small-scale and large-scale questions within the field of pentateuchal studies. It is important to realize that literary allusion and the exegesis of pentateuchal traditions or texts within prophetic literature are not simply a matter of aesthetics; that is, it is not simply the beauty of locating this or that pentateuchal phrase or concept within the prophets, while battling over the directions of influence, that matters. Rather, the search for literary references in prophecy may bring scholars of the Pentateuch to at least two main observations. First, by working carefully through literary relationships, literary allusion and exegesis in most cases show a clear picture of relative diachrony between the evoked text and the invoking one. Thus, outside, indirect work on the Prophets has quite a lot to say about the relative dating of traditions, and at times texts, in the Pentateuch and in prophecy. Second, and of more interest to me, is the methodological assumption that the prophets and their assumed audiences share the evoked literature as part of their common heritage. Tracing pentateuchal traditions in the prophets may thus shed light on the knowledge of pentateuchal traditions or even (and probably to a lesser extent) the breadth or depth of acquaintance with pentateuchal texts in the late preexilic and exilic periods. This search for literary references serves scholarly interest in revealing (or learning more about) the intellectual – cultural and theological – worlds and at times even the social arenas of these prophets (and likewise, their followers, tradents, and editors). Jeremiah belong to the stage of pentateuchal traditions/literature and thus to a prior stage than their current compilation in the Pentateuch. 15  Carr, Formation of the Hebrew Bible (see n. 2), 34, found these processes of memorization and performance to serve “long-duration texts/literature,” a term he defined in his earlier book (D. M. C arr , Writing on the Tablet of the Heart: Origins of Scripture and Literature [Oxford: Oxford University Press, 2005]) as literature recognized by a certain group to be their cultural/religious heritage and thus transmitted from one generation to another “to achieve cultural continuity” (10). This traditional-textual heritage is based on an authoritative corpus that may exist long before texts are considered canonized.

Introduction

839

Furthermore, scholars of the Pentateuch might gain from this outside or indirect information glimpses of the Sitz im Leben (setting) – the oral or written usage or performance of pentateuchal materials in the “real life” of Israel and Judah in the preexilic and exilic/postexilic eras. This might additionally challenge the common scholarly trend to portray the evolution of pentateuchal literature as an isolated literary discussion among circles of scribes in the late Persian period. To close, I hope that the twelve articles that follow are intriguing enough to encourage students and scholars of the Pentateuch to grapple with the question of whether and how the notion of interaction could be part of the conversation about the evolution of the pentateuchal corpus.

The Prophets after the Law or the Law after the Prophets? Terminological, Biblical, and Historical Perspectives Konrad Schmid This contribution deals with the formula lex post prophetas, i.e., “the law comes after the prophets,” which denotes the groundbreaking reordering of ancient Israelʼs religious history especially as it is linked with the name of Julius Wellhausen.1 My approach to this issue is threefold. First, I attempt to trace the origin of that often-cited formula: lex post prophetas. Who used it first? Then I will describe the basic biblical perspective on the problem, and a final section will bring up some historically informed examples.2

1 The Historical Origins of the Formula lex post prophetas If we investigate the historical origins of the formula lex post prophetas (“the law comes after the prophets”), we have to distinguish between the concept of dating “the law after the prophets” and the formula itself. The concept as such is usually associated with Julius Wellhausen, who opens his Prolegomena to the History of Israel (first published 1878 under the title Geschichte Israels, vol. I) by describing the problem he had understanding the prophets in light of the Pentateuch: Endlich faßte ich mir Mut und arbeitete mich hindurch durch Exodus Leviticus und Numeri und sogar durch Knobels Kommentar dazu. Aber vergebens wartete ich auf das Licht, welches von hieraus auf die geschichtlichen und prophetischen Bücher sich ergiessen sollte.3

 See R. S mend , Julius Wellhausen: Ein Bahnbrecher in drei Disziplinen (Munich: Carl Friedrich von Siemens Stiftung, 2004). 2  For the difference between “historical” and “biblical” Israel, see R. G. Kratz, Historisches und biblisches Israel: Drei Überblicke zum Alten Testament (Tübingen: Mohr Siebeck, 2013), 141–143. 3  J. Wellhausen, Prolegomena zur Geschichte Israels (3rd ed.; Berlin: Reimer, 1886), 3. 1

842

Konrad Schmid

At last, I took courage and made my way through Exodus, Leviticus, Numbers, and even through Knobelʼs Commentary to these books. But it was in vain that I looked for the light which was to be shed from this source on the historical and prophetical books.4

But then, in the wake of a personal encounter and communication, Wellhausen found a way out of his aporia: Da erfuhr ich gelegentlich im Sommer 1867, daß Karl Heinrich Graf dem Gesetze seine Stelle hinter dem Propheten anweise, und beinahe ohne noch die Begründung seiner Hypothese zu kennen, war ich für sie gewonnen: ich durfte mir gestehen, daß das hebräische Altertum ohne das Buch der Thora verstanden werden könne.5 At last, in the course of a casual visit in Göttingen, in the summer of 1867, I learned through Ritschl that Karl Heinrich Graf placed the Law later than the Prophets, and, almost without knowing his reasons for the hypothesis, I was prepared to accept it; I readily acknowledged to myself the possibility of understanding Hebrew antiquity without the Torah.6

The concept of the law coming after the prophets is attested here, but Wellhausen does not use the formula lex post prophetas here or elsewhere in his writings. As for the origin of the concept, Wellhausen traces the historical roots from Graf back to Reuss, George, and Vatke: Die Hypothese, die man nach Graf zu benennen pflegt, stammt nicht von ihm, sondern von seinem Lehrer Eduard Reuss. Am richtigsten wäre sie aber zu benennen nach Leopold George und Wilhelm Vatke; den sie haben dieselbe zuerst literarisch vertreten, unabhängig von Reuss und unabhängig voneinander.7 The hypothesis usually associated with Grafʼs name is really not his, but that of his teacher, Eduard Reuss. It would be still more correct to call it after Leopold George and Wilhelm Vatke, who, independent alike of Reuss and of each other, were the first to give it literary currency.8

So Reuss is the immediate predecessor of Graf, but Reuss himself was not the first to conceive of the idea; George and Vatke had written about it before him. But neither do any of these three authors use the formula lex post prophetas. The situation is especially difficult with Reuss. According to his own recounting, he first expounded on the idea in his 1834 lecture class on the introduction to the Old Testament. He published it only much later, however, in 1881. Reuss explains the delay of his publication as follows:9  J. W ellhausen , Prolegomena to the History of Israel (trans. J. Sutherland Black and A. Menzies; Cambridge: Cambridge University Press, 2013), 3. 5  Wellhausen, Prolegomena zur Geschichte Israels (see n. 3), 4. 6  Wellhausen, Prolegomena to the History of Israel (see n. 4), 4–5. 7  Wellhausen, Prolegomena zur Geschichte Israels (see n. 3), 4. 8  Wellhausen, Prolegomena zur Geschichte Israels (see n. 3), 5. 9  E. R euss , Die Geschichte der Heiligen Schriften Alten Testaments (Braunschweig: Schwetschke, 1881), VII (English translation mine). 4

The Prophets after the Law or the Law after the Prophets?

843

[D]er Entwurf ist zum ersten Mal im Sommersemester 1834 Gegenstand einer Vorlesung gewesen. . . . Wer die Literatur jener Zeit sich vergegenwärtigt, nicht die conservative blos, sondern namentlich auch die kritische, der wird es begreiflich finden dass ich mich scheute sofort der gelehrten Welt die Herausforderung hinzuwerfen, die Propheten für älter als das Gesetz anzuerkennen, und die Psalmen jünger als beide. [T]he draft was first the topic of a lecture course in the summer semester of 1834. [. . .] Whoever reflects on the literature of that time, not only the conservative, but especially also the critical one, will understand that I was immediately hesitant about challenging the academic world to consider the prophets older than the law, and the Psalms later than both.

The concept of lex post prophetas is clearly attested in that statement, but not the formula as such. Independently of one another, both Leopold George and Wilhelm Vatke argued in 1835 for the late date of the cultic laws from Exodus through Numbers.10 Interestingly enough, Vatke strongly criticized George and even rebuked him for being simplistic:11 Nach unserer Ansicht ist der Vf. in einem Irrthume befangen, wenn er überall voraus­ setzt, daß das leere Ceremonienwesen sich erst nach dem Exile gebildet habe. Stellen der älteren Propheten bezeugen hinlänglich, daß es in der Praxis lange vor dem Exile herrschte; später wurde es nur gesetzlich fixiert und kam als Moment zur älteren Gesetzgebung, welche einen wesentlich sittlichen Mittelpunkt hatte. In our view, the author [i.e., George] is mistaken if he presupposes generally, that the vapid ceremonial cult was only formed after the exile. Verses in the earlier prophets prove sufficiently that a praxis of ceremonies existed long before the exile. It was only later codified in legal terms and appeared as an [additional] element to the earlier legislation, which had a basically moral center.

Wellhausen, of course, thought more along the lines of Vatke than of George, although he is a bit ambiguous in his writings. In his Prolegomena, he builds up a camouflage for all biblical laws that predate P: Wenngleich das Deuteronomium und der Priesterkodex erst in sehr später Zeit auf­ge­ zeich­net worden sind, so bleibt doch noch die jehovistische Gesetzgebung (Exod. 20–23 Kap. 34), die als schriftlicher Ausgangspunkt der israelitischen Religions­geschichte betrachtet warden könnte.12

 L. George, Die älteren Jüdischen Feste mit einer Kritik der Gesetzgebung des Pentateuch (Berlin: Schroeder, 1835); W. Vatke , Die biblische Theologie wissenschaftlich dargestellt, Teil 1: Die Religion des Alten Testaments nach den kanonischen Büchern entwickelt (Berlin: Bethge, 1835). 11  W. Vatke, review of Die älteren Jüdischen Feste mit einer Kritik der Gesetzgebung des Pentateuch, by L. George, Jahrbücher für wissenschaftliche Kritik 1 (1836), 857–863, here 860. 12  Wellhausen, Prolegomena zur Geschichte Israels (see n. 3), 410. 10

844

Konrad Schmid

Even if it be the case that Deuteronomy and the Priestly Code were only reduced to writing at a late period, still there remains the Jehovistic legislation (Exodus xx.–xxiii. xxxiv.) which might be regarded as the document which formed the starting-point of the religious history of Israel.

But, as Wellhausen states in the Prolegomena, this is not in fact the case: Wol fehlte es auch im alten Israel nicht an gottgegebenen Grundlagen für die Ordnung des menschlichen Lebens, nur waren sie nicht schriftlich fixirt.13 Ancient Israel was certainly not without God-given bases for the ordering of human life; only they were not fixed in writing.

It is difficult to determine precisely what Wellhausen means here by “ancient Israel,” but one gets the impression that he is simplifying things in order to provide a clear reconstruction. His Composition is unambiguous about the existence of law traditions, the Decalogue in E, the Covenant Code in J, and the stand-alone piece in Exodus 34 in the monarchic period: Ausser den Berichten der beiden fortlaufenden Quellen E und J habe ich einen dritten völlig selbständigen Bericht in Kap. [sc. Exodus] 34 nachweisen zu können geglaubt, so dass also drei verschiedene Erzählungen des Vorgangs und drei verschiedene Auf­ zeich­nungen des Inhalts der Gesetzgebung vorlägen, der Dekalog in E, das Bundesbuch in J, das Goethesche Zweitafelgesetz in Exod. 34.14 Besides the accounts of the two continuous sources J and E, I think I have been able to reconstruct a third, completely stand-alone account in chapter 34. As a consequence, there are three different narratives of the event and three different records of the content of the legislation, the Decalogue in E, the Covenant Code in J, the two tablets as mentioned by Goethe in Exod. 34.

So only George formulates the notion of the law after the prophets as a clear-cut division: all laws are later than the prophets. Vatke and Wellhausen present a more differentiated view: the cultic laws usually attributed to P or, as Wellhausen names it, to Q are later than the prophets but not Exod 20–23 or Exod 34. So far, we have only discussed the concept of lex post prophetas. We have not found a single appearance of the formula itself in Vatke, George, Reuss, or even Wellhausen. The question of the formulaʼs origin is very sparsely discussed in the scholarly literature. Morgan and Bartonʼs Biblical Interpretation from 1988 devotes only half a page to inform us (correctly) that the “phrase lex post prophetas was more used about Wellhausen than by him.”15 They credit Ernst Wilhelm

 Wellhausen, Prolegomena zur Geschichte Israels (see n. 3), 411.  J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (3rd ed.; Berlin: Reimer, 1899), 95; translation mine. 15  R. Morgan and J. Barton, Biblical Interpretation (Oxford: Oxford University Press, 1988), 79. 13 14

The Prophets after the Law or the Law after the Prophets?

845

Hengstenberg with coining the phrase in his critique of Vatke.16 Unfortunately, they provide no clear reference. Hengstenberg writes: Wie der Christus des N.T. nach Strauß ein Produkt der christlichen Gemeinde ist, so ist der Moses des A.T. nach Vatke ein Produkt der Israelitischen, an dessen Erzeugung sie eine lange Reihe von Jahrhunderten gearbeitet hat. Er rühmt dieser Ansicht nach, daß die Propheten bedeutend dabei gewinnen. S. 481. Die bisherige Ansicht, wonach das Prophetentum aus dem Gesetze erwuchs, wirft er mit einem Schlag zu Boden, als der natürlichen Entwickelung widersprechend.17 Just as the Christ in the New Testament is a product of the Christian community according to Strauß, so Moses in the Old Testament according to Vatke is a product of the Israelite [community]. On the construction [of the Moses character], they worked for many centuries. He [i.e., Vatke] boasts that the prophets grow much more important thereby, p. 481. He knocks down the traditional view that prophecy grew out of the law with a single blow, as this would contradict the natural development.

Therefore, again, we have a clear attestation of the concept but not of the formula lex post prophetas. It is unclear who coined that term – maybe it was indeed Reuss, as Rudolf Smend (oral communication) suggests, but he never wrote it down.18

2 Biblical Perspectives Regarding the relationship between the law and the prophets in the purview of the Hebrew Bible itself, several levels of approach could be distinguished. First of all, according to the imagined scenery of the Hebrew Bible,19 of course, Moses predates all prophets – or at least the prophets that we now know from their books. (Abraham is termed a prophet in Gen 20:9, but does not have a book.) According to the biblical perspective, this is true not only in terms of the narrative timeline but also in terms of theological quality. Deuteronomy 34:10 states that, since Moses, “no prophet has arisen in Israel whom God knew face to face.” This statement is, of course, heavily discussed, most recently in a monograph by Jeffrey Stackert,20 and many divergent interpretations have been  Vatke, Die biblische Theologie (see n. 10).  E. W. Hengstenberg, Die Authentie des Pentateuches, 1. Band (Berlin: Ludwig Oehmigke, 1836), li. 18  A terminus ante quem for the term is supplied by M. K egel , Bruno Bauer und seine Theorien über die Entstehung des Christentums (Leipzig: Quelle & Meyer, 1908), 11, n. 2: “Von hier aus ergab sich für Bauer ein scharfer Gegensatz zu Vatke, der bekanntlich aus vielen Gründen die These ‘lex post prophetas’ vertrat.” 19  On the difference between “biblical” and “historical” Israel, see Kratz, Historisches und biblisches Israel (see n. 2). 20  J. Stackert, A Prophet Like Moses: Prophecy, Law, and Israelite Religion (New York: Oxford University Press, 2014), 117–122; C. Nihan, “ ‘Moses and the Prophets’: Deuteronomy 16

17

846

Konrad Schmid

offered. Nonetheless, it is obvious that it relates to Deut 18:15, where Israel receives the promise that a prophet like Moses will be raised up in order to provide guidance for the people. Whatever the diachronic relationship between Deut 34:10 and 18:15 might be, 34:10 draws a sharp line between the prophet Moses and subsequent prophets. They belong to the same group, but there is a difference in quality between Moses and all others. Deuteronomy 34:10 seems to want to distinguish the prophecy of Moses from all later prophecy. We have the fitting counterparts to that conception in Josh 1:7–8, 13 and Mal 3:22, which subordinate the whole complex of the Neviʾim to the point of view of Deut 34:10.21 Malachi 3:22

Joshua 1:7–8, 13 Only be strong and very courageous, being careful to act in accordance with all the law that my servant Moses commanded you; do not turn from it to the right hand or to the left, so that you may be successful wherever you go. This book of the law shall not depart out of your mouth; you shall meditate on it day and night, so that you may be careful to act in accordance with all that is written in it. For then you shall make your way prosperous, and then you shall be successful. [. . .]

Remember the teaching of my servant Moses, Remember the word that Moses the the statutes and ordinances that I commanded servant of YHWH commanded you, saying, “YHWH your God is providing you a place him at Horeb for all Israel. of rest, and will give you this land.”

According to this line of argumentation, Moses is the first and incomparable member of a sequence of prophets that follow him. First come the prophets of the nəbîʾîm rīʾšônîm, then the prophets of the nəbîʾîm ʾaḥărônîm.22

18 and the Emergence of the Pentateuch as Torah,” SEÅ 75 (2010), 21–55; M. Köckert, Leben in Gottes Gegenwart: Studien zum Verständnis des Gesetzes im Alten Testament (FAT 43; Tübingen: Mohr Siebeck, 2004), 195–215. 21  Cf. O. H. Steck, Der Abschluß der Prophetie: Ein Versuch zur Frage der Vorgeschichte des Kanons (BTSt 17; Neukirchen-Vluyn: Neukirchener Verlag, 1991), 134–136; see also L. M. M c D onald , The Biblical Canon: Its Origin, Transmission, and Authority (Peabody: Hendrickson, 2007), 78–80. 22  This distinction dates to the eighth century CE; see W. Dietrich et al., Die Entstehung des Alten Testaments (Theologische Wissenschaft 1/1; Stuttgart: Kohlhammer 2014), 20.

The Prophets after the Law or the Law after the Prophets?

847

3 Historical Perspectives At this point, a very basic distinction needs to be introduced that is often not sufficiently taken into account in biblical studies: the difference between the world of the narrative and the world of the narrator. For the Pentateuch, the world of the narrative is basically the second millennium BCE (if we bracket for a moment the primeval history), whereas the world of the narrator belongs basically to the first millennium BCE. Despite all divergences in pentateuchal scholarship, it is fair to say at least that much. Regarding the prophetic books, we have a similar constellation: according to the books themselves, the world of the narrative is the lifetime of the prophets. The world of the narrators may, but does not necessarily, coincide completely with the world of the narrative. It can possibly reach down to the closure of the Neviʾim and the end of literarily productive reinterpretation of the prophetic books. If we are a little bolder – and this is contested to a greater extent – we can say that the written texts of the Pentateuch may have originated between the ninth and the fourth centuries BCE.23 There may be some earlier24 and some later texts,25 and there are probably even older oral traditions reworked in the Pentateuch, but that is basically the time span we should reckon with.26

23  Cf., e.g., J. L. S ka , Introduction to Reading the Pentateuch (Winona Lake, IN: Eisenbrauns, 2006), 184–234; J. C. Gertz et al., T&T Clark Handbook of the Old Testament (New York: T&T Clark, 2012), 235–351. 24  Cf. V. Maag, “Zum Hieros Logos von Beth-El,” in Kultur, Kulturkontakt und Religion: Gesammelte Studien zur allgemeinen und alttestamentlichen Religionsgeschichte (ed. H. H. Schmid and O. H. Steck; Göttingen: Vandenhoeck & Ruprecht, 1980), 29–37; H. M. Wahl, Die Jakobserzählungen: Studien zu ihrer mündlichen Überlieferung, Verschriftung und Historizität (BZAW 258; Berlin: de Gruyter, 1997). 25  Cf., e.g., for parts of Num 22–24, H. R ouillard , La péricope de Balaam (Nombres 22–24) (EBib n.s. 4; Paris: Gabalda, 1985), 467; F. Crüsemann, Die Tora (Munich: Kaiser, 1992), 403; H.-C. S chmitt , “Der heidnische Mantiker als eschatologischer Jahweprophet: Zum Verständnis Bileams in der Endgestalt von Num 22–24,” in “Wer ist wie du, Herr, unter den Göttern?” Studien zur Theologie und Religionsgeschichte Israels; Festschrift für Otto Kaiser zum 70. Geburtstag (ed. I. Kottsieper et al.; Göttingen: Vandenhoeck & Ruprecht, 1994), 180–198, here 185; or for Gen 5, K. S chmid , Genesis and Exodus: Israelʼs Dual Origins in the Hebrew Bible (Siphrut 3; Winona Lake, IN: Eisenbrauns, 2010), 17–18; but see the important remarks of R. Hendel, “A Hasmonean Edition of MT Genesis?: The Implications of the Editions of the Chronology in Genesis 5,” HBAI 1 (2012), 448–464. 26  Cf., e.g., K. Schmid, Schriftgelehrte Traditionsliteratur: Fallstudien zur innerbiblischen Schriftauslegung im Alten Testament (FAT 77; Tübingen: Mohr Siebeck, 2011), 159–184; idem , “Der Pentateuch und seine Theologiegeschichte,” ZTK 111 (2014), 239–271; T. R ömer , “Der Pentateuch,” in Die Entstehung des Alten Testaments (Theologische Wissenschaft 1/1; Stuttgart: Kohlhammer 2014), 53–110.

848

Konrad Schmid

We get a similar, but not identical, picture if we look at the prophets. Amos, Hosea, and Isaiah as the earliest prophets belong to the eighth century, so the growth of the literary tradition may have started a little later than in the Pentateuch. Yet some of the prophetic books were redactionally expanded as late as the third century BCE, although this is contested.27 Taken together, there is a historical realm of possible mutual influence reaching from approximately the eighth to the fourth centuries. It is likely not only that the Pentateuch influenced the prophets28 but that influences ran in the other direction as well.29 A good example for the reception of prophecy in the Pentateuch is the quote of Amos 8:2 in Gen 6:13, already identified in the early 1980s by Rudolf Smend.30 Godʼs statement in Gen 6:13, “The end has come” – an undisputed P text – seems to allude to Amos 8:2. The topic is also present in Ezek 7:2–3. Why should P (if we allow for a post-Amos date of P) take up Amos 8:2? This should be explained within the overall political-theological message of P. P advocates  Cf., e.g., R. G. Kratz, Die Propheten Israels (Munich: Beck, 2003); K. Schmid, The Old Testament: A Literary History (Minneapolis: Fortress, 2012). 28  Cf., e.g., E. O tto , “Jeremia und die Tora: Ein nachexilischer Diskurs,” in Die Tora: Studien zum Pentateuch, Gesammelte Aufsätze (BZABR 9; Wiesbaden: Harrassowitz, 2009), 515–560 (see the bibliography in 517, n. 10); C. M aier , Jeremia als Lehrer der Tora: So­ zia­le Gebote des Deuteronomiums in Fortschreibungen des Jeremiabuches (FRLANT 196; Göttingen: Vandenhoeck & Ruprecht, 2002); D. R om -S hiloni , “Torah Interpretation in Jeremiah: Exegetical Techniques and Ideological Intentions,” Shnaton 17 (2007), 43–87 (Hebrew); idem , “Actualization of Pentateuchal Legal Traditions in Jeremiah: More on the Riddle of Authorship,” ZABR 15 (2009), 254–281; T. K rüger , “Das menschliche Herz und die Wei­sung Gottes: Elemente einer Diskussion über Möglichkeiten und Grenzen der Tora-Rezeption im Alten Testament,” in Das menschliche Herz und die Weisung Gottes: Stu­dien zur alttestamentlichen Anthropologie und Ethik (ATANT 96; Zürich: Theologischer Verlag, 2009), 107–136; see also T. Römer, “La rédaction des trois grands prophètes comme réaction à la crise de lʼexil baby­lonien,” Transeu 42 (2012), 69–80. Cf. further the broader reflections of H. Najman, Seconding Sinai: The Development of Mosaic Discourse in Second Temple Judaism (JSJSup 77; Leiden: Brill, 2003), 1–40. 29  Cf., e.g., C. L evin , “The ‘Word of Yahweh’: A Theological Concept in the Book of Jeremiah,” in Re-Reading the Scriptures: Essays on the Literary History of the Old Testament (FAT 87; Tübingen: Mohr Siebeck, 2013), 221–243, here 224, n. 7 (on the relationship between Jer 1:7, 9 and Deut 18:18). 30  R. S mend , “ ‘Das Ende ist gekommen’: Ein Amoswort in der Priesterschrift,” in Die Botschaft und die Boten: Festschrift für Hans Walter Wolff zum 70. Geburtstag (ed. J. Jeremias and L. Perlitt; Neukirchen-Vluyn: Neukirchener Verlag, 1981), 67–74, reprinted in idem , Die Mitte des Alten Testaments: Exegetische Aufsätze (Tübingen: Mohr Siebeck, 2002), 238–243; cf. also J. C. G ertz , “Noah und die Propheten: Rezeption und Reformulierung eines altorientalischen Mythos,” Deutsche Vierteljahrsschrift für Literaturwissenschaft und Geistesgeschichte 81 (2007), 503–522; and, foremost, T. P ola , “Back to the Future: The Twofold Priestly Concept of History,” in Torah and the Book of Numbers (ed. C. Frevel et al.; FAT 2/62; Tübingen: Mohr Siebeck, 2013), 39–65. 27

The Prophets after the Law or the Law after the Prophets?

849

a very peaceful world in which God envisions no violence or judgement toward his creation. Maybe the most blatant illustration of this is Godʼs bow in the clouds according to Gen 9:31 God puts his weapon away, and he will never again destroy the earth. But why then this interaction with Amosʼs prophecy of doom? P had to come to terms with the prophetic tradition that was known at its time. And Gen 9 demonstrates Pʼs solution to this issue: Yes, there was an end of the world decreed by God, but this is a crisis that has been resolved. It happened a very long time ago and has been settled by God once and for all. In order to interact in such a subversive way with the biblical prophecy of doom, P transformed Amos 8:2 from a divine statement about the present into a primeval action. The reworking of the Pentateuch in the prophets is a more familiar perspective in biblical scholarship on the direction of influence. Some examples are nearly undisputed, for example, the reception and reworking of Deut 24:1–4 in Jer 3:1–532 and of Deut 23:1–9 in Isa 56:1–7.33 I would like to add one more example here in order to show a less evident but nevertheless important example of an innerbiblical reinterpretation of the Torah in the prophets: Jer 30:18, promising the rebuilding of the town on its ruins, is very close to Deut 13:17,34 the law of an apostate town that shall be burnt and remain a ruin for ever. Apparently the authors of Jer 30:18, a promise directed toward Jerusalem, did not dare to develop a promise that directly contradicted the Torah. Therefore, they took up the relevant Torah text, Deut 13:17, quoted it, and thereby updated it with prophetic authority. Yes, an apostate town needs to be burned down, but, in the 31  U. Rüterswörden, “Der Bogen in Genesis 9: Militärhistorische und traditionsgeschichtliche Erwägungen zu einem biblischen Symbol,” UF 20 (1988), 247–263. 32  Cf., e.g., K. Schmid, Buchgestalten des Jeremiabuches: Untersuchungen zur Redaktionsund Rezeptionsgeschichte von Jer 30–33 im Kontext des Buches (WMANT 72; Neukir­ chen-Vluyn: Neukirchener Verlag, 1996), 277–294; G. F ischer , Jeremia 1–25 (HThKAT; Freiburg: Herder, 2005), 184–185; R om -S hiloni , “Torah Interpretation” (see n. 28); idem , “Actualization” (see n. 28). 33  H. Donner, “Jesaja LVI 1–7: Ein Abrogationsfall innerhalb des Kanons – Implikationen und Konsequenzen,” in Congress Volume, Salamanca (ed. J. A. Emerton; VTSup 36; Leiden: Brill, 1985), 81–95, reprinted in idem , Aufsätze zum Alten Testament aus vier Jahrzehnten (BZAW 224; Berlin: de Gruyter, 1994), 165–179; C. Maier, “Ist Versöhnung möglich? Jeremia 3,1–5 als Beispiel innerbiblischer Auslegung,” in “Gott bin ich, kein Mann”: Beiträge zur Hermeneutik der biblischen Gottesrede; Festschrift für Helen Schüngel-Strautmann zum 65. Geburtstag (ed. I. Riedel-Spangenberger and E. Zenger; Paderborn: Ferdinand Schöningh, 2006), 295–305; R om -S hiloni , “Torah Interpretation” (see n. 28), 43–87; M. I. G ruber , “Jeremiah 3:1–4:2 between Deuteronomy 24 and Matthew 5: Jeremiahʼs Exercise in Ethical Criticism,” in Birkat Shalom: Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Judaism Presented to Shalom M. Paul on the Occasion of his Seventieth Birthday (ed. C. Cohen et al.; 2 vols.; Winona Lake, IN: Eisenbrauns, 2008), 1:233–249. 34  Cf., e.g., Schmid, Buchgestalten des Jeremiabuches (see n. 32), 119–125; G. Fischer, Jeremia 26–45 (HThKAT; Freiburg: Herder, 135).

850

Konrad Schmid

case of Jerusalem, rebuilding is allowed, as God himself has promised through his prophet Jeremiah. In sum, the prophets come after the law, and the law comes after the prophets.35 In a diachronically differentiated approach, there are no easy solutions to that problem. This conclusion may be a burden to some of us but a relief to others.

 This is even true for the formation of Torah and Neviʼim, cf. S. B. Chapman, The Law and the Prophets: A Study in Old Testament Canon Formation (FAT 27; Tübingen: Mohr Siebeck, 2000). 35

Hoseaʼs Reading of Pentateuchal Narratives A Window for a Foundational E Stratum Marvin A. Sweeney I Hosea 12 includes references to three major narrative complexes in the Pentateuch, namely, the Jacob narratives in Gen 25–35, including references to the account of the birth of Jacob and Esau, Jacobʼs wrestling with an angel, his establishing territorial boundaries with Laban, and his journey to Aram to find a bride; the exodus narrative, including a reference to the prophet who brought Israel up from Egypt; and the wilderness narratives, including a reference to Israelʼs dwelling in tents following the exodus from the land of Egypt. Some interpreters maintain that these references must be later additions to the book of Hosea.1 Considerations supporting such a view would include difficulties in establishing the compositional history of the Pentateuch, coupled with the view that the final forms of both the Pentateuch and the Prophets are late redactional products of the Second Temple period. Some interpreters, citing the references to the Pentateuch among other arguments, now maintain that Hosea was composed entirely in the Persian period.2 But recent changes in scholarly understandings of both Hosea and the Pentateuch make it possible to realize that Hosea cited an early E or Ephraimite stratum of the pentateuchal narrative to make a point concerning Israelʼs relations with the Assyrian Empire. In the case of Hosea, reassessment of the Black Obelisk of Shalmaneser III and the vassal list of Adad-Nirari III indicate 1  See, e.g., the paper by Jakob Wöhrle in this volume. For the context of the study of Hosea in relation to the redactional formation of the Book of the Twelve, see J. Wöhrle, Die frühen Sammlungen des Zwölfprophetenbuche: Entstehung und Komposition (BZAW 360; Berlin: de Gruyter, 2006), 54–58, 229–240; idem, Der Abschluss des Zwölfprophetenbuches: Buchübergreifende Redaktionsprozesse in den späten Sammlungen (BZAW 361; Berlin: de Gruyter, 2006), 429–437; see also A. S chart , Die Entstehung des Zwölfprophetenbuchs (BZAW 260; Berlin: de Gruyter, 1998), esp. 101–155; J. D. Nogalski, Literary Precursors to the Book of the Twelve (BZAW 217; Berlin: de Gruyter, 1993), 58–73. 2  E.g., J. M. Bos, Reconsidering the Date and Provenance of the Book of Hosea: The Case for Persian-Period Yehud (LHBOTS 580; London: Bloomsbury, 2013).

852

Marvin A. Sweeney

that Israel was a vassal of Assyria during the reign of the Jehu dynasty.3 Hosea was opposed to that alliance and called for a realignment of Israelʼs foreign policy to forge a new alliance with Aram, from which Israelʼs ancestors had come.4 In the case of the Pentateuch, recent shifts in pentateuchal source theory indicate that the J or Judean stratum of the Pentateuch, formerly viewed as the earliest of the strata, actually dates to the late monarchic or even the exilic period.5 Such a redating of J leaves open the possibility that the E stratum of the Pentateuch, written during the time of the Northern Kingdom of Israel, was the earliest stage of pentateuchal composition and that the J stratum, written during the later period of the Judean monarchy and the aftermath of its destruction, was a redactional layer that reread and rewrote E after it was brought south to Judah following the collapse of the Northern Kingdom of Israel in the late eighth century BCE.6 This paper argues that Hosea cited an early Ephraimite (E) version of the Pentateuch to support his contention that Israel should abandon its alliance with Assyria in order to forge a new alliance with Aram, the homeland of Israelʼs ancestors. The paper proceeds in three major parts. First, it examines Hoseaʼs hostility to Assyria and Egypt in relation to the evidence from the Black Obelisk of Shalmaneser III and the vassal list of Adad-Nirari III to argue that the prophet calls for an end to Israelʼs alliance with Assyria. Second, it examines the pentateuchal citations in Hos 12 to establish their relationship to the current pentateuchal narrative in Genesis and their rhetorical functions in the book of Hosea. Third, it examines the pentateuchal narratives concerning Jacob, the exodus, and the wilderness journey to establish the likelihood that an E version of each narrative existed in the mid- to late eighth century when Hosea would have been active. The paper demonstrates that it is possible to posit an E or Ephraimite composition of the Pentateuch that would have included early 3  For the Black Obelisk of Shalmaneser III, see J. B. P ritchard , Ancient Near Eastern Texts Relating to the Old Testament (Princeton: Princeton University Press, 1969), 281; idem, The Ancient Near East in Pictures (Princeton: Princeton University Press, 1969), fig. 351. For the tributary list of Adad Nirari III, see S. Page, “A Stela of Adad Nirari III and Nergal-ereš from Tell al Rimlah,” Iraq 30 (1968), 139–153. 4  M. A. Sweeney, The Twelve Prophets (Berit Olam; 2 vols.; Collegeville, MN: Liturgical, 2000), 1:1–144, esp. 118–130. 5  E.g., J. Van Seters, Abraham in History and Tradition (New Haven, CT: Yale University Press, 1975); idem , Prologue to History: The Yahwist as Historian in Genesis (Louisville: Westminster John Knox, 1992); idem, The Life of Moses: The Yahwist as Historian in Exodus– Numbers (Louisville: Westminster John Knox, 1994); C. Levin, Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993); see also E. W. Nicholson, The Pentateuch in the Twentieth Century: The Legacy of Julius Wellhausen (Oxford: Oxford University Press, 1998), 132–160; T. B. Dozeman and K. Schmid (eds.), A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (SBL Symposium Series 34; Atlanta: Society of Biblical Literature, 2006). 6  E.g., T. Yoreh, The First Book of God (BZAW 402; Berlin: de Gruyter, 2010).

Hoseaʼs Reading of Pentateuchal Narratives

853

versions of the Jacob, exodus, and wilderness narratives. It further demonstrates that Hosea cited this narrative to support his argument that Israel should abandon its alliance with Assyria to reestablish its alliance with Aram.

II Interpreters generally recognize that Hosea is opposed to religious apostasy in northern Israel.7 The opening narrative in Hos 1:2–3:5 concerning Hoseaʼs marriage to Gomer and its implications for understanding YHWHʼs metaphorical marriage relationship with Israel make this concern quite clear. Just as the prophet charges his wife with adultery, claiming that she has pursued other lovers, and proposes to divorce her, so YHWH charges Israel with having pursued other gods and proposes to bring punishment upon the nation for having done so. Of course we will never know if either Gomer or Israel is actually guilty of the infidelity with which she is charged; neither Gomer nor Israel ever gets the chance to speak or to defend herself. Furthermore, we must recognize that the text is committed to defending divine integrity by charging Israel rather than YHWH with wrongdoing. This is particularly the case in view of Israelʼs destruction at the hands of the Assyrian Empire in 722–721 BCE despite YHWHʼs covenant with Israel that promised to make it a great nation with a population as numerous as the sands of the sea. Interpreters are less able to recognize the political dimensions of the prophetʼs message, namely, his opposition to Israelʼs vassal relationship with the Assyrian Empire from the time of King Jehu of Israel (842–815 BCE) through the last years of the nationʼs existence in 746–722 BCE.8 Part of the reason for their reluctance is the failure of Kings to mention Israelʼs alliance with Assyria during this period; another reason is the well-established view, correctly argued by both Cogan and McKay, that Assyria did not impose worship of its own gods on nations subjugated to its rule.9 But the Black Obelisk of Shalmaneser III portrays 7  For current major commentaries on Hosea, see F. I. A ndersen and D. N. F reedman , Hosea (AB 24; Garden City, NY: Doubleday, 1980); G. I. Davies, Hosea (NCB; Grand Rapids, MI: Eerdmans, 1992); J. Jeremias, Der Prophet Hosea (ATD 24/1; Göttingen: Vandenhoeck & Ruprecht, 1983); F. Landy, Hosea (Readings; Sheffield: Sheffield Academic Press, 1995); A. A. Macintosh, Hosea (ICC; Edinburgh: T&T Clark, 1997); H. W. Wolff, Hosea (Hermeneia; Philadelphia: Fortress, 1974); G. A. Y ee, “Hosea,” in The New Interpreters Bible (ed. L. E. Keck et al.; 12 vols.; Nashville: Abingdon, 1996), 7:197–297; Sweeney, Twelve Prophets (see n. 4), 1:1–144. 8  See especially my treatment of Hosea in Sweeney, Twelve Prophets (see n. 4), 1:1–144; idem , King Josiah of Judah: The Lost Messiah of Israel (Oxford: Oxford University Press, 2001), 256–272. 9  J. McKay, Religion in Judah under the Assyrians (SBT 2/26; Naperville, IL: Allenson, 1973); M. C ogan , Imperialism and Religion (SBLMS 19; Missoula, MT: Scholars Press,

854

Marvin A. Sweeney

King Jehu of Israel bowing at the feet of the Assyrian monarch in submission during the late ninth century BCE, and the vassal list of Adad-Nirari III includes Jehuʼs grandson, Joash ben Jehoahaz, among his tributaries. 10 Furthermore, Assyrian vassal treaties typically included lists of the gods of all the nations that are party to the treaty as witnesses as well as ritual ceremonies that committed the parties to the treaty before all of the gods named therein.11 The Assyrians may not have required the worship of their gods by nations bound to them by such treaties, but they certainly required recognition of their gods. Although unmentioned in the book of Kings, such an alliance is fully plausible for the late ninth through the mid-eighth centuries BCE, as it would have protected Israel from the advances of the Arameans. Aram had become vulnerable during the course of the late-ninth and eighth centuries BCE, and therefore sought to force their own alliance on Israel in order better to defend themselves against the Assyrians who began probing their borders during the reign of Shalmaneser III.12 Kings chose not to mention the Assyrian treaty, not because it did not exist but because Kings is designed to portray Assyria as the agent of YHWHʼs punishment against Israel. From the standpoint of Kingsʼ historiographical perspective, such a move would explain Israelʼs destruction in 722–721 BCE.13 With these considerations in mind, it is important to note that Hosea appears to have a special concern with the royal house of Jehu and the Assyrian Empire.14 The superscription for the book of Hosea in Hos 1:1 mentions, of the northern Israelite kings, only King Jeroboam ben Joash (786–746 BCE), the fourth monarch of the house of Jehu, and a sequence of Judean kings including Uzziah, Jotham, Ahaz, and Hezekiah (783–687/6 BCE). The account of Hoseaʼs failed marriage begins in Hos 1:2–5 with a divine command to marry a woman of harlotry so that she might bear children of harlotry, beginning with the first-born, Jezreel. This name is quite propitious because it speaks to the origins of the house of Jehu. Jehu overthrew the preceding house of Omri by assassinating King Jehoram ben Ahaz of Israel, King Ahaziah ben Jehoshaphat of Judah, and Jezebel bat Ittobaal, the widow of King Ahab ben Omri of Israel at Jezreel. Jezreel, of course, also signals Ahabʼs murder of Naboth, instigated by Jezebel, so that he might seize Nabothʼs property at Jezreel for himself. Fundamentally, Jezreel signals the 1974), 55–96. 10  See n. 3 above. 11  D. J. Wiseman, “The Vassal Treaties of Esarhaddon,” Iraq 20 (1958) 1–99; Pritchard, Ancient Near Eastern Texts (see n. 3), 534–541. 12  M. A. S weeney , 1 and 2 Kings: A Commentary (OTL; Louisville: Westminster John Knox, 2007), 330–340. 13  M. A. S weeney , “The Portrayal of Assyria in the Book of Kings,” in The Bible as a Human Witness to Divine Revelation: Hearing the Word of God through Historically Dissimilar Traditions (ed. R. Heskett and B. Irwin; LHBOTS 469; New York: T&T Clark, 2010), 274–284. 14  Sweeney, Twelve Prophets (see n. 4), 1:1–40.

Hoseaʼs Reading of Pentateuchal Narratives

855

origins of the house of Jehu, and the subsequent children born to Hosea and Gomer, the daughter, Lo-ruhamah (No Mercy), and the second son, Lo-ammi (Not My People), symbolize the dissolution of the marriage and the dissolution of YHWHʼs relationship with the Jehu-led Northern Kingdom of Israel. We may also observe the prophetʼs statements about the Assyrians.15 After accusing Israel of treachery in Hos 5:8–12, the prophet states in v. 13, “and Ephraim saw his sickness, and Judah his sore, and Ephraim went to Assyria and he sent to a contentious king. But he was not able to heal you, and he will not heal from you a sore.” Here we see a scenario in which the prophet portrays Ephraimʼs unsuccessful turn to Assyria for aid. The circumstances are not clear, but the turn to Assyria is unmistakable. In Hos 7:8–10, the prophet accuses Ephraim of seeking aid everywhere among the nations, but not from YHWH. In vv. 11–12, the prophet states, “And Ephraim has acted like a silly dove without sense; Egypt, they called; to Assyria they went. When they go, I will spread my net upon them; like a bird of the heavens, I will bring them down; I will chastise them when hearing of their treaty.” The term laʿădātām is problematic, but it is best rendered as a form of ʿēdût, “testimony, treaty.” Again, Israelʼs relationship with Assyria comes into question. The prophet accuses Israel of fruitlessly seeking recourse among the nations in Hos 8:7–14, “for they have gone up to Assyria like a wandering ass, Ephraim has hired lovers” (v. 9). In Hos 9:3, the prophet charges that “they shall not dwell in the land of YHWH, but Ephraim shall return to Egypt, and in Assyria, unclean food they will eat.” Hosea claims in Hos 10:5–8 that the idolatrous calf of Beth Aven/Bethel “shall be carried to Assyria” (v. 6). He refers in Hos 10:14 to defeat at Beth Arbel by Shalman (i.e., Shalmaneser III), the Assyrian monarch whom King Ahab fought and to whom Jehu submitted. In Hos 11:5, he claims, “No, he (Israel) returns to the land of Egypt, and Assyria is his king.” When YHWH rescues them in Hos 11:11, “They will flutter like birds from Egypt and like doves from the land of Assyria.” In Hos 12:2b, the prophet charges, “And a covenant with Assyria they make, and oil to Egypt they carry.” When the prophet calls for Israel to return to YHWH in Hos 14:2–9, he states in v. 4, “Assyria will not save us; upon horses we will not ride, and we will not say again, ‘our God,’ to the work of our hands.” Clearly, Hosea has a problem with Assyria – and Egypt. When one considers the alliance between the Jehu dynasty and Assyria, the cause of that problem becomes clear: Hosea objects to Israelʼs alliance with Assyria. And indeed, Egypt enters into the picture because scholars have long recognized that Assyriaʼs foreign policy in the late ninth through the eighth centuries BCE was to take control of the trade routes to Egypt and thereby dominate the economy of western Asia and the eastern Mediterranean.16 The question then becomes, “Why?” 15  For treatment of the following passages, see S weeney , Twelve Prophets (see n. 4), 1:1–40. 16  M. Elat, “The Economic Relations of the Neo-Assyrian Empire with Egypt,” JAOS 98 (1978), 20–34.

856

Marvin A. Sweeney

III The answer to our question lies in Hoseaʼs citations of the pentateuchal narratives, including the Jacob narratives and the exodus/wilderness narratives.17 Hosea 12 appears as a subunit within the larger structure of the book of Hosea as a whole. As my previous studies of Hosea demonstrate, the book is designed as a prophetic instruction that calls for Israel to return to YHWH. Following the superscription in Hos 1:1, the balance of the book in Hos 1:2–14:9 appears as Hoseaʼs parenetic appeal for Israelʼs return. The report of YHWHʼs speech to Hosea concerning marriage and the birth of children in Hos 1:2–2:3 sets the tone and parameters for the report of Hoseaʼs speech appealing for Israelʼs return in Hos 2:4–14:9. Hosea 12:1–15 appears as an element within this larger structure in which the prophet speaks but quotes YHWH throughout to make the appropriate points. Overall, the argument of Hos 12:1–15 is that YHWH has acted on Israelʼs behalf to bring them up from Egypt to a secure and well-defined land as indicated by Jacobʼs treaty with Laban/Aram as related in Gen 31; Israel must therefore abandon its relationship with Assyria and return to YHWH.18 Structure Diagram of Hosea Prophetic Instruction: Parenetic Appeal for Israelʼs Return to YHWH

1–14

  I. Superscription

1:1

 II. Parenetic Appeal for Israelʼs Return

1:2–14:9

  A. Report of YHWHʼs speaking to Hosea: marriage and birth of     children

1:2–2:3

  B. Report of Hoseaʼs speech concerning appeal for Israelʼs return

2:4–14:9

   1. report of Hoseaʼs appeal to children for motherʼs return:     restoration of united people under Davidic monarch

2:4–3:5

   2. report of YHWHʼs controversy against Israel

4:1–19

   3. detailed report of YHWHʼs call for Israelʼs return

5:1–14:9

    a. initial statement of issues: Israelʼs harlotry/alien children

5:1–7

    b. concerning Israelʼs alliance with Assyria/Egypt

5:8–7:16

    c. concerning Israelʼs kings and cultic apostasy

8:1–14

    d. concerning Israelʼs pagan cultic practice/rejection of YHWH

9:1–14:1

     (1) announcement of punishment

9:1–9

     (2) Israel as grapes found in wilderness/rebellion at Baal Peor

9:10–17

     (3) Israel as luxuriant vine that will be grown over

10:1–8

17 18

 For discussion, see Sweeney, Twelve Prophets (see n. 4), 1:116–130.  See Sweeney, King Josiah of Judah (see n. 8), 267.

857

Hoseaʼs Reading of Pentateuchal Narratives

     (4) Israel as trained heifer that must repent/return

10:9–11:11

     (5) Ephraimʼs need for exile and return like Jacob

12:1–15

     (6) Ephraimʼs exalted status turned to punishment and death

13:1–14:1

   4. Appeal for Israel to return to YHWH/abandon alliance with     Assyria

14:2–9

III. Postscript: Instruction for wise to understand (Hosea) and to act          properly

14:10

Hosea 12 begins in Hos 12:1–2 with the prophetʼs above-noted objections to Israelʼs covenant with Assyria, which causes them to present oil to Egypt. Hosea 12:3–7 then follows with the first set of references to the Jacob narratives of the Pentateuch.19 The pericope begins with the notice that “YHWH had a controversy with Judah” in v. 3a. Some interpreters view this notice with suspicion,20 but they overlook the role that Judah played in the tribal structure of the Northern Kingdom of Israel during the reigns of the Omride and Jehu dynasties. Judah was a vassal of northern Israel during this time. Although the Davidic monarch continued to sit on the throne in Jerusalem, Judah was obligated to give support to northern Israel in times of need. Examples would include Jehoshaphatʼs accompanying the king of Israel in battle against the Arameans at Ramot Gilead (1 Kgs 22) and against the Moabites (2 Kgs 3); the presence of the Judean prophet and sheep broker Amos of Tekoa at Bethel to present tribute offerings (Amos 7–9); the action of King Jehoash of Israel against King Amaziah of Judah when the latter attempted to revolt (2 Kgs 14:1–22); and the claims of King Jeroboam ben Joash/Jehoash of Israel to rule a kingdom that extended from Lebo-hamath in Aram to the sea of the Arabah in Judah (2 Kgs 14:23–29). Hosea 12:3b continues with a focus on YHWHʼs treatment of Jacob for his deeds. Verses 4–9 present a series of acts drawn from the pentateuchal narrative beginning with the references to Jacobʼs attempt to supplant his brother in the womb (Gen 25:19–26), continuing with his struggle with G-d (Gen 32:4–33), and culminating in the reference to YHWH finding Jacob at Bethel to speak with him (Gen 28:10–22). Each citation serves as a means to remind the reader of Jacobʼs experience with YHWH. In the case of his struggle with his twin brother Esau, Jacob overcame Esau early on by obtaining the rights of the firstborn (Gen 25:27–34) and the blessing of his father Isaac (Gen 27:1–45), but he was forced into exile to Aram in order to find a bride and to avoid Esauʼs wrath (Gen 27:46–28:9). Even when Jacob returned to the land of Israel with his wives and children, he still had to contend with an angry Esau bent on vengeance. These 19  For earlier studies, see H. U tzschneider , Hosea: Prophet vor dem Ende (Freiburg: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht, 1980), 186–230; H.-D. Neef, Die Heilstraditionen Israels in der Verkündigung des Propheten Hosea (BZAW 169; Berlin: de Gruyter, 1987), 15–49. 20  E.g., Wolff, Hosea (see n. 7), 206, 211.

858

Marvin A. Sweeney

examples point to YHWHʼs support of Jacob during his struggle with Esau to prepare him to serve as Israelʼs eponymous ancestor. In the case of Jacobʼs wrestling with G-d, the language of wrestling with G-d is certainly preserved in Gen 32:29 at the point when the “man” with whom Jacob wrestled declared that his name would henceforth be “Israel” to signify his status as one who struggled with men and with G-d and prevailed, as well as his status as the eponymous ancestor of Israel. Interpreters have been troubled by the reference to an “angel” in Hos 12:5, but such a statement need not reflect the text of an early version of the Genesis narrative as some contend. Rather, it is an attempt to qualify the statement that Jacob had wrestled with G-d and prevailed when, in fact, the text in Gen 32:25 states that his opponent was a “man.” By introducing the term angel, Hos 12:5 resolves the ambiguity inherent in the (Proto-)Genesis presentation as well as the theological problem posed by a statement that Jacob had overcome G-d/YHWH. As a result of the encounter, YHWH rewarded Jacob with his new status as eponymous ancestor of Israel and likewise explained the name of the city Penuel in Transjordan as well as the Israelite practice of devoting the thigh portion of a sacrifice to YHWH. Finally, Hos 12:5b refers to YHWHʼs appearance to Jacob at Bethel in Gen 28:10–22, where YHWH promised that Jacob would become a great nation in the land of Israel with descendants like the dust of the earth spreading throughout the land and blessing the families of the land. We may also note that Jacobʼs second encounter with YHWH at Bethel in Gen 35:1–15 again results in the promise that Jacob would become a great nation and the reiteration of his name as Israel. Hosea 12:6–7 follows up with a reminder that YHWH guided and blessed Jacob throughout all of these struggles and an appeal to Jacob, therefore, to return to YHWH. In the context of Hos 12:1–2, such an appeal to return to YHWH would be a reconsideration of the covenant with Assyria. A key consideration that ties these references together is not only YHWHʼs support of Jacob, but Jacobʼs growing from a contentious young man who rivals his older twin brother into a man who becomes the ancestor of his nation.21 Each reference – the conflict with Esau, who is the eponymous ancestor of Edom; the wrestling with the man/G-d/angel to found Penuel and to obtain the name Israel; and the vision of YHWH at Bethel – results in elements that define the national experience and character of the Northern Kingdom of Israel, namely, its conflicts with Edom during the eighth century BCE that result in Edomʼs breaking away from Israel and its vassal Judah (2 Kgs 8:20–22; 14:7; cf. 16:6) during the late ninth through the eighth centuries BCE. The references to YHWHʼs granting the name Israel to Jacob constitute etiologies for the name of the nation. The reference to Penuel likewise explains the origins of a Transjordanian city that once served as King Jeroboam ben Nebatʼs capital for ruling the nation (1 Kgs 21  For discussion of the Jacob narratives, see M. A. Sweeney, “Puns, Politics, and Perushim in the Jacob Cycle,” Shofar: An Interdisciplinary Journal of Jewish Studies 9 (1991), 103–118.

Hoseaʼs Reading of Pentateuchal Narratives

859

12:25). The reference to Jacobʼs wounded thigh explains the origins of one aspect of northern Israelʼs sacrificial practice, and the account of the vision at Bethel explains the origins of northern Israelʼs royal sanctuary, not in polemical terms as in the case of Jeroboam ben Nebat (1 Kgs 12:25–13:34) but as a site that celebrated Jacobʼs encounter with YHWH and YHWHʼs promise that Jacob and his descendants would become a great nation. These are all elements of northern Israelʼs national identity. They do not reflect the Judean polemic against northern Israel in 1 Kgs 12:25–13:34 and 2 Kgs 17; instead, they make better sense as foundational narratives concerning the origins of the Northern Kingdom of Israel. Such narratives would find their setting in the Northern Kingdom of Israel itself, at some point following the foundation of the kingdom under Jeroboam ben Nebat and prior to its demise under Hoshea ben Elah. The prophet returns to accusations against Ephraim/Israel in Hos 12:8–11, where he charges in vv. 8–9 that Ephraim should be likened to a dishonest merchant who uses false balances to enrich himself to gain power and thereby commits transgression and sin. The prophet cites YHWH in vv. 10–12 to remind the nation of what YHWH has done for them. YHWH begins in v. 10 with a brief reference to bringing the people up from Egypt and enabling them to dwell in tents once again during the period of wilderness wandering. Both of these references refer to foundational events in Israelʼs history and a self-understanding in which YHWH led the people to freedom from Egyptian bondage. Verse 11 continues with YHWHʼs statement of speaking to prophets, granting them visions, and speaking with them through parables. Such statements would have to refer to Moses and perhaps also to Miriam and Aaron, who would have spoken on behalf of YHWH as prophets during the exodus from Egypt and during at least the first part of the wilderness wanderings. After all, Aaron and the Levites were not appointed as priests until Num 17–18, and both Miriam and Aaron were dead by Num 20–21. Again, the pentateuchal tradition is cited to support the prophetʼs contention of YHWHʼs support of Israel and to convince them to return to YHWH by rejecting the alliance with Assyria. Interpreters have struggled with Hos 12:12 insofar as its third-person form suggests that YHWHʼs speech has ended and the prophet has resumed with a reference to Israelʼs apostasy. But v. 12 has an unspecified subject for its third-person verbs: “If Gilead is iniquity, indeed, they are worthless/nothing. In Gilgal, they sacrificed bulls also (on) their altars like heaps upon the furrows of the field.” The only antecedent for the two instances of “they” in this enigmatic verse is YHWHʼs reference to the prophets in v. 11: “if Gilead is iniquity, then the prophets are worthless.” What does such a statement mean? We must recognize that Gilead marks the eastern border between Israel and Aram. King Ahaz of Israel died defending Gilead from Aram in 1 Kings 22, and his son, Jehoram, was assassinated by Jehu during the course of the deteriorating military situation in Gilead (2 Kgs 9–10). But Gilead is also the site of Jacobʼs treaty with Laban over their respective boundaries in Gen 31:22–54. Following Jacobʼs departure

860

Marvin A. Sweeney

from Haran with his family, Laban, the eponymous ancestor of Aram in the Jacob narratives, pursued Jacob and overtook him in Gilead, claiming that Jacob had stolen Labanʼs household gods. When Laban was unable to prove the charge, the two men came to an agreement over their respective territories and marked them by erecting a pillar and a heap of stones as a “Heap of Witness” (Gen 31:47; Hebrew, galʿēd; Aramaic, yəgar śāhădûtāʾ), in other words, as a boundary marker. The citation suggests that YHWHʼs prophets would be worthless if they led Israel to a land that would otherwise have been lost to Laban or anyone else. The point is reinforced with the reference to Gilgal. Gilgal was the site where Joshua crossed the Jordan River into the land of Israel and celebrated the first Passover in the land with circumcision and the required sacrificial offerings of firstborn bulls from the herd (Josh 5). Gilgal was also known for the twelve pillars that represented the twelve tribes of Israel. Insofar as the tribes of Manasseh, Gad, and Reuben are assigned land in the Transjordanian Gilead region, it would appear that Hos 12:12 is concerned with the reliability of YHWHʼs promise of the land of Israel, including the territory of Transjordan; in other words, “if Gilead is iniquity, they/the prophets are worthless; in Gilgal they sacrificed bulls upon their altars as heaps/boundary markers on the furrows of the field.” Hoseaʼs point is that prophets brought the people of Israel/Jacob into the land of Israel that has secure, recognized borders in Gilead as well as elsewhere. Should that land be lost, YHWHʼs efforts would be for naught and YHWHʼs prophets would be worthless. The point is further reinforced by the prophetʼs remarks in Hos 12:12–13. First, Jacob fled to Aram to find a bride, and he served Laban by tending sheep to pay the bride price. Jacobʼs sojourn in Aram appears in Gen 29–31. Second, YHWH brought Israel up from Egyptian bondage by the agency of a prophet who watched over them. Third, Ephraim gave bitter offense to YHWH, and YHWH punished Ephraim for it. The Hebrew term tamrûrîm, “bitter offenses,” is based on the root mrr, “to murmur, rebel,” and recalls Israelʼs rebellion against Moses and YHWH in the wilderness, for which they were punished. These statements make several key points. First, Jacobʼs brides came from Aram, which means that Aram – and not Assyria – constitutes Israelʼs family base as well as a suitable partner for a treaty, as indicated by past experience. Second, YHWH brought Israel up from Egypt by the hand of a prophet to inhabit the land. YHWH employed a prophet to act on Israelʼs behalf by bringing them up from Egypt into the land of Israel as related in the exodus narratives. Third, the Israelites rebelled in the wilderness, were punished for it, and will suffer punishment again if they do not return to YHWH. The initial premise of this unit was that the alliance with Assyria takes Israel back to Egypt to trade oil. Such an act constitutes an offense against YHWH, who brought them out of Egypt. By referring back to the Jacob, exodus, and wilderness traditions, Hosea argues that the people should abandon their treaty with Assyria and return to a relationship with Aram like that of Jacob in order to return to YHWH.

Hoseaʼs Reading of Pentateuchal Narratives

861

Hosea 13:1–14:1 goes on to make the same point in religious terms, namely, that Israel is pursuing idols (in its relationship with Assyria) and needs to return to YHWH, who brought them up from Egypt and into the land of Israel. Hosea 14:2–9 appeals for Israelʼs return and emphasizes that “Assyria will not save us” (Hos 14:4). Hoseaʼs point is not based on the relative power of Assyria and Aram; clearly, Assyria was the more powerful. Instead, it is based in Hoseaʼs understanding of Israelite ethnic and religious identity.

IV When we turn to the question of dating the pentateuchal narratives that Hosea would have cited, we must begin with the recognition that Wellhausenʼs source-critical model of J, E, D, and P is severely flawed. Wellhausenʼs sources are not independent narrative strands that have been woven together by a redactor but instead constitute redactional strata that expanded and reworked earlier layers of the narrative. The Priestly stratum appears to have composed the final edition of the work as a whole in the early Persian period, even though some of its elements, such as the Holiness Code in Lev 17–26, appear to be older.22 D appears to derive from the late seventh-century reign of King Josiah rather than earlier northern Israelite reform circles,23 but it constitutes most of the book of the Deuteronomy and therefore does not affect significantly the character of the tetrateuchal narrative. Although these issues raise a host of questions that have yet to be settled, the primary issue for the present is the interrelationship between the J and E strata. Wellhausen considered J to be the earliest of the sources, dating to the early Davidic monarchy during the ninth century BCE, and von Rad moved J back even earlier to the origins of the house of David in the tenth century BCE.24 E was placed in the early to mid-eighth century BCE, at the height of northern Israelite power prior to its collapse due to Assyrian invasions in 734–732 and 724–721 BCE. But Wellhausen had a problem with J and E in that he so frequently was unable to distinguish them. Based upon his view that 22  Nicholson, Pentateuch in the Twentieth Century (see n. 5), 196–221; A. F. Campbell and M. A. OʼB rien , Sources of the Pentateuch (Minneapolis: Fortress, 1993), 1–20; R. G. Kratz, “The Pentateuch in Current Research: Consensus and Debate,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 31–61. 23  Sweeney, King Josiah of Judah (see n. 8), 137–169. 24  J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (Berlin: Georg Reimer, 1889); G. von Rad, “The Form-Critical Problem of the Hexateuch,” in The Problem of the Hexateuch and Other Essays (London: SCM, 1966), 1–78; cf. Nicholson, Pentateuch in the Twentieth Century (see n. 5), 3–92.

862

Marvin A. Sweeney

a primitive, anthropomorphic understanding of G-d must date to the earlier source and that the house of David must be a focal point for the earliest source due to the founding role of the Davidic dynasty, Wellhausen judged J to be earlier. Consequently, J must constitute a coherent narrative that E, as the later source, would have supplemented. Neither Wellhausen nor von Rad was able to identify such a complete and coherent narrative at the level of either J or E. Ironically, Wellhausen thereby laid the foundations for later redaction-critical understanding of the sources with this move. The inability to reconstruct a coherent and complete foundational narrative remained a major problem of pentateuchal scholarship for decades, but it was later historical research that identified Assyrian influence in the so-called J stratum that began to change the character of the discussion. Scholars such as Thomas Thompson and John Van Seters began to recognize that elements of J presupposed Assyrian or even Babylonian elements that could not have been known in the early Davidic period.25 The narrative concerning the Tower of Babel in Gen 11 could have been known beginning only with the late eighth-century reign of Hezekiah, who first allied with Babylon in an effort to revolt against Assyria. The reference to Abramʼs march through the main sites of the land of Canaan in Gen 12 presupposes the Assyrian practice of Palu campaigns, when the Assyrian king would traverse his empire together with his army in an effort to ensure his rule over potentially wayward vassal nations.26 The patterns of Israelite law, including both the casuistic forms known from Hammurabiʼs law code and the apodictic forms known from seventh-century Assyrian treaty texts point to Assyrian influence in the composition of the so-called Covenant Code in Exod 20–24 and other Israelite legal texts.27 All of these features and more point to Assyrian and possibly even Babylonian influence in the composition of the J stratum of the Pentateuch, which pushes the composition of the J stratum into the late eighth, seventh, and possibly even the early sixth centuries BCE.28 But such a scenario raises questions concerning the E stratum. If J is later, E cannot constitute a supplementary stratum for J. Given Wellhausenʼs difficul25  T. L. T hompson , The Historicity of the Patriarchal Narratives: The Quest for the Historical Abraham (BZAW 133; Berlin: de Gruyter, 1974); Van Seters, Abraham (see n. 5). 26  H. Tadmor, “The Campaigns of Sargon II of Asshur: A Chronological Historical Study,” JCS 12 (1958), 22–40, 77–100. 27  For casuistic or case law, see A. Alt, “The Origins of Israelite Law,” in Essays on Old Testament History and Religion (Garden City, NY: Doubleday, 1967), 171; for apodictic law, see the stipulations in the Assyrian Vassal Treaties discussed in Wiseman, “Vassal Treaties of Esarhaddon” (see n. 11). 28  Cf. D. J. M c C arthy , Treaty and Covenant: A Study in Form in the Ancient Oriental Documents and in the Old Testament (AnBib 21A; Rome: Biblical Institute Press, 1978), which argues for an early date for such parallels based on his early dating of J; see also idem, Old Testament Covenant: A Survey of Current Opinions (Atlanta: John Knox, 1972), which surveys the state of scholarship at the time.

Hoseaʼs Reading of Pentateuchal Narratives

863

ties in distinguishing J and E and the redactional character of the pentateuchal strata, we must consider that E might be the foundational text and that J might be a redactional stratum.29 Interpreters have already demonstrated the largely Judean characters of the Primeval History of Gen 1–11 and the Abraham-Sarah cycle of Gen 11–25: the Primeval History presupposes characteristics of the Jerusalem temple, and the Abraham-Sarah narratives presuppose a Davidic-like dynastic concern in their focus on the birth of a son to Abraham and Sarah.30 But can the same be said of the Jacob, exodus, and wilderness narratives? Indeed, closer reading of each of these narratives indicates their northern or Ephraimite character which presupposes a setting in the late ninth or eighth century BCE. The Jacob narratives in Gen 25–35 have long been recognized as an account of the origins of Israel as recounted through the life of Jacob, Israelʼs eponymous ancestor.31 Throughout the narrative, Jacob is in conflict with his fraternal twin brother, Esau, the eponymous ancestor of Edom, and his uncle and father-in-law, Laban, the eponymous ancestor of Aram. Running parallel to the accounts of Jacobʼs conflicts is the sibling rivalry between his wives, Rachel and Leah, and their respective handmaidens, Bilhah and Zilpah, who became the mothers of the tribes of Israel. Interpreters have long understood this account to address the origins of Israel as a whole, long before the division of the nation into the Northern Kingdom of Israel and the Southern Kingdom of Judah. But there are problems with this view. One would expect Judah to play a more important role among the sons of Jacob insofar as Judah was the tribe of David and Solomon, who united the tribes under a single monarchy that disintegrated only after the death of Solomon. Instead of appearing in a prominent position among the sons of Jacob, Judah appears only as the fourth-born son of Leah, and indeed, Judah is grouped together with Leahʼs other sons, who represent geographically peripheral tribes in Israel: Reuben from northern Moab, Shimon from the Negev, landless Levi, Judah from the southern hill country, Issachar from Galilee, and Zebulun also from Galilee. The sons of the handmaidens are also geographically peripheral. The sons of Bilhah, Rachelʼs handmaid, are Dan and Naphtali from the region north of Galilee, and the sons of Zilpah, Leahʼs handmaid, are Gad and Asher from Transjordan and the region by Phoenicia, re Cf. Yoreh, First Book of God (see n. 6).  J. D. Levenson, “The Temple and the World,” JR 64 (1984), 275–298; J. Rosenberg, King and Kin: Political Allegory in the Hebrew Bible (Bloomington: Indiana University Press, 1986); R. E. Clements, Abraham and David: Genesis XV and its Meaning for Israelite Tradition (SBT 2/5; London: SCM, 1967). 31  E.g., C. Westermann, Genesis 12–36 (Continental Commentary; Minneapolis: Augsburg, 1985), 518. See M. A. Sweeney, “Puns, Politics, and Perushim” (see n. 21); N. Sarna, Genesis (JPS Torah Commentary; Philadelphia: Jewish Publication Society, 1989), 177–178; J. G ammie , “Theological Interpretation by Way of Literary and Tradition Analysis,” in Encounter with the Text: Form and History in the Hebrew Bible (ed. M. Buss; SBL Semeia Supplements; Philadelphia: Fortress, 1979), 117–134. 29

30

864

Marvin A. Sweeney

spectively. The sons of Jacobʼs beloved Rachel – Joseph, the father of Manasseh and Ephraim, and Benjamin – are located in the central and northern Israelite country as well as the strategically important Gilead region of Transjordan. Indeed, Ephraim and Manasseh constitute the core of the Northern Kingdom of Israel, and Benjamin was the first royal tribe of Israel under the house of Saul. The emphasis on the central roles of the tribes descended from the sons of Rachel and their identification as the key tribes of northern Israel becomes even more important when considering the northern features of the Jacob narrative. Jacob founds Bethel, later the royal sanctuary of the Northern Kingdom of Israel under Jeroboam ben Nebat, in Gen 28. He receives YHWHʼs covenant at Bethel in Gen 28 and in Gen 35 confirms his name, Israel, and receives the divine promise once again. Following his marriages to Rachel and Leah and the birth of his sons in Haran, Jacob returns to the land of Israel to found key locations for northern Israel in Transjordan: Penuel, where he wrestles with the “man” and receives the name Israel; Sukkot, where he camps on his journey home; and Mahanaim, where he confronts Esau and ultimately separates from him. It is noteworthy that these cities would have been lost to Israel in the aftermath of Jehuʼs coup overthrowing the house of Omri ca. 842 BCE. In the aftermath of the coup, 2 Kgs 10:32–33 relates that Hazael of Aram was able to strike Transjordan, including the tribes of Manasseh, Gad, and Reuben. The Moabite Stone, the Deir Alla Inscription, and the Tel Dan Inscription all confirm that Transjordan was lost to Israel at this time.32 Furthermore, 2 Kgs 8:16–24 notes that Edom broke away from King Jehoram of Judah, who ruled during the early reign of King Jehoram of Israel. It is also noteworthy that Judah was a vassal of Israel throughout the reigns of the house of Omri and the house of Jehu in the ninth and eighth centuries BCE.33 Indeed, King Jehoshaphat accompanied King Ahab of Israel to battle against Aram in 1 Kgs 22, and he again accompanied King Jehoram of Israel in battle against Moab in 2 Kgs 3. Jeshoshaphatʼs grandson, Ahaziah, later joined Joram of Israel in battle against the Arameans at Ramoth-gilead as his grandfather had done. Ahaziah was assassinated together with Jehoram of Israel by Jehu in 2 Kgs 8 when Ahaziah came to visit his wounded counterpart at Jezreel. Indeed, the house of Omri was closely tied to the house of David by the marriage of Athaliah, described in Kings both as the daughter of Omri (2 Kgs 8:26) and 32  The Moabite Stone, which relates the ninth-century defeat of Israel by King Mesha of Moab, appears in P ritchard , Ancient Near Eastern Texts (see n. 3), 320–321. The Deir Alla Inscription, which presents an eighth-century inscription concerning Balaam and the celebration of Aramean victory over Israel, appears in M. Dijkstra, “Is Balaam Also among the Prophets?,” JBL 114 (1995), 43–64. The Tel Dan Inscription, which celebrates an Aramean kingʼs defeat of Israel in the ninth century, appears in A. Biran and J. Naveh, “An Aramaic Stele Fragment from Tel Dan,” IEJ 43 (1993), 81–98; idem, “The Tel Dan Inscription: A New Fragment,” IEJ 45 (1995), 1–18. 33  For treatment of the Kings passages, see Sweeney, 1 and 2 Kings (see n. 12), 330–340.

Hoseaʼs Reading of Pentateuchal Narratives

865

the daughter of Ahab (2 Kgs 8:18), to King Jehoram of Judah, making her the mother of Ahaziah and the rest of the Davidic kings. The house of David was closely tied by alliance to the house of Omri; given Judahʼs smaller size, it was a suzerain-vassal alliance in which Israel was the suzerain and Judah was the vassal.34 Judahʼs vassalage to Israel would explain Judahʼs place as the fourth son born to Leah. Like his brothers, Judah would have represented a peripheral region in the Northern Kingdom of Israel.35 It would appear that the Jacob narratives of Gen 25–35 were written to portray the experience of the Northern Kingdom of Israel during the reigns of the house of Omri and the house of Jehu during the ninth and eighth centuries BCE.36 The Jacob narratives presuppose the central role of the Bethel sanctuary, the loss of Transjordan during the wars with Aram, the loss of Edom, the role of Judah as a vassal to both dynasties, and ultimately the restoration of Israelʼs borders with Aram during the reigns of the later Jehu monarchs.37 All of these events are played out in the interpersonal relations of the major characters of the narrative: Jacob, Esau, Laban, Rachel, Leah, and the sons of Jacob/Israel. The narrative appears to constitute a form of reflection on this period in which the eponymous ancestors of Israel, Edom, and Aram – together with the mothers of the tribes of Israel – play the major roles. These observations support the contention that the Jacob narratives in Gen 25–35 constitute an E (or Ephraimite) narrative concerning northern Israel that would have been written toward the end of the reign of the house of Jehu in the eighth century BCE to give expression to Israelʼs history and experience. The reign of Jeroboam ben Joash, 786–742 BCE, would provide a suitable setting, insofar as Israel was secure once again during his reign. Northern Israelite authors would then take the opportunity to write the Jacob narratives as a means to reflect upon Israelʼs history over the course of the last century. 34  Even later during the reign of the house of Jehu, King Amaziah of Judah attempted to revolt against the Jehu monarch, Jehoash of Israel. Amaziah was defeated by Jehoash, Jerusalem was breached, and Amaziah was ultimately assassinated for his miscalculation (2 Kgs 14:1–22; Sweeney, 1 and 2 Kings [see n. 12], 361–367). 35  It was only during the reigns of the Jehu monarchs, Jehoash ben Jehoahaz and Jeroboam ben Jehoash, that the Arameans were finally defeated and Israelʼs borders with Aram were secured once again (2 Kgs 14:22–25, 23–29). 36  See also Sweeney, “Puns, Politics, and Perushim” (see n. 21). 37  Avigdor Hurowitz argues that the account of Jacobʼs vision at Bethel in Gen 28 must date to the Babylonian period due to the correspondence of many of its details with the Babylonian ziggurat Entemenanki and other Mesopotamian parallels, but by the end of his study he acknowledges that the Entemenanki was widely imitated, even when it was under Assyrian control. He consequently states that the account of Jacobʼs dream could have been written as early as the reign of Jeroboam ben Nebat of Israel who reigned in the late tenth century BCE; see V. A. H urowitz , “Babylon in Bethel – New Light on Jacobʼs Dream,” in Orientalism, Assyriology, and the Bible (ed. S. W. Holloway; Sheffield: Sheffield Phoenix Press, 2007), 436–448.

866

Marvin A. Sweeney

But we must also note that the Jacob narrative has been edited to fit a larger J (or Judean) framework. Genesis 26, the account of the endangered matriarch, Rebekah, the wife of Isaac, takes place in Gerar along the borders of Judah and Philistia. Although Gen 27–28 does not mention Gerar, the placement of Gen 26 in its present position means that Jacob leaves from the vicinity of Judah and passes through Israel on his way to Aram.38 Indeed, Jacob returns through Israelite Transjordan to Bethel on his way home. He returns to Hebron in Judah only in Gen 35:27–29 to bury his father, Isaac, which would then serve together with Gen 26 as a redactional framework to locate Jacob in Judah rather than in Israel. It would appear that the E (or Ephraimite) Jacob narrative has been placed into a J (or Judean) narrative framework. We might add the observation that the narrative in Gen 34 concerning the rape of Dinah portrays Shechem, Israelʼs early gathering point on the borders of Ephraim and Manasseh, in a very bad light. Insofar as Gen 34 constitutes another example of an endangered-matriarch narrative, it appears to be the product of J redaction.39 The exodus and wilderness narratives are generally recognized as a combination of J, E, and P materials.40 They present an account of Israelʼs emergence as a nation as a result of its experience in the exodus from Egypt and the journey through the wilderness to the land of Canaan. Many interpreters recognize that these stories function as a type of creation narrative, insofar as natural features 38  See M. A. S weeney , “Form Criticism: The Question of the Endangered Matriarchs in Genesis,” in Method Matters: Essays on the Interpretation of the Hebrew Bible in Honor of David L. Petersen (ed. J. M. LeMon and K. H. Richard; RBS 56; Atlanta: Society of Biblical Literature, 2009), 17–38, which points to the potential redactional role of the endangered-matriarch narratives in Genesis. 39  The Joseph narrative in Gen 37–50 also appears to be a northern or E narrative insofar as it focuses on the emergence of Joseph as the father of northern Israelʼs leading tribes, Ephraim and Manasseh. It also appears to be subject to J redaction at two key points, Gen 37:18–36 (see esp. vv. 26–27) and Gen 42:35–38 // Gen 43:8–14, when Judah emerges as the figure who supplants Reuben, first when he protects Joseph from death when he is thrown into the pit by his brothers, and later when he protects Benjamin by demanding that the brothers bring Benjamin to appear before him. We may also note the appearance of another endangered-matriarch narrative in Gen 38, when Tamar ensures that Judahʼs descendants are not born to a Canaanite mother, and the placement of the blessing of Jacob over his sons in Gen 49, which ensures that Judah is recognized as the royal tribe. All together, these features suggest J editing of an earlier E narrative concerning Joseph. 40  Cf. T. B. D ozeman , “Hosea and the Wilderness Wandering Tradition,” in Rethinking the Foundations: Historiography in the Ancient World and in the Bible; Essays in Honor of John Van Seters (BZAW 294; ed. S. L. McKenzie and T. Römer; Berlin: de Gruyter, 2000), 55–70, who argues that the wilderness allusions in Hosea cannot be recognized as an early form of the pentateuchal narrative because they lack the theme of divine guidance. But the Hosea references do recognize the prophet (Moses) who served as the divine agent in these narratives. Even if they were presented orally at the Bethel sanctuary as Dozeman suggests, such oral forms would still have formed the basis for an early proto-pentateuchal narrative in the Northern Kingdom of Israel.

Hoseaʼs Reading of Pentateuchal Narratives

867

of the lands of Egypt, Sinai, and Israel play significant roles: the burning bush, the first nine plagues against Egypt, the parting of the Red Sea, the appearance of quails and manna in the wilderness, and the water from the rock.41 But the exodus and wilderness narratives also point to the origins of important features of Israelite national and religious life, such as the celebration of Passover as the first of the three major holidays, the establishment of courts of justice, the institution of the Levitical priesthood, and of course the revelation of Torah at Sinai, which serves as a model for the role played by Israelite temples in serving as the loci for the revelation of YHWH and divine Torah. There is little overt evidence that would identify Israel as presented in the exodus and wilderness narratives as either the Northern Kingdom of Israel or southern Judah. Even when the Song of the Sea indicates that YHWH will lead the people to a sanctuary, there is no hint that the sanctuary in any way represents Jerusalem or any of the northern sanctuaries, such as Gilgal, Shiloh, Bethel, or Dan. Rather, the narrative appears to presuppose all Israel. Nevertheless, one feature of the narrative is quite distinctive: the tenth plague, which calls for the slaying of the firstborn of Egypt and the redemption of the firstborn of Israel. Unlike the other nine plagues, the focus on the firstborn does not appear to explain the origins of the natural features of creation. Instead, it appears to explain the origins of an early form of Israelite priesthood, especially as practiced in northern Israel.42 The tenth plague appears in Exod 11, and it precedes the instructions concerning the celebration of Passover in Exod 12. Most interpreters view it as the culmination of the plagues that result in Israelʼs exodus, but they do not note the distinctive nature of YHWHʼs instruction to Moses in Exod 13:2 concerning the significance of the firstborn: “Sanctify to me every first born issue of every womb from the people of Israel, including humans and animals, it is mine.” This statement then serves as the basis for the instructions in vv. 1–16 concerning the observance of Passover and the recognition of YHWHʼs mighty act in delivering Israel from Egyptian bondage. This statement also serves as the basis for the law concerning the firstborn in Exod 34:19–20: “Every first issue of the womb is mine, from all of your cattle that are born male, the first issue of cattle or sheep. And the first issue of an ass you shall redeem with a sheep, and if you do not redeem it, you shall break its neck. All the firstborn of your sons you shall redeem, and you shall not appear 41  E.g., N. M. S arna , Exploring Exodus: The Heritage of Biblical Israel (New York: Schocken, 1986), 77; U. Cassuto, A Commentary on the Book of Exodus (Jerusalem: Magnes, 1967), 177–182; cf. W. H. C. Propp, Exodus 1–18 (AB 2; Garden City, NY: Doubleday, 1999), 554–561. 42  M. A. Sweeney, “Israelite and Judean Religions,” in The Cambridge History of Religions in the Ancient World, Volume 1: From the Bronze Age to the Hellenistic Age (ed. M. Salzman and M. A. Sweeney; Cambridge: Cambridge University Press, 2013), 151–173.

868

Marvin A. Sweeney

before me empty-handed.” Although the instructions concerning Passover in Exod 12 make it clear that the cattle and sheep are to serve as firstborn offerings to YHWH at Passover, the purpose of the redemption of the firstborn sons is not explained, either in Exod 34 or in Exod 12–13. Nevertheless, the narratives concerning Israelʼs departure from Sinai in Num 1–10 explain the function of the firstborn sons of Israel. A major issue in the Numbers account of the wilderness journey is the selection of Aaron and the Levites to serve as priests in Israel (see esp. Num 17–18). Such a selection was necessary to provide leadership for Israel and to put an end to the constant challenges of Moses and YHWH that appear throughout the wilderness narratives. At the outset of the narrative, it is noteworthy that YHWH states to Moses three times that the Levites will replace the first-born of Israel: “And YHWH spoke to Moses, saying, ‘And I, behold, I, have taken the Levites from the midst of the sons of Israel, and they shall be to Levites to me, for all the first-born were mine on the day that I smote all the first-born of the land of Egypt. I sanctified for myself all the firstborn in Israel. From humans to animals, they are mine. I am YHWH’ ” (Num 3:11–13; cf. Num 3:40–51; 8:13–19). The narratives in Num 1–10 make it clear that the Levites are to serve as the priests of YHWH for the people of Israel in place of the firstborn who formerly served in this role. The Levites are “to perform the service of the sons of Israel in the tent of meeting and to atone for the sons of Israel so that there shall not be among the sons of Israel a plague when the sons of Israel approach the sanctuary” (Num 8:19). Although most interpreters assume that the Levites are the legitimate priests of Israel, the priest and prophet Samuel serves as an example of a firstborn son who functions as a priest at Shiloh in 1 Sam 1–7.43 When his barren mother, Hannah, finally bears Samuel as her firstborn son, she takes him to the sanctuary at Shiloh where he is raised as a priest by Eli, the high priest of the sanctuary (1 Sam 1). He has a visionary experience of YHWH while sleeping by the ark of the covenant in 1 Sam 3, and he functions as the chief priest and leader of Israel following the death of Eli and his sons during the wars with the Philistines. It is noteworthy that Samuelʼs father, Elkanah, is identified as an Ephraimite in 1 Sam 1:1, although 1 Chr 6:1–15 identifies both Elkanah and Samuel as Levites in keeping with its general tendency to identify those from Samuel or Kings who take on holy functions as Levites. The phenomenon is all the more interesting when we recognize that King Jeroboam ben Nebat, the founding king of the Northern Kingdom of Israel, is criticized for a number of cultic sins, such as installing the golden calves for worship in the sanctuaries of Bethel and Dan, establishing a festival (Sukkot) on the fifteenth day of the eighth month, and appointing priests from among  See M. A. Sweeney, “Samuelʼs Institutional Identity in the Deuteronomistic History,” in Constructs of Prophecy in the Former and Latter Prophets and Other Texts (ed. L. L. Grabbe and M. Nissinen; ANEM 4; Atlanta: Society of Biblical Literature, 2011), 165–174. 43

Hoseaʼs Reading of Pentateuchal Narratives

869

the people who were not Levites. As I have pointed out elsewhere, Jeroboamʼs practices were not deliberate sins; rather, they represented an understanding of the religious institutions and practices of ancient Israelite religion different from that of Judah.44 In northern Israel, firstborn sons could serve as priests; in southern Judah, the Levitical priesthood was the norm. We do not know if all northern priests were firstborn sons. Eli and his sons appear to represent a Levitical dynastic priesthood, but the example of Samuel shows that the service of firstborn sons as priests was accepted in northern Israel. And later, when Solomon came to power, Abiathar (the surviving priest of the house of Eli) was expelled to Anathoth, whereas Zadok (a descendant of Aaron) was retained as the high priest in Jerusalem. It would seem then that the house of Eli did accept firstborn sons within its ranks, whereas the Zadokite line of Jerusalem did not. It is perhaps no accident then that the golden calf narratives of Exod 32–34, in which the Levites demonstrate their zeal for YHWH, portray Israelʼs sins of worshipping the golden calf as an analogy for the sins of Jeroboam.45 Jeroboamʼs golden calves likely served as symbolic mounts for an invisible YHWH at Bethel and Dan, much as the ark of the covenant served a similar function for YHWH in Jerusalem. But in Exod 32–34 and 1 Kgs 12:25–13:34, the golden calves are portrayed as idolatrous gods, and Jeroboamʼs decision to select priests who were not Levites is portrayed as a sin. But the introduction of the golden-calf narrative in Exod 32–34, which provides the justification for the later decision to appoint the Levites as priests in Num 1–10 and 17–18, appears to be a redactional move that shifted the focus of the exodus and wilderness narratives. The exodus narratives were originally composed to explain the origins of Israelʼs observance of the Passover festival and the role of the firstborn as priests. But the present form of the golden-calf narratives in Exod 32–34 and those concerned with the selection of the Levites as priests in Num 1–10 and 17–18 appear to represent a J (or Judean) revision that justifies the Judean practice of recognizing only the Levites as priests. Such a move indicates that an early form of the exodus narratives concerned with the origins of Passover and Israelʼs firstborn priesthood would have been extant in the Northern Kingdom of Israel. 44  Sweeney, 1 and 2 Kings (see n. 12), 172–182; idem, Reading the Hebrew Bible after the Shoah: Engaging Holocaust Theology (Minneapolis: Fortress, 2008), 67–72. 45  See M. A. Sweeney, “The Wilderness Traditions of the Pentateuch: A Reassessment of their Function and Intent in Relation to Exodus 32–34,” in Society of Biblical Literature 1989 Seminar Papers (ed. D. Lull; Atlanta: Scholars Press, 1989), 291–299, which demonstrates the dependence of Exod 32–34 on Dtr and DtrH texts, thereby laying a foundation for a late-monarchic J narrative; cf. C. E. Anderson, “The Tablets of Testimony and a Reversal of Outcome in the Golden Calf Episode,” Hebrew Studies 50 (2000), 41–65, who points to the P redaction of the narrative.

870

Marvin A. Sweeney

We may recognize this early form of the exodus narrative as an E (or Ephraimite) narrative that would have been available to Hosea in the latter portion of the eighth century BCE. We do not know if he would have known these narratives in oral or written form, but they at least provide the basis for his use of pentateuchal elements in Hos 12 to support his argument that Israel should reject its alliance with Assyria, that led it to trade oil with Egypt from which Israel fled in the exodus, in favor of an alliance with Aram, where its founding ancestor, Jacob, found his wives and made peace with his father-inlaw, Laban.

V The preceding discussion demonstrates that an early E (or Ephraimite) form of the Jacob/Joseph, exodus, and wilderness narratives must have existed in northern Israel during the height of the power of the Northern Kingdom. It follows then that Hosea was in a position to know these narratives, either in oral or written form, and to cite them as part of his argument that Israel should abandon its alliance with Assyria in favor of an alliance with Aram. The question of Israelʼs alliance with Assyria was a major issue that ultimately led to a struggle for control of the throne in northern Israel that spanned the reigns of its last six monarchs, and finally to the destruction of the kingdom itself in 724–721 BCE when Israel revolted against Assyria and Assyria invaded to put down the revolt. The debate over Israelʼs future saw the demise of four of Israelʼs monarchs as the pro-Assyrian and pro-Aramean factions fought it out. King Zechariah ben Jeroboam, the last monarch of the pro-Assyrian house of Jehu, was assassinated only six months into his reign (2 Kgs 15:8–12). His assassin and successor, the pro-Aramean Zimri, ruled for only a month, until he in turn was assassinated by Menahem ben Gadi, who submitted to King Pul (Tiglath Pileser III) of Assyria (2 Kgs 15:13–16, 17–24). Menahem died of natural causes, but his son Pekahiah was assassinated by Pekah ben Remaliah, who proceed to establish the Syro-Ephraimite alliance as part of a larger strategy to throw off Assyrian rule (2 Kgs 15:23– 26, 27–31). Pekah was assassinated by Hoshea ben Elah when the Assyrians first invaded in 734–732 (2 Kgs 15:30), but he in turn revolted against Assyria in 724–721 and was deposed and imprisoned for his actions, and Israel was destroyed (2 Kgs 17:1–6). Given the chaos that engulfed the Northern Kingdom of Israel during the last years of its existence, it is likely that Hosea – or at least his book – came to Judah either to escape pro-Assyria Israelite monarchs who would have been hostile to his views or to escape the Assyrians who were not likely to show mercy to anyone in Israel, especially during the course of their second invasion in 724–721 BCE. The anti-Assyrian views of the book of Hosea

Hoseaʼs Reading of Pentateuchal Narratives

871

would have found a welcome home in Judah during the reigns of Hezekiah, who unsuccessfully revolted against Assyria in 705–701 BCE, and his great grandson, Josiah, who attempted to restore Judah and Israel following the collapse of the Assyrian Empire in the late seventh century BCE. We may posit a similar scenario for the proposed E stratum of the Pentateuch, which itself would have been brought south to Judah as the Northern Kingdom of Israel was destroyed. In the aftermath of that destruction, a J (or Judean) redaction of the Ephraimite narratives would have produced an EJ edition of the work that reflected Judean interests and ultimately formed the basis for later expansion by the addition of D (or Deuteronomy) and the final redaction by the P (or Priestly) stratum which ultimately produced the final form of the Pentateuch in the Persian period.

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah Reinhard Achenbach 1 Old Rules with Respect to the Seventh-Day Rest and with Respect to the Sabbath Day The oldest version of the commandment to rest on the seventh day for the sake of the cattle and the personae miserabiles – (female) slaves and foreign residents – is encountered in Exod 23:12. It is formulated in parallel with the commandment to leave the soil fallow after six years (Exod 23:10–11). Both rules belong to the traditional Privilegrecht, the “law of divine privilege of YHWH,” who owns and protects soil, humans, and beasts (“ox and ass,” Exod 23:12). The reason offered is to provide the animals that support the agricultural work with rest. “To do work” in that context implicitly means to bear burdens. In addition, the weakest members of the social community – children of female slaves and resident aliens – are mentioned among those who should enjoy the recreation. The age and origin of the text are under continuous debate.1  That Exod 23:12 is the oldest version of the seven-day rest rule is commonly accepted among scholars. For the tradition of a seventh-day rule in preexilic times, see A. G rund , Die Entstehung des Sabbats: Seine Bedeutung für Israels Zeitkonzept und Erinnerungskultur (FAT 75; Tübingen: Mohr Siebeck, 2011), 19–66. The “seventh day” in Deut–2 Kgs (except Deut 5) is mentioned in Deut 16:8 (“After eating unleavened bread six days, you shall hold a solemn gathering for YHWH, your God, on the seventh day: you shall do no work!”); Josh 6:4, 15, 16 (the fall of Jerichoʼs walls on the seventh day); Judg 14:15 (the riddle is solved on the seventh day); 2 Sam 12:18 (the kingʼs child dies on the seventh day); and 1 Kgs 20:29 (battle against the Arameans on the seventh day). On the special meaning of the number seven in ritual contexts see E. Otto, “‫ ֶשׁ ַבע‬u.a.,” ThWAT 7:1000–1027. For further research on the text in connection to the Sabbath see J. B riend , “Le sabbat en Jr 17,19–27,” in Mélanges bibliques et orientaux en lʼhonneur de M. Delcor (ed. A. Caquot et al.; AOAT 215; Neukir­ chen-Vluyn: Neukirchener Verlag, 1985), 23–35; G. Robinson, The Origin and Development of the Old Testament Sabbath: A Comprehensive Exegetical Approach (BBET 21; Frankfurt am Main: Peter Lang, 1988); T. Veijola , “Die Propheten und das Alter des Sabbatgebots,” in Prophet und Prophetenbuch: Festschrift O. Kaiser (ed. V. Fritz et al.; BZAW 185; Berlin: de Gruyter, 1989), 246–264; M. W einfeld , “Jeremiah and the Spiritual Metamorphosis of Israel,” ZAW 88 (1976), 17–56; C. Houtman, Das Bundesbuch: Ein Kommentar (DMOA 24; Leiden: Brill, 1997), 274–293; for a most profound analysis of Jer 17:19–27, see C. Maier, 1

874

Reinhard Achenbach

Exodus 34:21

Exodus 23:12

‫ששת ימים תעבד‬ ‫וביום השביעי תשבת‬

‫ששת ימים תעשה מעשיך‬ ‫וביום השביעי תשב‬

‫בחריש ובקציר תשבת‬

‫למען ינוח שורך וחמרך וינפש בן אמתך והגר‬

In his recent study of the festival calendars, Shimon Gesundheit has shown again that Exod 34:18–26 is a post-D and post-P composition.2 However, the text partially repeats the sentence from the Covenant Code (Exod 23:12a // 34:21a), leaving out Exod 23:12b, and deliberately adds that the rule of the seventh day should also be observed during the intensive periods of plowing and harvesting (Exod 34:21b). During the exilic period, the Deuteronomistic and the Priestly reformulations of the rule for resting on the seventh day are explicitly connected with the day of Sabbath (Akk. šabbatu/šapattu/šabattu, “full moon day”).3 The Dtr version of the Decalogue has been expanded under the influence of Priestly scribes.4 Exodus 20:8–11 (D and P s)

Genesis 2:2–3 (P s)

‫שמור את יום‬ ‫השבת לקדשו‬ ‫כאשר צוך יהוה אלהיך‬

‫זכור את יום‬ ‫השבת לקדשו‬ ‫ששת ימים תעבד‬ ‫ועשית כל מלאכתך‬

Deuteronomy 5:12–15 (D)

‫ויכל אלהים ביום‬ ‫השביעי מלאכתו אשר עשה‬

‫ששת ימים תעבד‬ ‫ועשית כל מלאכתך‬

Jeremia als Lehrer der Tora: Soziale Gebote des Deuteronomiums in Fortschreibungen des Jeremiabuches (FRLANT 196; Göttingen: Vandenhoeck & Ruprecht, 2002), 205–224. Grund, Die Entstehung des Sabbats (see above), 67–149, 305–308, finds that the verb šābat in Exod 23:12 and 34:21 does not refer to the noun Sabbath; it means “to finish doing something.” The seven-day rule wants to introduce a balance of workload and rest according to the šəmiṭṭâ. The preexilic Sabbath day is not identical with the rule intended in Exod 23; Sabbath and new moon are, together with other days of ritual performance, part of a cultic calendar; see 2 Kgs 4:23; Hos 2:13; Isa 1:13–14. For the halakah on the Sabbath in Jewish antiquity, see L. Doering, Schabbat: Sabbathalacha und -praxis im antiken Judentum und Urchristentum (TSAJ 78; Tübingen: Mohr Siebeck, 1999). 2  S. Gesundheit, Three Times a Year: Studies on the Festival Legislation in the Pentateuch (FAT 82; Tübingen: Mohr Siebeck, 2012), 21–43. 3  AHw 1172; the postexilic texts in which it appears are Isa 66:23; Ezek 45:17; Lev 23:32. 4  T. Krüger, “Schöpfung und Sabbat in Genesis 2,1–3,” in Sprachen – Bilder – Klänge: Dimensionen der Theologie im Alten Testament und in seinem Umfeld; Festschrift für Rüdiger Bartelmus (ed. C. Karrer-Grube et al.; AOAT 359; Münster: Ugarit-Verlag, 2009), 155–169, has convincingly shown that the seven-day scheme and the theology of Sabbath have been introduced secondarily into the composition by Priestly scribes. The concepts of these scribes are connected to the Holiness Code and the school behind it; see Exod 16:23, 25, 29; 31:13–16; 35:2; Lev 19:3, 30; 23:3, 11, 15, 32, 38; 24:8; 25:2, 4, 6, 8; 26:2, 34, 43.

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

‫ויום השביעי שבת‬ ‫ליהוה אלהיך לא‬ ‫תעשה כל מלאכה‬ ‫אתה ובנך ובתך‬ ‫עבדך ואמתך‬

‫וישבת ביום השביעי‬ ‫מכל מלאכתו אשר עשה‬

‫ובהמתך וגרך אשר בשעריך‬

‫כי ששת ימים עשה‬ ‫יהוה את השמים ואת‬ ‫הארץ את הים ואת‬ ‫כל אשר בם וינח‬ ‫ביום השביעי על כן‬ ‫ברך יהוה את יום‬ ‫השבת ויקדשהו‬

‫ויברך אלהים את יום‬ ‫השביעי ויקדש אתו‬ ‫כי בו שבת מכל‬ ‫מלאכתו אשר ברא‬ ‫אלהים לעשות‬

875

‫ויום השביעי שבת‬ ‫ליהוה אלהיך לא‬ ‫תעשה כל מלאכה‬ ‫אתה ובנך ובתך‬ ‫ועבדך ואמתך‬ ‫ושורך וחמרך וכל‬ ‫בהמתך וגרך אשר בשעריך‬ ‫למען ינוח עבדך‬ ‫ואמתך כמוך‬ ‫וזכרת כי עבד היית‬ ‫בארץ מצרים ויצאך‬ ‫יהוה אלהיך משם‬ ‫ביד חזקה ובזרע‬ ‫נטויה על כן צוך‬ ‫יהוה אלהיך לעשות‬ ‫את יום השבת‬

The core of the presumably oldest version of the Decalogue – the Dtr version – in Deut 5:13–14 expands the wording of the commandment to accord with the other laws of protection and ritual participation in the Deuteronomic collection (cf. Deut 16:11, 14; 24:19–21) that integrate the personae miserabiles into the house of the citizen in the community with the sons and daughters and even includes all sorts of cattle (Heb. ‫)וכל בהמתך‬. At the core of the Deuteronomic law collection, the rule of the “seventh” is only applied to the šəmiṭṭâ (Deut 15:1), the manumission (Deut 15:12), and the seventh day of Matsot (Deut 16:8). The seventh-day rule concerning rest – Heb. ‫ שבת‬5 – is neither repeated nor explained anywhere in the Deuteronomic corpus (Deut 12–26). The seventh-day rest has no special position among the narratives of the Deuteronomistic History either, nor is it mentioned in the non-P pentateuchal or Deuteronomistic narratives. However, alongside the seventh-day rule, a preexilic tradition to celebrate the full moon as the Sabbath is mentioned (Heb. ‫)שבת‬6 in 2 Kgs 4:23 as a day when one can properly visit a holy man and in 2 Kgs 11:5, 7, and 9 as a day on which the royal guards have special service rules. The existence of a facility called mûsak, an umbrella roof or “Sabbath shelter” (Heb. ‫מוסך השבת‬, 2 Kgs 16:18), in the temple precinct seems to be a marker for the special meaning of Sabbaths  W. Gesenius, Hebräisches und Aramäisches Handwörterbuch über das Alte Testament (18th ed.; Heidelberg: Springer, 2013), 1320–1321. 6  W. G esenius , Hebräisches und Aramäisches Handwörterbuch, 1321–1322; cf. 2 Kgs 4:23 (“He said: Why do you want to go to the man of God? Today is not new moon day nor is it a Sabbath!”); 2 Kgs 11:5, 7, 9 (royal guards, coming in on the Sabbath to keep watch in the kingʼs palace); 2 Kgs 16:18 (with respect to the king of Ashur Ahaz changing the “Hall of Sabbath” [‫ ]מוסך השבת‬in the temple! the term ‫[ מוסך‬ketib ‫ ]מיסך‬may be derived from ‫סכך‬, “to cover”; cf. Ps 121:5–6). 5

876

Reinhard Achenbach

in preexilic times. Of course, for each datum of evidence in the DtrH narrative, scholars may debate the dating. During the exile the rule of resting (Heb. ‫ )שבת‬on the seventh day was connected with the rule to keep the Sabbath (or full-moon day) as a holy day (Deut 5:12a and 14a); the versions of Exod 23:12a and 34:21 were combined. A. Grund argues that the Sabbath was integrated into the concept of the seventh-day rest under the influence of priestly circles.7 The Deuteronomists who inserted the Decalogue as a frame into their new edition of the Deuteronomic law in the late exilic period8 assumed a long prehistory of both rules and stated that the seventh-day rule and the Sabbath-day rule had their roots in the time of Moses. The authors of the creation myth in the Priestly narrative have produced an etiology for the Sabbath day and identified it with the seventh day of creation (Gen 2:2–3). This idea was added to the Dtr Decalogue (Exod 20:8, 11) and adjusted to this new perspective (Deut 5:12b) when Deuteronomy was integrated into the composition of the Hexateuch.9 According to the narrative of the Priestly code in Exod 24:16, the model of the holy tabernacle is revealed to Moses on the seventh day after the kābôd of YHWH has descended upon Mount Sinai. Here the term Sabbath is not mentioned, but the account seems to be correlated to the calendar system of Sabbaths. The Priestly legend of Godʼs instructions about the Sabbath in the desert is an explication of the commandment in the Decalogue (cf. Exod 16:26–27, 29–30). The rule derived from this legend is not to go out from dwelling places on the seventh day in order to collect and bear something home. In the Priestly regulations concerning offerings (Lev 1–7), neither a seventh-day rule nor the Sabbath is mentioned. And where a seventh-day rule has to be observed according to the regulations on ritual purity (Lev 13–15), this has nothing to do with the Sabbath. Remaining in the realm of Deuteronomistic and non-Priestly texts, there is no further explication either of the seventh-day rule or of the Sabbath-day commandment in combination. It seems that the weekly Sabbath had mainly a memorial function (‫זכור‬, Exod 20:8) during the exile and allowed the Judean-Israelite community regular religious conventions with prayers and spiritual rituals that reinforced the memory of the God of the exodus (‫שמור‬, Deut 5:12).10 No traces of any ritual impact from the preexilic full-moon Sabbath remain  Grund, Die Entstehung des Sabbats (see n. 1), 151–176.  At least two settings can be discerned for the Dtr frames of the preexilic Deuteronomic law: according to the first concept, the law was revealed on Mount Horeb (Deut 5:9–10* + 6–8 and 11); according to the second, the law was promulgated by Moses in the plains of Moab (Deut 1–3; 28:69 etc.). E. O tto , Deuteronomium 1–11, Erster Teilband: 1,1–4,43 (HThKAT; Freiburg: Herder, 2012), 238–248, differentiates between a “Horebredaktion” and a “Moabredaktion.” 9  Grund, Die Entstehung des Sabbats (see n. 1), 166–171. 10  Grund, Die Entstehung des Sabbats (see n. 1), 182–187. 7 8

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

877

on the exilic or early postexilic seventh-day Sabbath.11 According to Exod 34, the postexilic temple community in Jerusalem was understood as a community under the condition of a renewed covenant that, along with the traditional festivals of pilgrimage to the temple three times a year (Exod 34:23), particularly stressed the observance of the seventh-day Sabbath during the times of plowing and planting and the times of harvesting connected to the main festivals (Exod 34:21). Ritual obligations for burnt offerings and meal offerings at the Second Temple for every Sabbath were connected with the traditional commandment to rest on the holy day (Lev 23:3, H) only under the influence of Priestly groups around the Zadokites and Ezekiel (cf. Ezek 46:4–5), together with the Holiness Code and the later priestly calendar in Num 28:9–10. A sanction against profaning the Sabbath day is mentioned not in the Covenant Code, Deuteronomy, or the Holiness Code but in the redactional parenetic text of Exod 31:12–17. But what about the sermon of Jeremiah on the Sabbath in Jer 17? Does this text convey any data about a teaching on some preexilic seventh-day Sabbath?

2 Jeremiahʼs Sermon on the Sabbath in Jer 17 2.1 Introduction Many scholars doubt that Jer 17 has anything to say about the historical Jere­ miah. The Deuteronomists do not mention him. The book was redactionally shaped as evidence of prophetic teaching in accordance with the late Dtr theory on the function of prophets in the salvation history of YHWH (2 Kgs 17:23; Deut 18:15–22) only under the late influence of the Deuteronomistic scribes. Thus the picture of Jeremiah one gleans from the book is that of a continuous teaching in the tradition of the teaching of Moses (cf. Jer 1:9; Deut 18:15). The composition wants us to read the book as a continuation of the Torah in the shape of an originally inspired prophetic speech. As Moshe Weinfeld, Wilfried Thiel, and – some years ago – Christl Maier in her book Jeremia als Lehrer der Tora12 have already shown, the style of the sermons of Jeremiah follows schematic rhetorical paradigms. The introduction of Jeremiahʼs sermon follows a certain scheme of narration that lets the prophetʼs work appear as a paradigmatic form with several parallels within the book itself. Similar passages to this form also exist in the Deuteronomistic frame of Deuteronomy (Deut 5:1, 28b, 31; 6:3). 11  M. Köckert, “Ein Palast in der Zeit: Wandlungen im Verständnis des Sabbatgebotes,” in Leben in Gottes Gegenwart: Studien zum Verständnis des Gesetzes im Alten Testament (FAT 43; Tübingen: Mohr Siebeck, 2004), 109–151, 130–131; G rund , Die Entstehung des Sabbats (see n. 1), 183–184. 12  See n. 2.

878

Reinhard Achenbach

Jeremiah 17:19–20

Jeremiah 19:1–3

‫כה אמר יהוה אלי‬ ‫הלך‬ ‫ועמדת בשער בני עם‬ ‫אשר יבאו בו מלכי יהודה‬ ‫ואשר יצאו בו ובכל שערי‬ ‫ירושלם‬

‫כה אמר יהוה‬ ]. . .[ ‫הלוך וקנית בקבק‬ ]. . .[ ‫ויצאת אל גיא בן הנם‬ ]. . .[ ‫וקראת שם את הדברים‬

‫ואמרת אליהם‬ ‫שמעו דבר יהוה‬ ‫מלכי יהודה‬ ‫וכל יהודה‬ (LXX: και ανδρες Ιουδα) ‫וכל ישבי ירושלם הבאים‬ ‫בשערים האלה‬

‫ואמרת‬ ‫שמעו דבר יהוה‬ ‫מלכי יהודה‬ ]. . .[ ‫וישבי ירושלם‬

Jeremiah 22:1–2 ‫כה אמר יהוה‬ ‫רד בית מלך יהודה‬

‫ודברת שם את הדבר הזה‬

‫ואמרת‬ ‫שמע דבר יהוה‬ ‫מלך יהודה‬ ‫הישב על כסא דוד אתה‬ ‫ועבדיך ועמך הבאים‬ ‫בשערים האלה‬

Jeremiah 17:19–20 Thus said YHWH unto me: “Go, and stand in the gate of the children of the people, whereby the kings of Judah come in, and by which they go out, and in all the gates of Jerusalem; and say to them: ‘Hear the word of YHWH, ye kings of Judah, and all Judah, and all the inhabitants of Jerusalem, that enter in by these gates!’ ”

Prophetic Messenger Formula Self-Report of the Prophet (LXX, but see 4QJera) Mission Place

Additional Places Message Call to Listen Addressees: Kings Judeans Citizens of Jerusalem “Coming In”

The story follows the conventional structure of a prophetic mission narrative. The introductions of these narratives all have stylistic similarities (cf. Jer 7:1–15, 16–26; 7:27–8:3; 11:1–17; 17:19–27; 19:1–15; 22:1–5; 26:1–24; 32:8–22). Often the place of mission is a gate, which served as a place of public communication, commerce, legal debates, and trials (Jer 1:15, the cities of Jerusalem and Judah; 7:2, the gate of the temple; 22:1, the gate of the royal palace; etc.; cf. Jer 19:1–2; 20:2; 26:10; 36:10).13 The place of the mission in Jer 17 is a gate with a special name mentioned nowhere else, the “gate of the children of the people” (LXX, “your people”).14 This place has a symbolic meaning, as it allows the general populace to move into the city and its markets. Consequently, 13 14

 Gates are mentioned in Deut 16:18; 17:2, 5, 8; 26:12, etc.  ‫ ;בני העם‬cf. Jer 17:19; 26:23; 2 Chr 35:5.

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

879

according to v. 20, the mission is extended to all gates of Jerusalem and all the cityʼs inhabitants. The sermon is clearly fictional, because a sermon cannot be directed toward the plurality of the “kings of Judah” who “sit on the throne of David” (cf. 17:25; 22:2, 4), i.e., the dynasty of David after Jehoiakim (Jer 36:30) and Jehoiachin (Jer 22:30). We find sermons for the kings of Judah only in the book of Jeremiah, because the composition interprets his words as prophetic warnings for all the kings of Judah during his lifetime (Jer 1:18; 8:1; 17:19–20; 19:3–4, 13; 20:4; 33:4; 44:9).15 The introduction of the sermon with the call to listen – ‫שמעו דבר‬ ‫ – יהוה‬is frequent in the book of Jeremiah at the opening of prophetic speeches (Jer 2:4; 7:2; 19:3; 21:11; 22:2, 29; 29:20; 31:10; 34:4; 42:15; 44:24, 26).16 It is always part of the narrative framing; in the Dtr context, the prophetʼs speeches announce the downfall of the Northern and Southern Kingdoms, the same as in the prophetic scriptures. The common concern of the addressees, be they kings or citizens from Jerusalem or Judah, is their “coming in” and “going out.”17 A special cluster of symbolic ritual interaction seems to prevail in the narrative of King Joashʼs coronation ceremony, which says that the Sabbath guards of the temple had to attend the king “in his comings and goings” (2 Kgs 11:7–9, 10–12). If the full-moon day in preexilic times was a day of special advertence of the deity appropriate for issuing oracles (2 Kgs 4:23) and perhaps connected with certain taboos (Amos 8:5), the special date of the ceremony may have played an important role during the crisis of rebellion. However, we simply do not know the role of the king in preexilic Sabbath rituals,18 and the literary unity of 2 Kgs 11 can be doubted. It 15  Jeremiahʼs speech toward the king of Judah (‫)מלך יהודה‬: Jer 22:1–2, 6, 11, 18, 24; 27:18, 20; Zedekiah: Jer 21:7, 11; 24:8; 27:3, 12; 28:1; 29:2; 32:1–4; 34:2, 4, 6, 21; 37:1, 7; 38:22; 39:1, 4, 6; 44:30; 49:34; 51:59; Jeconiah: 24:1; Jehoiakim the son of Josiah: 25:1; 26:1; 27:1, 20; 28:4; 35:1; 36:1, 9, 28, 29, 30, 32; 45:1; 46:2; Amon: 24:3; Hezekiah: 15:4; 26:18–19; 34:2, 4, 6, 21; Jehoiachin: Jer 52:31; about the kings of Judah: Jer 1:18; 8:1; 17:19; 19:3–4, 13; 44:9; 45:1–2, 34; 51:59; 52:31. 16  The formula also appears in the introduction to the prophetic speeches of Micaiah (1 Kgs 22:19 // 2 Chr 18:18), Elisha (2 Kgs 7:1), Hosea (Hos 4:1), Amos (Amos 7:16), and Isaiah (2 Kgs 20:16; Isa 1:10; 28:14; 39:5; 66:5), and as an exhortation to listen to the word just related; see Ezek 13:2; 16:35; 34:9; 36:1; 37:4. The formula is never an essential part of the prophetic speech but part of a narrative style of scribal presentation. The Dtr texts have the announcement of the downfall and the exile of the population of Israel and Judah in mind. 17  “Coming in and going out” are events in everyday life that need special religious attention and awareness, see Deut 28:6: ‫ברוך אתה בבאך וברוך אתה בצאתך‬, and Deut 28:19: ‫ארור‬ ‫אתה בבאך וארור אתה בצאתך‬. 18  In late postexilic times, the order of guards in the temple was an obligation of the Levites (1 Chr 9:17–27). C. Levin, Der Sturz der Königin Atalja: Ein Kapitel zur Geschichte Judas im 9. Jahrhundert v.Chr. (SBS 105; Stuttgart: Katholisches Bibelwerk, 1982), 5–43, argues that vv. 7 and 9 are part of a postexilic reworking of the text that already has in mind the identification of the seventh-day rest and the Sabbath.

880

Reinhard Achenbach

does not really seem possible to illuminate the textual relationship between Jer 17 and 2 Kgs 11:7 and 9. After all, the central concern of the Jeremiah sermon does not allude to any special position of the kings in the context of the Sabbath ritual but focuses only on the kingʼs responsibility for the divine law; the text is written as a fulfillment of the conditions (cf. 2 Kgs 17:13–23) and already has the unity of the rules for rest on the seventh day and for the Sabbath day in mind. In this regard it is anachronistic, because it demands that kings should have supervised the prohibition against bearing burdens on the Sabbath day and through the gates into the city of Jerusalem in order to ensure that even in later times Davidic kings would reign in the city (Jer 11:25; 22:4). The reader of Jeremiahʼs sermon who is aware of the fact that the Judean kingdom has collapsed is encouraged to think over the consequences of this failure for the future observance of the Sabbath in the postexilic period. Now the main concern is the claim that commercial activity inside the city should be forbidden on the Sabbath day. 2.2 The Observance of Sabbath and the Prohibition of Trade Jeremiah 17:21–23 and Dtr Parallels Thus saith YHWH: Take heed to yourselves, and bear no burden on the Sabbath day, nor bring it in by the gates of Jerusalem; neither carry forth a burden out of your houses on the Sabbath day, neither do ye any work: but hallow ye the Sabbath day, as I commanded your fathers. But they hearkened not, neither inclined their ear, but made their neck stiff, that they might not hear, and might not receive instruction.

Nehemiah 13:15, 19b

‫כה אמר יהוה‬ ‫השמרו בנפשותיכם‬

‫בימים ההמה ראיתי‬ ‫ביהודה דרכים גתות בשבת‬ ‫ומביאים הערמות ועמסים על החמרים‬ ‫ואף יין ענבים ותאנים‬ ‫וכל משא ומביאים ירושלם ביום השבת ואל תשאו משא ביום השבת‬ ‫והבאתם בשערי ירושלם‬

‫ואעיד ביום מכרם ציד‬

‫ולא תוציאו משא מבתיכם‬

‫ומנערי העמדתי על השערים‬ ‫לא יבוא משא ביום השבת‬

‫ביום השבת‬ ‫וכל מלאכה לא תעשו‬ † ‫וקדשתם את יום השבת‬ *

‡

‫כאשר צויתי את אבותיכם‬ § ‫ולא שמעו‬ ‫ולא הטו את אזנם‬ ‫ויקשו את ערפם‬ ‫לבלתי שומע‬ ‫ולבלתי קחת מוסר‬

 ‫לא תעשה כל מלאכה‬: Deut 5:14 // Exod 20:10  ‫שמור את יום השבת לקדושו‬: Deut 5:12 ‡  ‫כאשר צויתי את אבותיכם‬: Judg 2:20; 3:4; 1 Kgs 8:58; 2 Kgs 17:13; Jer 11:4 §  ‫ולא שמעו‬: Jer 7:24, 26; 11:8; 25:4; 34:14; 44:5 * †

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

881

There is no clear reference in the Pentateuch or in any other part of the Ennea­ teuch for the prohibition against bearing burdens on the Sabbath mentioned in Jer 17:21b–22 as a practice connected to the holiness and sanctification of the Sabbath (cf. Jer 17:24, 27). There is evidence from some Elephantine texts that the Sabbath day – ‫ – יום שבת‬was known there (CG 205), that people were expected to travel and to “arrive” on the Sabbath day (Berlin P 11 367), and that people were involved in trade, as is evidenced in Ostracon CG 152: “Blessing, Yislah! Now, see, tomorrow I shall send you vegetables. Meet the boat tomorrow, Sabbath, so that it will not be lost. By the life of YHH! Otherwise I shall take you!”19 The fact that the specific time of the Sabbath day is noted shows that it was a special occasion in the calendar; however, there is no further taboo mentioned explicitly that could exclude open trade. We find the only parallel for Jer 17 in the so-called memorial of Nehemiah: Nehemiah 13:15, 19 In those days I saw in Judah some men treading wine-presses on the Sabbath, and bringing in sheaves, and loading asses therewith; as also wine, grapes and figs and all manner of burdens, which they brought into Jerusalem on the Sabbath day, and I testified against them in the day wherein they sold victuals. And it came to pass that, when the gates of Jerusalem began to be dark before the Sabbath, I commanded that the doors should be shut, and commanded that they should not be opened till after the Sabbath: and some of my servants I set over the gates, that there should no burden be brought in on the Sabbath Day.

Nehemiah has no torah to rely on, except that there is a taboo on the Sabbath that prevents all sorts of work. “Work” is defined in several terms, one of which is the “bearing of burdens.” The text shows that there is no fundamental tendency among the population of Jerusalem to confine trade even within the gates of the city or the holy precinct. Jerusalem during the Persian period was open to international trade. Isaiah 60:4–11 suggests that hope for the cityʼs increasing welfare due to trade from Arabian and Mediterranean countries led to support for a buildup and reorganization that was rather extensive at times. 20 The text 19  Doering, Schabbat (see n. 1), 23–36 and literature cited on 590, on ostraca 42; 44; 152; 186; and 205 of the Clermont-Ganneau collection (with reference to A. D upont-S ommer , “Lʼostracon araméen du Sabbat [Collection Clermont-Ganneau no 152],” Semitica 2 [1949], 29–39; idem , “Sabbat et Parascève à Éléphantine [Collection Clermont-Ganneau no 186],” RSO 23 [1957], 403–409; idem, “Un ostracon araméen inédit dʼÉléphantine [Collection Glermont-Ganneau no 44],” in Hebrew and Semitic Studies presented to G. R. Driver in Celebration of his 70th Birthday, 20 August 1962 [ed. D. W. Thomas and W. D. McHardy; Oxford: Oxford University Press, 1963], 53–58), and on the ostracon Berlin P 11367 (according to E. Sachau, Aramäische Papyrus und Ostraka aus einer jüdischen Militär-Kolonie zu Elephantine: Altorientalische Sprachdenkmäler des 5. Jh. v.Chr., Bd. I–II [Leipzig: J. C. Hinrichs, 1911], no. 67:6). 20  Isa 61, “4 Raise your eyes and look about: They have all gathered and come to you. [. . .] 5 As you behold, you will glow, your heart will throb and thrill, for the wealth of the sea shall

882

Reinhard Achenbach

expresses the hope of the pious community of the Trito-Isaianic confessors of a Zion-oriented piety for economic effect, namely, that it would be “aliens” who “shall rebuild your walls” (Isa 61:10), and has the optimistic vision that “your gates shall always stay open – day and night they shall never be shut – to let in the wealth of the nations.” However, after the wall ascribed to Nehemiah had been built together with the sanctuary in the central area of Jerusalem, the small town of Jerusalem demanded protection of the temple precinct and of the cultic performances on the holy days, as well as restriction of the inflow of foreign merchants and trade.21 The only historical evidence for a discussion concerning official regulations on trade and commerce concluding a prohibition of trade within the walls of Jerusalem on the Sabbath can be found in the literary core of the public ʾămānâ (“obligation”) from the time when Nehemiah was tiršātāʾ of the mədînâ of Yehud, when the assembly of Jerusalem decided not to accept merchandise in the area of Jerusalem:22

pass on to you, the riches of the nations shall flow to you, 6 dust clouds of camels shall cover you, dromedaries of Midian and Ephah, they all shall come from Sheba, they shall bear gold and frankincense, and shall herald the glories of YHWH. 7 All the flocks of Kedar shall be assembled for you, the rams of Nebioth shall serve your needs, they shall be welcome offerings on my altar, and I will add glory to my glorious House! [. . .] 9 Behold, the coastlands await me, with ships of Tarshish in the lead, to bring your sons from afar, and their silver and gold as well, for the name of YHWH your God, for the Holy One of Israel, who has glorified you.” (Translation according to A. Berlin and M. Z. Brettler [eds.], The Jewish Study Bible [Oxford: Oxford University Press, 2004].) 21  Limitations of public dealings on holy days in cities with sanctuaries is nothing special in itself. In Athens, for example, trials or judicial proceedings were not allowed on certain public holidays; see Pseudo-Xenophon, Constitution of the Athenians 3.8 (the constitution was written in the second half of the fifth century BCE, the report around 400 BCE; see G. Weber, Pseudo-Xenophon: Die Verfassung der Athener – griechisch und deutsch [Texte zur Forschung 100; Darmstadt: Wissenschaftliche Buchgesellschaft, 2010], 60–61). A public decision about the offering of the firstlings (aparchai) to the sanctuary of Eleusis (Athens, ca. 422 BCE, stela from Eleusis) defines measures and modalities of the delivery and determines that the archōn has to protect the sanctuaries and define the demarcations of the holy precincts, whence no one was allowed to remove stones and earth; see ML no. 73 l. 54–60 in R. Meiggs and D. Lewis, A Selection of Greek Historical Inscriptions (rev. ed.; Oxford: Oxford University Press, 1988), German translation in Historische griechische Inschriften in Übersetzung (ed. K. Brodersen, et al.; Darmstadt: Wissenschaftliche Buchgesellschaft, 2011), no. 123, pp. 98–100. 22  Texts similar to this ʾămānâ (obligatory public commitment) are known from Asia Minor, where several poleis or amphictyonic communities left inscriptions about their dedication to a common sanctuary or decisions of peopleʼs assembly about religious, political, or legal obligations; see J. von Prott and L. Ziehen, Leges Graecorum sacrae e titulis collectae I–II (Leipzig: B. G. Teubner, 1896–1906); F. Sokolowski, Lois sacrées de lʼAsie mineure (Paris: E. de Boccard, 1955, 1969); Brodersen et al., Historische griechische Inschriften (see n. 21). A reconstruction of the historical core of Neh 10 has been proposed by T. Reinmuth, “Reform

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

883

Nehemiah 10:32a: Merchandise on Sabbaths and on Holy Days ‫ועמי הארץ המביאים את המקחות וכל שבר ביום השבת למכור לא נקח מהם בשבת וביום קדש‬ And if the local people of the land bring merchandise or any kind of grain to sell on the Sabbath day, we will not buy from them on the Sabbath or on a holy day!

Neither the products of Judeans nor the wares of other “peoples of the land” would be accepted. The text states that Levites and priests and forty-five leading heads of families made an agreement about self-obligations with respect to the city of Jerusalem. Inside the newly built wall, which does not serve a defensive function but marks the area of the temenos of Zion, the commandment of Sabbath should be observed and obeyed. We do not find this regulation in the Pentateuch. There is no reference to the Decalogue or to the extensive regulation of Exod 31:12–17.23 Although the Nehemiah memorial identifies as desecration all activities of buying and selling, taking products into and out of the city (‫מהללים את יום השבת‬, Neh 13:17), the ʾămānâ does not include any sanctions or rules about the punishment of violations of the stipulation in Neh 10:32. The deliberate transgression against a common and public covenant probably would have had a sanction of kārēt (i.e., exclusion from the community) as its consequence (cf. Lev 17:9, 14; 19:8; 20:17, 18; 22:3; 23:29). The Nehemiah memorial textʼs overreaching on the precept of the ʾămānâ seems to underlie later interpretation of the circumstances in Jerusalem; the original authorship of Nehemiah has been doubted for extensive parts of it.24 Whereas the ʾămānâ is a self-obligation of the citizens to rest on the Sabbath without bargaining in the city of Jerusalem, the measures described in Neh 13:19–20 go beyond that: Nehemiah says that he has shut the gates and left them guarded by young men in order to prevent the merchants from entering the city and that this measure has been successful. According to this account, Nehemiah und Tora bei Nehemia: Neh 10,31–40 und die Autorisierung der Tora in der Perserzeit,” ZABR 7 (2001), 287–317. 23  The redactional explication of the Sabbath commandment in the context of the legend about the construction of the tabernacle in Exod 31:12–17 maintains the taboo against any sort of physical labor and demands execution for those who profane (‫ )מהלל‬the Sabbath. The legend of Num 15:32–35 is a later scribal explication of that commandment; M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1988), 98–102; H. S eebass , Numeri: Teilband Numeri 10,11–22,1 (BKAT 4/2; Neukirchen-Vluyn: Neukirchener Verlag, 2003), 154–157; C. Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25, Tübingen: Mohr Siebeck, 2007), 512–520; R. Achenbach, “Complementary Reading of the Torah in the Priestly Texts of Num 15,” in Torah and the Book of Numbers (ed. C. Frevel et al., FAT 2/62; Tübingen: Mohr Siebeck, 2013), 200–232, 225–227. 24  J. L. W right , Rebuilding Identity: The Nehemiah-Memoir and Its Earliest Readers (BZAW 348; Berlin: de Gruyter, 2004); for a more conservative evaluation, see K.-D. Schunck, Nehemia (BKAT 23; Neukirchen-Vluyn: Neukirchener Verlag, 2009), 396, 403–409, who considers vv. 19–21, 22b to be a part of Nehemiahʼs original memorial but v. 22a a secondary addition from a “levitical editor.”

884

Reinhard Achenbach

supported a position among Jerusalemʼs citizens who wanted to keep at least the traditional inner area of the town under the control of those people determined to keep the Sabbath commandment as it was written in the Decalogue and in the Covenant Code. At that time, the seventh-day, resting Sabbath was known by all Jews in the empire as a day of rest and of religious practice and memorial. As we can see from the Elephantine documents, this did not mean precluding economic practices in general. Whoever was determined to keep the covenantal law in Jerusalem must have felt obliged to refrain from “all his work” (cf. Deut 5:14) and to remember the liberation from slavery in Egypt (Deut 5:15). In addition, that meant not laying any burden on oxes or asses and not receiving any support from the personae miserabiles (Exod 23:12). Keeping merchandise beyond the wall that enclosed the city and the temple ensured that no work would be done by the citizens themselves, who promised not to buy anything and thus were forced to bear the burden home. The rule also includes the issue of circumvention, leaving the burden to the foreigner in order to avoid an offense by oneself. The sermon of Jeremiah thus fits with the intentions of those exclusivistic Yahwistic circles who – with the support of Nehemiah – wanted to have evidence from the highest authority that their interpretation of the Sabbath rule should be observed in Jerusalem. As the existing scriptures that were believed to go back to the time of Moses did not offer this help, the sustaining torah was dispersed as a prophetic voice with Mosaic authority – the voice of Jeremiah (Jer 1:9; Deut 18:15). The consequent interpretation of Deut 5:12–15 par. in a complementary reading together with Exod 23:12 to mean that bearing any burden on the holy day breaks the Sabbath rest was introduced into a fictional sermon of the prophet Jeremiah in Jer 17:19–27. The conviction behind this ingenious scribal construct was that Jeremiah was a prophet sent by God in order to remind Israel of the divine will and commandments in compliance and accordance with the Mosaic Torah. Here we find the reasoning around the theme and even the threat of severe punishment for Israel. Jer 17 is clearly influenced by the post-Dtr framing of Deuteronomy that sees the prophet as a Mosesʼs successor, as a prophetic teacher of Torah.25  E. Otto, “Der Pentateuch im Jeremiabuch: Überlegungen zur Pentateuchrezeption im Jeremiabuch anhand neuerer Jeremialiteratur,” ZABR 12 (2006), 245–306; idem , “Jeremia und die Tora: Ein nachexilischer Diskurs,” in Tora in der Hebräischen Bibel: Studien zur Redaktionsgeschichte und synchronen Logik diachroner Transformationen (ed. R. Achenbach et al.; BZABR 7; Wiesbaden: Harrassowitz, 2008), 134–182; G. F ischer , Der Prophet wie Mose: Studien zum Jeremiabuch (BZABR 15; Wiesbaden: Harrassowitz, 2010); R. A chenbach , “ ‘A Prophet like Moses’ (Deuteronomy 18:15) – ‘No Prophet like Moses’ (Deuteronomy 34:10): Some Observations on the Relation between the Pentateuch and the Latter Prophets,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 435–458. Ezek 20 seems to have been composed later than Jer 17, as Ezek 20:10–13 clearly refers to pentateuchal 25

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

885

This observation has already led to a general assumption that Jer 17:19–27 is part of a later reworking of former collections of prophetic texts.26 Thus the prophetic Torah of Jeremiah supports the idea of the observance of a religious rule at a time long after Jeremiahʼs lifetime. Whereas in Nehemiah and in the Elephantine ostraca the “bearing of burdens” has a contextual focus on trade and – in the case of Nehemiah – selling in the city of Jerusalem, the Jeremiah text remains more general. If we did not know the text from Neh 10, we would have to guess what “bearing burdens” could mean in the immediate context of Jeremiah. The text of Jeremiah gains its religious authority and legitimation by hinting about the divine commandment and the Decalogue. The text thus clearly takes its position as part of those texts that God had already promised to reveal at Mount Horeb:27 Deuteronomy 18:16–18 This is just what you asked of YHWH your God at Horeb, on the day of the Assembly, saying, “Let me not hear the voice of YHWH, my God, any longer or see this wondrous fire any more, lest I die.” Whereupon YHWH said to me (Moses), “They have done well in speaking thus. I will raise up a prophet for them from among their own people, like yourself: I will put my words in his mouth and he will speak to them all that I command him; and if anybody fails to heed the words he speaks in my name, I myself will call him to account.”

The statement that the commandment had already been given to the fathers alludes to similar texts in Judg 2:20; 3:4; 1 Kgs 8:58; 2 Kgs 17:13; and Jer 11:4. These parallels focus on the idea that it was the task of the prophets to warn the people of Judah and Jerusalem not to continue in their evil ways. Jeremiah does exactly what prophets had to do according to 2 Kgs 17:13. The negative reaction – ‫ – ולא שמעו ולא הטו את אזנם‬to the sermon is defined only in the book of Jeremiah (Jer 7:24, 26; 11:8; 25:4; 34:14; 44:5) and in 2 Kgs 17:14, where the traditions from several backgrounds: the Sabbath was proclaimed to Israel after the exodus (Exod 12:37–42) and in the desert (Exod 16), the commandment to keep the Sabbath was part of the laws “by the pursuit of which a man shall live” (see Lev 18:5; Deut 8:2–3; 30; Mic 6:8); it is viewed as a sign (Exod 31:13, 17; cf. Gen 1:14; 2:3); and desecration of the Sabbath is interpreted as a rebellious act against Godʼs law (Exod 20:8–11). 26  C. Körting and H. Spieckermann, “Sabbat I: Altes Testament,” TRE 29:518–521, here 520, assume that the text is late Deuteronomistic. K öckert, “Ein Palast in der Zeit” (see n. 11), 146, argues that the text is a sermon from the time of Nehemiah. The analysis of Veijola, “Die Propheten und das Alter des Sabbatgebots” (see n. 1), 255–258, goes in the same direction. 27  Cf. Exod 20:18–21; Deut 5:2, 22–32. The text of Deut 18:16–18 incorporates parts of the prophetic scriptures into the narrative of a divine revelation of Torah and thus helps to legitimate the production of a series of prophetic scrolls under late-Dtr influence during the fifth century in addition to the Hexateuch; see R. A chenbach , “ ‘The Unwritten Text of the Covenant’: Torah in the Mouth of the Prophets,” in Covenant in the Persian Period (ed. R. J. Bautch et al.; Winona Lake, IN: Eisenbrauns, 2015), 93–108.

886

Reinhard Achenbach

sentence “they did not listen” is combined with the critique against the “stiffnecked” opposition (‫)ויקשו את ערפם‬, as in Jer 7:26; 17:23; 19:15; Neh 9:16, 17. The combination with the sentence ‫ לקחת מוסר‬/ ‫ ולא לקחו מוסר‬is documented in Jer 7:28; 17:23; 32:33; and 35:13. Some elements allude to other late texts in the Pentateuch that try to more densely structure teaching about the Sabbath into the Torah as a whole. The introduction ‫“( השמרו בנפשותיכם‬Guard yourself,” Jer 17:21a; cf. Deut 4:9, 15; Josh 23:11; Ps 25:20) seems to allude to the threat of the death penalty for transgressions against the Sabbath (‫ושמרתם את השבת כי קדש הוא לכם מחלליה‬ ‫מות יומת‬, Exod 31:14; cf. Exod 35:2; Num 15:32–36),28 but there is no explicit connection in Jeremiah to any of the pentateuchal texts that demand the death penalty for desecrating or defiling the Sabbath (cf. Exod 31:12–17; 35:2; Num 15:32–36). Nor is there any evidence that the authors shared the view of Lev 26:34–43* that the land will remain desolate and the people will remain in the diaspora until all the Sabbath years and Sabbath days that have been neglected are atoned for. The only text from the Pentateuch comparable to Jer 17 in this sense is the tale of the wondrous provision of manna in the wilderness and the Sabbath torah in Exod 16:25–28 resulting from the prohibition “not to gather” manna on the seventh day (v. 26 // Exod 20:9–10; v. 27 interprets ‫יצאו ללקט‬, “they went out to collect,” as transgression against the Torah and the commandment of YHWH in v. 28). But Exod 16:19–36* seems instead to be a pious legend that helps to protect resting on the Sabbath in a very radical way. This legend belongs to the late Priestly layers in the Pentateuch, so it may be later than Jer 17. Another prophetic text that deals with festivals and Sabbaths can be found in Amos 8:4–7. Some scholars argue that this oracle could be a preexilic text, but more recent analyses come to the conclusion that the final composition was not written before the time of Nehemiah, because it seems to describe the possible background of the restrictive rules in Neh 10:32; 13:15–16.29

 Maier, Jeremia als Lehrer (see n. 1), 214.  G rund , Die Entstehung des Sabbats (see n. 1), 80–82. For a preexilic dating, see G. F leischer , Von Menschenverkäufern, Baschankühen und Rechtsverkehrern: Die Sozialkritik des Amosbuches in historisch-kritischer, sozialgeschichtlicher und archäologischer Perspektive (BBB 74; Frankfurt am Main: Athenäum, 1989), 444. J. J eremias , Hosea und Amos (FAT 13; Tübingen: Mohr Siebeck, 1996), 231–243, here 242–243; idem, Amos (ATD 24/2; Göttingen: Vandenhoeck & Ruprecht, 1995), 114, argues that the text is close to Mic 6:9–12 and Deut 25:13–16 and therefore assumes a date at the time of Jeremiah. Veijola , “Die Propheten und das Alter des Sabbatgebots” (see n. 1), 252–255, and K örting and Spieckermann, “Sabbat I” (see n. 26), 519, argue for a dating at the time of Nehemiah. 28

29

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

887

Amos 8:4–7 ‫שמעו זאת השאפים אביון‬ ‫ולשבית ענוי ארץ‬ ‫לאמר מתי יעבר החדש‬ ‫ונשבירה שבר‬ ‫והשבת ונפתחה בר‬ ‫להקטין איפה‬ ‫ולהגדיל שקל‬ ‫ולעות מאזני מרמה‬ ‫לקנות בכסף דלים‬ ‫ואביון בעבור נעלים‬ ‫ומפל בר נשביר‬ ‫נשבע יהוה בגאון יעקב‬ ‫אם אשכח לנצח כל מעשיהם‬

“Listen to this, you who devour the poor, and thus annihilate the humble of the Land, saying: ‘If only the new moon were over, so we could sell grain, and the Sabbath, so that we could offer wheat for sale, using an ephah that is too small,* and a shekel that is too big, tilting a dishonest scale,† to buy the needy for silver and the poor for a pair of sandals,‡ and (selling) grain-refuse as grain!’ YHWH swears by the Pride of Jacob: ‘I will never forget any of their doings!’ ”

 Deut 25:13–14  Prov 11:1; 20:23 (!); cf. Hos 12:8; Mic 6:11 ‡  Am 2:6–7 * †

Amos 8 fits into the Deuteronomistic styling of prophetic sermons; 30 the prophet gives a sermon that refers to several traditional prohibitions to refute using different weights and measures, as mentioned in Deut 25:13–15.31 But the Deuteronomic text is already embedded in a parenesis that lets Moses threaten them with loss of the promised land. Reference to the Sabbath in connection with the new moon especially seems to be an argument for a preexilic dating of the text. But Isa 66:23 shows that ḥōdeš (“new moon”) was also celebrated in postexilic times.32 Another suggestion has been offered by J. Wöhrle, who observes that we find a secondary commentary to the vision-cycle (Amos 7–9) in Amos 8:4–14a.33 Amos 8:4 and 6a refer back to Amos 2:6–7 and 6:8; Amos 8:9–10 takes up Amos 5:18, 20, 21, 23; and Amos 8:13–14 repeats thoughts from Amos 5:2, 30  See A. Schart, Die Entstehung des Zwölfprophetenbuchs: Neubearbeitungen von Amos im Rahmen schriftenübergreifender Redaktionsprozesse (BZAW 260; Berlin: de Gruyter, 1998), 157–233, here 166. Schart dates the text to the exilic period (see also idem , “The Sabbath: In the Law, in the Prophets, and in Mark,” Verbum et Ecclesia 25 [2004], 253–273, here 262–264), but why should the Deuteronomistic shape have been applied to the collection of Amosʼs oracles at that time, referring to misuses that were somehow timeless? For wrong weights, see Prov 11:1; 16:11; 20:10, 23; Lev 19:35–36; Hos 12:8; Job 31:6; CH § 108 et al.; cf. Grund, Die Entstehung des Sabbats (see n. 1), 82, n. 449. 31  Deut 25:13: “You shall not have in your pouch alternate weights, larger and smaller! 14 You shall not have in your house alternate measures, a larger and a smaller! 15 You must have completely honest weights and completely honest measures! – if you are to endure long on the soil that YHWH your God is giving you!” 32  Versus Grund, Die Entstehung des Sabbats (see n. 1), 82–83. 33  J. W öhrle , Die frühen Sammlungen des Zwölfprophetenbuches: Entstehung und Komposition (BZAW 360; Berlin: de Gruyter, 2006), 105–110, 129–135.

888

Reinhard Achenbach

4–5. The critique of the oppression of the poor in Amos 8:4, 6a is extended (and interrupted) by an explanation about how we must imagine this exploitation: vv. 5 and 6b show that the Torah provides the material to sketch what could be meant by an offense against the law (i.e., Deut 25:13–14). But even here we must say that the Deuteronomic reference is combined with wisdom teaching about the “dishonest scale” (‫מאזני מרמה‬, which we find in Prov 11:1 and 20:23 but also in Hos 12:8 and Mic 6:11). Amos 8:7 alludes to the threat of the decline of the Northern Kingdom in Amos 6:8 and 8:2. Amos 8:4–7 is thus a composite text consisting of traditional elements from Amos as well as Deuteronomic and wisdom material and extensions of that material to the scribal composition of a so-called prophetic torah! Again we find that the text of the ʾămānâ can be interpreted as the result of Nehemiahʼs work and that the prohibition against bearing burdens on the Sabbath was added to the Pentateuch at a post-Nehemianic stage. Until the work of Nehemiah, the people of Jerusalem were unaware that trade close to the temple precinct could be forbidden. The radical prohibition against all work on the Sabbath because it was a day of sacred convocation (‫מקרא קדש‬, Lev 23:3) was thus not yet implanted firmly enough into the religious tradition. The term holy day is rooted not in the Pentateuch but in the mid-fifth century BCE, when the Nehemianic measures seem to receive support: Isaiah 58:12–14 Men from your midst shall rebuild ancient ruins, you shall restore foundations laid long ago, and you shall be called “Repairer of fallen walls, restorer of lanes for habitation.” If you refrain from trampling the Sabbath, from pursuing your affairs on my holy day, if you call the Sabbath “delight,” the Lordʼs holy day “honored,” and if you honor it and go not about your ways [. . .] I will set you astride the heights of the earth, and let you enjoy the heritage of your father Jacob.

The self-obligation to hold fast to the rules of foregoing the harvest in the seventh year is rooted in the ancient prexilic tradition as we find it in the Covenant Code (Exod 23:6–7). The principle of the sixth-/seventh-year rhythm was already applied to the law of manumission in Exod 21:2. It was extended to a regular release of debts in Deut 15:1 (“Every seventh year you shall practice remission of debts”). The difficulties in upholding these religious rules are reflected in the report on economic conflicts in Neh 5:1–12, especially 5:5.34 The ʾămānâ draws out the consequences of these measures.35 These conflicts are deeply reflected in the Holiness Code, and the scribes developed a totally new sacral order that  Neh 5:5: “Now, we are as good as our brothers, and our children as good as theirs; yet here we are subjecting our sons and daughters to slavery – some of our daughters are already subjected – and we are powerless, while our fields and vineyards belong to the others.” See H. G. M. Williamson, Ezra, Nehemiah (WBC 16; Waco, TX: Word, 1985), 335. 35  S chunck , Nehemia (see n. 24), 301; T. R einmuth , Der Bericht Nehemias: Zur literarischen Eigenart, traditionsgeschichtlichen Prägung und innerbiblischen Rezeption des Ich-Berichts Nehemias (OBO 183; Göttingen: Vandenhoeck & Ruprecht, 2002), 216. 34

The Sermon on the Sabbath in Jeremiah 17:19–27 and the Torah

889

makes debt slavery virtually impossible (cf. Lev 25:19–55). At the time of Nehemiah, we may therefore suppose that people could refer to the Covenant Code and Deuteronomy regarding legal questions but that even these texts were not sufficiently acknowledged as binding religious law. At the same time, we see that Nehemiah could not yet refer to central distinctions in the Holiness Code. Thus the formation of the Holiness Code may have taken place between the time of Nehemiah and the time of Ezra. 2.3 The Promise of Blessing and Curse Jeremiah 17:24–2736 But if you will really listen to me – declares YHWH – [cf. Jer 22:4a] and do not bring in burdens through the gates of this city on the Sabbath day, but hallow the Sabbath day and do not work on it, then through the gates of this city shall enter kings who sit upon the throne of David, with their officers – riding on chariots and horses, they and their officers – and the men of Judah and the inhabitants of Jerusalem, and this city shall be inhabited for all time [cf. Jer 22:4b], and the people shall come from the towns of Judah and from the environs of Jerusalem, and from the land of Benjamin, and from the Shephelah, and from the hill country, and from the Negeb, bringing burnt offerings and sacrifices, meal offerings and frankincense, and bringing offerings of thanksgiving to the temple of YHWH. But if you do not obey My command to hallow the Sabbath day and to carry in no burdens through the gates of Jerusalem on the Sabbath day, then I will set fire to its gates, it shall consume the fortresses of Jerusalem and it shall not be extinguished! [cf. Jer 22:5]

This text is formulated like a sermon on a bərît in cases of obedience or disobedience (cf. Deut 28:1, 15). The prophetic announcement of judgment and the destruction of Jerusalem by fire stands in the stead of curses (cf. 2 Kgs 25:9; Jer 21:14; 22:5; 39:8; Lam 4:11; Hos 8:14; Amos 2:5; Zech 7:11). The destruction of the city, then, is mainly described as the result of a punishment for disobeying the commandment to keep the Sabbath. We do not yet find this understanding in the context of Nehemiah, but it does appear in some different theories in the Holiness Code (Lev 23; 25; 26:43: “For the land shall be forsaken of them, making up for its Sabbath years by being desolate of them, while they atone for their iniquity; for the abundant reason that they rejected my rules and spurned my laws.” See also Ezek 20:20–25: “the people survived in the desert although they did not obey the Sabbath, but they were heavily punished by bad laws”; and Isa 56:2–6;37 58:13; Neh 9:6–37, 14!). The topography described in the blessing (Shephelah, hillside, Negev) corresponds to the province of Yehud as described in Neh 12:25b–36.38 The series of offerings mentioned in the promise of blessings (‫עולה‬, ‫זבח‬, ‫ )מנחה‬we find in  The translation follows The Jewish Study Bible (2004).  Non-Israelites who observed the Sabbath are named in Isa 52:2, 4, 6, 13. 38  M aier , Jeremia als Lehrer (see n. 1), 216; cf. also Jer 17:26; 32:44; 33:12–13; and corresponding Dtr and post-Dtr land descriptions (Deut 1:7; Josh 9:1; 10:40; 11:16; 12:8; Judg 1:9). 36 37

890

Reinhard Achenbach

the Holiness Code (Lev 23:37), where the thanksgiving offering (Heb. ‫)תודה‬ is also mentioned (Lev 22:29). Further references to the thanksgiving offering can be found in Jer 33:10–11, 18, and in Neh 12:27, 31, 38, 40. C. Maier has shown that the paradigmatic sermon of Jeremiah reflects on the narrative of Neh 2:1–10; 13:15–22.39 Consequently, after the late-Dtr period of the formation of the Hexateuch and after Nehemiah, the book of Jeremiah was used to add explanations about the Sabbath torah that later were reflected in a more explicit form in the Holiness Code and in the redaction of the Pentateuch. The Holiness Code does not condemn to death those who transgress the Sabbath (Lev 19:3, 30; 23:3; 26:2). The radical demand of the death penalty for those who violate the Sabbath commandment in Exod 31:12–17 and Num 15:32–35 was added as a rule concerning the holy tabernacle in the desert by the Pentateuch redaction and later Priestly rewritings.

3 Conclusion The example of the prophetic Torah in Jer 17 shows that the formation of the Pentateuch is the result of a dialogue between scribes who stood behind the composition of the Hexateuch and others who were responsible for the Prophets, the Holiness Code, and the late Priestly layers of Exodus–Numbers. The sociohistorical and ritual background of Jer 17 can be developed with respect to Neh 10 and 13. The history of Jerusalem and the Second Temple as related in the books of Nehemiah and Ezra has to be correlated with internal developments in both literary corpora.

39

 Maier, Jeremia als Lehrer (see n. 1), 218–223.

‫ותפשי התורה לא ידעוני‬ The Relationship of the Book of Jeremiah to the Torah Georg Fischer This article will investigate the theme of the conference – “The Pentateuch within Biblical Literature: Formation and Interaction”1 – with respect to a very specific area, namely the connections between the Torah and the book of Jeremiah. I will first deal with the use of ‫ תורה‬in the book of Jeremiah, then present some examples of intertextual relationships between the books of the Torah and the book of Jeremiah (hereafter abbreviated Jer), and, finally, sum up the results.

1 The Use of ‫ תורה‬in the Book of Jeremiah The word ‫ תורה‬occurs eleven times in Jer.2 This is a relatively small number in comparison to other words in Jer, like “to hear,” “prophet,” and “deceit.” 3 Nevertheless, ‫ תורה‬in Jer is of considerable importance, as it is nearly always connected with God and as some passages contain unique ideas. I will present briefly all the occurrences,4 in three steps. 1.1 The Beginning The title of this article is a quote of one stich of Jer 2:8: 1  I thank Jan Christian Gertz, Bernard M. Levinson, Dalit Rom-Shiloni, and Konrad Schmid, the organizers of the conference in Jerusalem in May 2014. The presentation here is an extended version of a paper read there. 2  These eleven occurrences constitute 5 percent of the 223 occurrences within the Hebrew Bible; it is thus less than what might be expected, as Jer contains more than 7 percent of the words in the Hebrew Bible. 3  Respectively 158, 135, and 37 occurrences; in all cases, Jer has the highest number of these words within the books of the Hebrew Bible. The most frequent word in Jer is the tetragrammaton (YHWH), which occurs 726 times. 4  The main investigation of ‫ תורה‬in Jer is C. Maier, Jeremia als Lehrer der Tora: Soziale Gebote des Deuteronomiums in Fortschreibungen des Jeremiabuches (FRLANT 196; Göttingen: Vandenhoeck & Ruprecht, 2002), esp. 282–352.

892

Georg Fischer

The priests have not asked: “Where is YHWH?,” and those handling/grasping the Torah donʼt know me, and the shepherds rebelled against me, and the prophets prophesied in (the name of?) Baal, and they went after what is useless.

This is the first instance of the use of ‫ תורה‬in Jer. It mentions four responsible groups in the community, three of which have religious functions.5 All are blamed for serious deviations from their relationship with the biblical God. The second group is named ‫תופשי התורה‬. The verb ‫ תפש‬indicates normally a forceful action,6 to be translated by “to seize, capture, snatch.” It may also be used with weapons, like a bow, swords, or a shield,7 as well as with a sickle, the oar of a ship, and even musical instruments.8 In these cases, the meaning seems to be close to “to handle, to use.” The NRSV therefore also translates Jer 2:8 as “Those who handle the law did not know me,” and the NIV as “Those who deal with the law did not know me.” Thus both translations render the Hebrew verb in a neutral, mild way, which does not bring out its underlying force. This is also true for several German translations.9 Within the Hebrew Bible this combination of the verb ‫ תפש‬with ‫ תורה‬is unique. Yet it occurs again in Sir 15:1, attested to, in identical form, by the two manuscripts A and B of the Cairo Geniza.10 The first stich reads, “For (who) fears YHWH, will do this,” and is continued by ‫תופש תורה ידריכנה‬, “and (the one)  The identification of the “shepherds” remains open. This term seems to refer, in Jer, mostly to political leaders, but other responsible groups are not excluded. 6  A. Even-Shoshan, A New Concordance of the Bible (Jerusalem: Kiryat Sepher, 1980), 2299, gives as equivalences for the qal stem ‫ אחז בחזקה‬and ‫לכד‬. For Jer 2:8 (listed as number 26), he attributes the first meaning, “to seize with force.” 7  Connected with “bow” is Amos 2:15, “he who handles the bow,” and probably also Jer 46:9, where the two participles ‫ תפשי‬and ‫ דרכי‬precede the bow. Earlier in the same verse, ‫תפשי‬ precedes “shield.” Ezek 38:4 mentions those “wielding swords” within the army of Gog. Jer seems to have a preference for this expression. It contains five out of the ten occurrences of the qal participle. All translations in the footnotes are from the NRSV except where otherwise noted. 8  In Jer 50:16, Babylon shall be deprived of “the one who handles the sickle,” and “all who handle the oar” wail about the sinking ship of Tyre in Ezek 27:29. Gen 4:21 names Jubal as “the father of all those who play the lyre and the pipe.” 9  I mention only two examples: Die Bibel: Altes und Neues Testament, Einheitsübersetzung (Freiburg: Herder, 1980), renders it as “Die Hüter des Gesetzes kannten mich nicht,” suggesting a group taking care of (hüten) and being responsible for the law, thus already interpreting it in one direction. Die Heilige Schrift: Aus dem Grundtext übersetzt; Elberfelder Bibel revidierte Fassung (4th ed.; Wuppertal: Brockhaus, 1995), translates, “Und die das Gesetz handhabten, kannten mich nicht,” in a similar vein to the English versions quoted above. 10  P. Boccaccio and G. Berardi, Ecclesiasticus: Textus Hebraeus secundum fragmenta reperta (Rome: Pontifical Biblical Institute, 1986). 5

The Relationship of the Book of Jeremiah to the Torah

893

handling/grasping tôrâ will get/achieve her.” The referent of “her” is ambiguous in the various versions; in Hebrew, the next referent is ‫תורה‬, but “her” could also refer to “wisdom” mentioned before, as in the Greek manuscripts. In this text, the context and the connotations of the verb are positive.11 Having dealings with ‫ תורה‬leads to good results, and this is equally true for Sir 15:1 in the Septuagint.12 The negative background of the reproaches in Jer 2:8, however, impels us to reflect on the meaning of the verb ‫ תפש‬there. As a designation of the group, it may be neutral; this option cannot be totally excluded. Yet, the unique use of a verb normally associated with force and often combined with weapons may also indicate a similar powerful action with regard to Godʼs instruction/law.13 The translation of ‫ התורה‬must be discussed, too. It could be simply “the law, the instruction.” However, “the Torah,” namely the first five books of the Bible, could also be in view, or at least one scroll of it.14 The noun is used with the definite article here and has no preceding referent that could explain it. For the moment we must leave open the exact meaning of ‫ תורה‬in Jer 2:8 and come back to this question at the end (see 1.4). Yet the position of this passage close to the beginning of Jer, the gravity of the accusations, and the important groups mentioned in the context already allow us to surmise that we are dealing with something more than simply a normal instruction. Another ambiguity of Jer 2:8 regards the translation of the tense of the verb ‫ידע‬. It is in the qatal form, like the verbs for the three other groups in v. 8, and could also be rendered by the perfect tense: “(they) have not known me.” However, the usual meaning of this verb ‫ ידע‬in the qatal is present tense: “(they) donʼt know me.”15 This would result also in maintaining the reproach of not knowing God for the actual moment. The “me” refers to God, who is introduced in v. 5 as the speaker. Jer 2:8 raises harsh accusations early in the book. One of them is connected with a group associated with instruction/law or the Torah and denominated in a unique way. Their identity remains open and poses the question of whether further on we will find a hint pointing to their identity. The introduction of the theme ‫ תורה‬in Jer is thus rather negative. 11  In this sense, P. W. Skehan and A. A. di Lella, The Wisdom of Ben Sira (AB 39; New York: Doubleday, 1987), 264, who offer an alternative rendering, “he who grasps, or is practiced in” (the Law), where the expression refers to the scribe. 12  J. Marböck, Jesus Sirach 1–23 (HThKAT; Freiburg: Herder, 2010), 188, translates the Greek version “wird sie erlangen,” where the female sie refers back to wisdom in Sir 14:20. 13  A modern equivalent to Jer 2:8 might be the title of the book by J. S. D uvall and J. D. Hays, Grasping Godʼs Word (3rd ed., Grand Rapids, MI: Zondervan, 2012), which also uses grasp in connection with Godʼs revelation. 14  M aier , Jeremia als Lehrer (see n. 4), 293–294, and similarly 310, thinks of a written scroll of the Torah. 15  P. Joüon, Grammaire de lʼHébreu Biblique (Rome: Pontifical Biblical Institute, 1965), § 112a.

894

Georg Fischer

1.2 Further Reproaches Referring to Instruction or Torah in the Book of Jeremiah The next passages about ‫ תורה‬in Jer continue on this line. I will deal with the next five texts together, picking out only some of the main aspects. (a) Jeremiah 6:19 says at the end: “because they didnʼt pay attention to my words, and my instruction/law/Torah – they despised it.”16 This accusation of despising/rejecting divine ‫ תורה‬also occurs in Amos 2:4 and Isa 5:24, and Jer could be dependent on either passage.17 The book of Jeremiah quotes Godʼs speech, introduced in Jer 6:16, which parallels “my words” and ‫ תורתי‬and announces evil as a consequence of the negative reaction of the people toward what he says and commands. (b) The next passage, Jer 8:8, is complex. A groupʼs words are quoted: “We are wise, and YHWHʼs instruction/law/Torah is with us!” This statement suggests a positive attitude toward Godʼs instruction. However, the introduction to it already questions this conviction when it says, “How can you [in the plural] say [. . .]?” And the continuation casts still more doubt on the self-perception of this group, so firmly related to Godʼs ‫תורה‬: “Really, behold, for/to deceit the pen of deceit of (the) scribes has made (it?).”18 Twice ‫ שקר‬is repeated, first as the outcome of their actions, set in front of the phrase after the two demands for attention, and second as designation of the scribesʼ instrument, to underline the complete perversion of Godʼs just words to the contrary, to treachery and deceit. The repetition stresses that the acts as well as the results are corrupted. If Jer 8:8 is linked to 2:8, this would further enhance the negative tone in 2:8. It might also suggest a possible identification of the ‫ תופשי התורה‬as scribes. (c) In Jer 9:12 (v. 13 in English versions), God himself answers the preceding question of v. 11 asking why the country perishes: “because they have abandoned my instruction/law/Torah, which I gave before them, and didnʼt listen to my voice, and didnʼt walk in it.” The combination ‫עזב תורה‬, “abandon the instruction/law/torah,” with reference to Godʼs teaching is relatively rare in the Bible. It is found also in 2 Chron 12:1; Pss 89:31; 119:53; and Prov  Instead of “despised,” the Hebrew verb ‫ מאס‬could also be translated as “rejected.”  Amos 2:4 and Isa 5:24 are addressed to Judah, and both texts are with high probability earlier than Jer. On the relationship between Amos and Jer, generally, see J. Pschibille, Hat der Löwe erneut gebrüllt? Sprachliche, formale und inhaltliche Gemeinsamkeiten in der Verkündigung Jeremias und Amosʼ (Biblisch-Theologische Studien 41; Neukirchen-Vluyn: Neukirchener Verlag, 2001). For the relationship between Isaiah and Jer, see U. W endel , Jesaja und Jeremia: Worte, Motive und Einsichten Jesajas in der Verkündigung Jeremias (Biblisch-Theologische Studien 25; Neukirchen-Vluyn: Neukirchener Verlag 1995). 18  The Hebrew phrase lacks a direct object, so that one can only guess what the scribes falsify and pervert in this way. The translation above follows the Masoretic accentuation, but Maier, Jeremia als Lehrer (see n. 4), 298 and 300, reads in another way, taking scribes as the direct object: “Ja, siehe, zur Lüge macht der Lügengriffel Schreiber.” 16 17

The Relationship of the Book of Jeremiah to the Torah

895

28:4.19 The first three texts, in all likelihood, are later than Jer; this is probably also true for Prov 28:4. It is possible that Jer stands at the origin of this expression. As early as Jer 1:16, God started to denounce the people for “having abandoned me.”20 This is picked up and further specified in 9:12, where, in addition to God himself, his instruction is set aside, too. The disdain of divine ‫ תורה‬logically follows the defiance of God, continuing and completing it. (d) Another instance of the question-answer scheme encountered in 9:11–12 can be found in Jer 16:10–11. The people ask why God has pronounced such great evil upon them (v. 10), and God tells Jeremiah to communicate to them: “Because your fathers have abandoned me – utterance of YHWH – and have gone after other ‘gods,’ and they have served them, and they have worshiped them, but me – they have abandoned; and my instruction/law/Torah – they have not preserved/kept.”21 The exact wording of this reproach is unique, and the front position of ‫ תורתי‬underlines the weight of Godʼs instruction and therefore also of the accusation. Other combinations of ‫שמר‬, “to keep,” with Godʼs ‫ תורה‬are found elsewhere. The closest passages are: • Exod 16:28, where God asks, with ‫מאן‬, “to refuse”: “How long do you refuse to keep my commandments and ‫”?תורתי‬ • Deut 17:19: The king should learn “to keep all the words of this ‫תורה‬.” • Deut 28:58, in a warning: “if you do not preserve/keep to do all the words of this ‫( תורה‬which) are written on this scroll [. . .] .” In a similar way, Deut 31:12; 32:46 are also expanded with “to do all the words” as commands for the people. • In 1 Chr 22:12, David instructs Solomon “to keep YHWHʼs ‫תורה‬.” • Ps 119:34, 44, 55: Three times the prayer states the speakerʼs intention to stick to Godʼs law; in v. 136 he is afflicted because others do not do so: “Streams of water flow down (from) my eyes, for they do not keep your instruction/law/Torah.” • Participles in Prov 28:4 (in the plural; see above in example c) and 29:18 (in the singular) speak of an attitude of “keeping ‫תורה‬.” Of these twelve passages, only three have a negative background, thus coming close to the reproach of Jer 16:11: Exod 16:28; Deut 28:58; and Ps 119:136. As Jer often picks up Deut 28,22 Jer 16:11 might be an actualization of the condi In Prov 4:2, ‫ עזב תורה‬refers to the instruction of the father, so it is excluded from the list here. M. V. Fox, Proverbs 10–31 (AB 18B; New Haven, CT: Yale University Press, 2009), 821 understands Prov 28:4 as “ignoring” (‫“ )עזב‬wisdom instruction” (‫)תורה‬. 20  This is often repeated with variations; examples include Jer 2:13, 17, 19; Jer has twenty-five occurrences of the verb “abandon,” more than any other book of the Bible. 21  The Hebrew verb is ‫שמר‬, “to preserve”; “kept” is the rendering for it in NRSV and NIV. 22  G. Fischer, “Fulfillment and Reversal: The Curses of Deuteronomy 28 as a Foil for the Book of Jeremiah,” Semitica et Classica 5 (2012), 43–49. 19

896

Georg Fischer

tional threat there, implying that the curses following in Deut 28:59–68 will come into effect here. (e) Jeremiah 18:18 is not actually a direct reproach. This opening of the fourth confession, however, characterizes the false attitude in the community that attacks the prophet Jeremiah, thinking: “because ‫ תורה‬does not perish from the priest, and council from the wise, and word from the prophet.” The quote presents the illusional confidence placed in the religious leaders and their divine inspiration or access to revelation. This last passage is interesting because it has very close ties with Jer 2:8. Common elements are priests, prophets, and the ‫תורה‬. In Jer 2:8 the priests were mentioned separately, as well as those “handling/grasping ‫התורה‬.” Jeremiah 18:18, however, links ‫ תורה‬with the “priest.” This group, too, seems to have a close relationship with Torah – at least on the level of the peopleʼs persuasion. Jeremiah 18:18 might indicate that ‫ תופשי התורה‬in 2:8 may be a section of the priests. Taking into consideration Jer 8:8 as well, it could refer to a group of scribes among them. Looking back at these five passages, three instances (a, c, and d) are explicitly very negative, using verbs like despise/reject, abandon, and not keep for the communityʼs attitude toward Godʼs ‫תורה‬. The book of Jeremiah describes, sometimes in a unique manner, the peopleʼs complete distance from it. In contrast to this remoteness, they themselves are convinced of having it as a kind of possession (b and e) and of being strongly bonded with it. 1.3 The Other Texts Five out of the overall eleven passages in Jer remain to be dealt with. Four of them draw on a concept already mentioned, namely “to walk” in Godʼs ‫;תורה‬23 one, however, develops a new idea (Jer 31:33). (a) Walking in Godʼs ‫תורה‬ This phrase occurs first in Jer 9:12, albeit separated by a distance of several words. There, “in it” refers back to ‫ תורתי‬in the previous context. The expression combines the verb ‫הלך‬, “to go, to walk,” with the preposition ‫ב׳‬, “in,” and ‫תורה‬. This wording is encountered four more times in Jer: i. In Jer 26:4, God renews24 his warning in the temple: “if you do not listen to me, to walk in my instruction/law/‫ תורה‬which I gave before you [. . .].”  The first instance is Jer 9:12 (see above 1.2, c): “and did not walk in it.”  The setting and the general orientation of Jer 26 are similar to the first temple sermon that Jeremiah is commanded to communicate in Jer 7, but the contents and wording are quite different. See T. S eidl , “Jeremias Tempelrede: Polemik gegen die joschijanische Reform? Die Paralleltraditionen Jer 7 und 26 auf ihre Effizienz für das Deuteronomismusproblem in 23

24

The Relationship of the Book of Jeremiah to the Torah

897

ii. Jeremiah prays in Jer 32:23, after the purchase of the land in Anathoth, and confesses the peopleʼs guilt: “and they did not listen to your voice, and in your ‫ תורה‬they did not go.” iii. Jer 44:10 expresses Godʼs reproach of those who immigrated to Egypt: “and they did not fear, and they did not walk in my ‫ תורה‬and in my statutes which I had given before you.” iv. A few verses later, in Jer 44:23, Jeremiah addresses the same audience, picking up Godʼs accusation: “and you did not listen to YHWHʼs voice, and in his ‫ תורה‬and in his statutes and in his ‘testimonies’25 you did not walk.” Both passages in Jer 44 are at the end of subunits, summarizing the main accusation and stressing it through repetition. Jeremiah 44 itself, too, is an end: it concludes the discourses of God and Jeremiah to the people within the book. Thus it is a kind of final statement on the people in Jer. This also underlines the weight of “walking in Godʼs ‫ ”תורה‬and the gravity of not doing it, as here. Of the five passages with this expression, four are preceded by “not listen to . . . voice/me.”26 This seems to indicate that attention to Godʼs speaking (“to listen”) is a first step, like a precondition, after which concrete acts of obedience (“to walk in his ‫ )”תורה‬should follow. Out of the five passages, four are negated and accuse the people of past disobedience. Only Jer 26:4, as a conditional warning for the future, remains open. (b) A New Hope The “scroll of consolation” in Jer 30–31 contains many divine promises. The most famous among them is probably the announcement of a “new covenant” in Jer 31:31–34. There, in v. 33, God pledges: “because this will be the covenant which I will conclude with the house of Israel after those days – uttering of YHWH – I will give ‫ תורתי‬in their midst, and on their heart I will write it.” This stands in contrast to the passages quoted above in (a), as it is a positive assurance granted by God and touching the core of the believers. The combination “to give (with ‫ )נתן‬. . . ‫ ”תורה‬appears in Jer 9:12 and 26:4. There it is set “before them/you.” Here it is no longer external to the people but gets right into their interior being and is inscribed on their hearts. The problem of the communityʼs not listening to God27 is thus overcome by an inward conJeremia befragt,” in Jeremia und die “deuteronomistische Bewegung” (ed. W. Groß; BBB 98; Weinheim: Beltz, 1995), 141–179, here 174–175. 25  NIV clarifies the expression by rendering it as “stipulations.” 26  Starting with 9:12, the only exception being 44:10. 27  See above, in subsection a, the fourfold repetition of “not listen” in Jer 9:12; 26:4; 32:23; 44:23, all of which mention (indirectly, in the negative mode) listening as a precondition for following Godʼs ‫תורה‬.

898

Georg Fischer

cordance with his will and intentions. The inscription of his ‫ תורה‬on the heart “overwrites” the sin engravings of Jer 17:1, and it also signifies that God goes beyond the writing of his commandments on scrolls or stones on Mount Sinai (Exod 24:4; 31:18). 1.4 Results The overview of the eleven passages with ‫ תורה‬in Jer reveals some interesting features: a. Nine times it explicitly refers to God. The only exceptions are Jer 2:8 and 18:18; there priests and another religious group are associated with it. b. There are eight different constructions, which points to a varied and creative use throughout the book. c. Most of them are negative, with verbs like despise/reject, abandon, (not) keep, (not) walk. This might indicate that there is a problem in the relationship of the people to Godʼs ‫ תורה‬and that Jer 2:8 has also to be understood in an adverse way. d. The first two negative verbs, “despise” and “abandon,” are also used for YHWH. This underlines the close connection of ‫ תורה‬and Israelʼs God. The peopleʼs behavior toward both is, to a large extent, nearly identical. e. Repeated combinations are “to walk in the ‫( ”תורה‬5×)28 and ‫( נתן תורה‬3×). The latter shows a progression toward the divine promise in Jer 31:33, bringing a positive solution to the theme of Godʼs ‫ תורה‬in the book of Jeremiah. It is to be regarded as the high point of Jerʼs message about it. f. The exact denotation of ‫ תורה‬in Jer cannot be determined with certainty. It may refer to divine instruction in general or to (one of) the first five scrolls of the Bible. However, the difference between those two options might not be that great, as Godʼs instructions are mainly found in these books.29 An argument for a written ‫ תורה‬might be taken from Jer 8:8. There the response to the peopleʼs affirmation mentions the “pen of scribes” as false and falsifying, maybe alluding to an already existing (“has made,” in the perfect tense) written document30 and to a group producing written documents. Eventually, the use of the verb ‫תפש‬, referring to something that can be “handled” like an instrument or  Jer 9:12 is included here, referring to ‫ תורה‬by preposition and suffix.  The Septuagint of Jer shows a tendency toward interpreting ‫ תורה‬as “law.” Its standard equivalent in the first half of the book is νόμος in the singular, whereas the second half always has plural forms and displays a variety of terms, employing νόμιμα in 33:4, νόμοι in 38:33, and προστάγματα in the three other instances (39:23; 51:10, 23). For the peculiarities of JerLXX, see G. A. Walser , Jeremiah: A Commentary based on Ieremias in Codex Vaticanus (Septuagint; Leiden: Brill, 2012). 30  Similarly Maier, Jeremia als Lehrer (see n. 4), 304: “dass es sich [. . .] um eine schriftliche Größe handelt.” 28 29

The Relationship of the Book of Jeremiah to the Torah

899

a weapon, may also point to a written form of the ‫תורה‬. Finally, the use of ‫תורה‬ with the definite article in Jer 2:8 – the only time within the whole book – might still add some weight to the consideration that “the Torah” is being referred to. Whether this is a realistic option will be investigated in part 2. 1.5 The Motif “To Know (God)” in the Book of Jeremiah As God in Jer 2:8 accuses the group “grasping/handling ‫ ”התורה‬of not knowing him, this introduces a second motif connected with what we have dealt with so far, and it might also offer some clues to understanding what ‫ תורה‬in Jer could mean. ‫ ידע‬occurs seventy-five times in Jer. Use of this motif is varied, starting in Jer 1:5 with God knowing Jeremiah. After Jer 2:8, the same reproach of not knowing God is found again in 9:2 and is intensified further in 9:5 by “they have refused to know me.” God encounters a deliberate rejection by his people. In contrast to this, there are passages in Jer that highlight the possibility and the importance of knowing God. In 9:23, God himself proposes that whoever wants to boast may do it by virtue of knowing him. Jeremiah 24:7 offers a divine promise to those exiled in 597 BCE: “And I will give them a heart to know me.” Finally, God announces in 31:34, “they all will know me, from their small (ones) to their great (ones).” The pervading knowledge of God in the whole community presents a kind of healing for the absence of knowing and its refusal earlier in 9:5. The scroll of consolation and therein especially the small unit with the new covenant in Jer 31:31–34 thus furnishes two important reversals for faults within the people: i. The inscription of divine ‫ תורה‬on the hearts of the people counters its disregard in the past (v. 33). ii. The promise of the knowledge of God,31 granted by him personally and to all (v. 34), overcomes the previous widespread refusal to know him. These two divine assurances belong to the same explanation of what is new in this covenant, starting in v. 33 and continuing in v. 34, in opposition to the old one (v. 32). ‫ תורה‬and “to know God” are connected here as well and (ii) can be regarded either as a kind of intensifying of (i) or as the highest fulfilment of it. Jeremiah 2:8, which brought together the two themes for the first time, shows their deficiencies and failures at the beginning of the book, thus providing a dark foil for what is to follow. Jeremiah 31:33–34 is the only other passage in Jer that again contains both topics, and it constitutes their focal point;32 in it, their negative dynamic is turned around and receives a good and lasting solution. 31  To know YHWH, the real God, contrasts with other passages where Israelʼs worship of “other Gods whom they/you did not know” (Jer 7:9; 19:4; 44:3) is mentioned. 32  For the knowledge of YHWH, Jer 31:34 is also the final point, although 44:3 also touches on this theme from the negative side (see n. 31). ‫תורה‬, however, occurs more often still later

900

Georg Fischer

The motif of “knowing God” in Jer shows that the mishandling of the ‫תורה‬, seen above in 1:1–4, is only part of a much broader defect, namely, the absence or even rejection of an intense relationship with God himself. To “know” in Hebrew can include an intimate relationship; thus the accusations in Jer 2:8 and the other passages with ‫ תורה‬in Jer reveal a fault deeper than mishandling of the ‫תורה‬: the lack of knowledge of God is at the root of the disastrous developments described in Jer. There is hope for the future, however, as the texts in Jer 24 and 31 show, a future which will even surpass what went before.

2 The Relationship of the Book of Jeremiah to the Torah The investigation of the use of ‫ תורה‬in the book of Jeremiah resulted, at least, in the possibility that ‫ תורה‬in Jer may already refer to (one of) the first five books of the Bible. This leads us further, to check whether other observations, on a literary level, may confirm that a connection between Jer and the Torah is not only possible but indeed real. I contend that the whole Torah (Genesis–Deuteronomy, in a form very close to the Masoretic Text) predates Jer and serves as a source for it. In order to show this, I will proceed in several steps. First, I will present several expressions that link the books of the Torah with Jer (2.1). An evaluation of these connections will follow (2.2), before the differences between the two literary corpora are dealt with (2.3). Finally, I will reflect on the significance of these observations (2.4). 2.1 Some Connections between the Torah and the Book of Jeremiah (a) With the Book of Genesis “Nations will bless themselves in you”33 in Jer 4:2 has its closest parallels in Gen 22:18 and 26:4. In Gen 22, the angel of the Lord promises this to Abraham and his descendants.34 In Jer, this motif is applied to the people who return to God, thus continuing and expanding what had been announced. Jeremiah 4:23–25

(see 1.3). The occurrences in Jer 44:10, 23 show that Godʼs promises, at least within the time span of Jeremiah, have not yet been realized. 33  The most detailed investigation of this expression is A. Flury-Schölch, Abrahams Segen und die Völker: Synchrone und diachrone Untersuchungen zu Gen 12,1–3 unter besonderer Berücksichtigung der intertextuellen Beziehungen zu Gen 18; 22; 26; 28; Sir 44; Jer 4 und Ps 72 (Forschung zur Bibel 115; Würzburg: Echter, 2007), 192. He arrives at the conclusion that ‫ ברך‬in the hithpael stem has a reflexive meaning (116) and sees this as valid for Jer 4:2 as well. 34  Gen 22:18 is, besides Ps 72:17, “der motivgeschichtliche Ausgangspunkt” for the other texts with this motif of blessing in Abrahamʼs descendants (F lury-S chölch , Segen [see n. 33], 239). This would indicate that Jer 4:2, too, is dependent on it.

The Relationship of the Book of Jeremiah to the Torah

901

describes a kind of reversal of the creation. Some words link it exclusively with the narratives in Gen 1–2.35 In Jer 5:1, God offers to spare the whole city if he finds one person “doing justice.” In Gen 18:32, God agrees to spare Sodom and Gomorrah, if ten just people can be found therein. Compared to that, the “price” for the rescue of the community in Jer 5 is tapered from ten to one. To “do justice” and the expression “way of YHWH” (Jer 5:4) further connect with Godʼs and Abrahamʼs dialogue in Gen 18:25, 19. Finally, Rachelʼs refusal to be consoled in Jer 31:15 comes nearest to Jacobʼs behavior in Gen 37:35, where his sons try to console him over the “death” of Joseph.36 (b) With the Book of Exodus Jeremiah 7:9 mentions various commandments of the Decalogue. It varies the order but nevertheless probably refers to Exod 20.37 In the same chapter (Jer 7), “leading out of Egypt” in v. 22 is connected in v. 23 with “listen to my (i.e., Godʼs) voice.” This has its source in Exod 19:5, Godʼs offer of the covenant at Mount Sinai. Several motifs link Jer 16 to the book of Exodus. Verse 14 alludes to the past exodus out of Egypt, and v. 15 surpasses it, referring to a future exodus out of all the countries of exile. In Jer 16:5, God has taken away ‫החסד‬ (“solidarity, grace”) and ‫“( הרחמים‬mercy”), in contrast to his self-revelation in Exod 34:6. The writing of Godʼs ‫ תורה‬on tablets of stone and his handing them over to Moses, with ‫נתן‬, in Exod 24:12 forms an exclusive relationship with Jer 31:33, the passage dealt with above in 1.3, where the ‫ תורה‬is inscribed on the hearts of the people. This overcomes the impasse of a merely superficial adherence to Godʼs instructions. Similarly, the next verse (Jer 31:34) seems to draw on the Sinai pericope, particularly Exod 34:9. The expression ‫סלח לעון‬, “to forgive the iniquity,” in connection with ‫חטאת‬, “sin,” is encountered only in these two passages of the Hebrew Bible.

35  B. Trimpe, “Ich schaue auf die Erde und siehe, sie ist tohu-wabohu (Jer 4,23),” BK 53 (1998), 135–139, esp. the table on 137, and still more extended, idem, Von der Schöpfung zur Zerstreuung: Intertextuelle Interpretationen der biblischen Urgeschichte (Gen 1–11) (Osnabrück: Rasch, 2000), 55–84. Another reversal of a motif of Gen 1 can be seen in Jer 16:2: the command not to take a wife and not to have children runs counter to Godʼs initial advice in Gen 1:28 to be fruitful and multiply. 36  This has been noted by S. E. Brown-Gutoff, “The Voice of Rachel in Jeremiah 31: A Calling to ‘Something New,’ ” USQR 45 (1991), 177–190, and K. Schmid, Buchgestalten des Jeremiabuches (WMANT 72; Neukirchen-Vluyn: Neukirchener Verlag, 1996), 133–136. Both have also pointed to further relationships between the Joseph story in Genesis and Jer. 37  There is only a slight indication that Exod 20, and not Deut 5, might be in view: the use of ‫ שקר‬occurs only in the Exodus version of the Decalogue.

902

Georg Fischer

(c) With the Book of Leviticus Chapter 26 in particular, with Godʼs promises and threats, has many connections with Jer. God is portrayed as ‫נתן גשם‬, “giving rain shower(s),” in Lev 26:4 // Jer 5:24.38 In Lev 26:23, he warns of consequences if the people donʼt let themselves be disciplined;39 in Jer 6:8 he asks them to accept discipline. The so-called covenant formulas in Lev 26:12 and Jer 7:23 are completely identical. The threat that the “country will become a ruin/desert (‫ ”)חרבה‬occurs in Lev 26:33 and eight times in Jer, from 7:34 onwards.40 The phrase ‫געלה נפש‬, “the soul abhors,” forms an exclusive link between Jer 14:19 and Lev 26, where it is used four times (vv. 11, 15, 30, 43); v. 30 comes closest, as God is also the subject of it there. (d) With the Book of Numbers The most intensive connections between Jer and Numbers can be found in the oracles on Moab in Jer 48. Many names of places in vv. 21–24 correspond with locations mentioned in Num 21 as well as Num 32–33 in connection with the desert wanderings of the Israelites; some of them, like Diblathaim, form exclusive links between the two books.41 Moreover, Jer 48:45–46 concurs to a large extent with Num 21:28–29 and 24:17 and seems to have been taken from there.42 (e) With the Book of Deuteronomy Deuteronomy is the main source for Jer, as has been widely recognized.43 Recent research has brought to the fore further aspects. Although the whole book of Deuteronomy has links with Jer, its frame, namely, Deut 1–11 and 26–34, offers the bulk of the relationships. Especially prominent are Deut 4–5; 7; 10; 28;

38  The only other instance for this phrase, Ps 105:32, uses it with another meaning (turning rain showers to hail). 39  ‫ יסר‬in the niphal is used for “disciplined.” 40  For a list of the passages where this expression occurs, and also of other recurring phrases in Jer, see H.-J. Stipp, Deutero-jeremianische Konkordanz (Arbeiten zu Text und Sprache im Alten Testament 63; St. Ottilien: EOS, 1998), 50–51. 41  It occurs in Jer 48:22 and together with two other place names elsewhere only in the topographical list of Num 33:46–47; see G. F ischer , Jeremia 26–52 (HThKAT; Freiburg: Herder, 2005), 514. 42  Fischer, Jeremia 26–52 (see n. 41), 525. 43  See, among others, the two volumes on the Deuteronomistic redaction of the book of Jeremiah by W. Thiel, Die deuteronomistische Redaktion von Jer 1–25 (WMANT 41; Neukirchen-Vluyn: Neukirchener Verlag, 1973); idem, Die deuteronomistische Redaktion von Jer 26–45 (WMANT 52; Neukirchen-Vluyn: Neukirchener Verlag, 1981); and W. L. Holladay, Jeremiah 2 (Chapters 26–52) (Hermeneia; Philadelphia: Fortress, 1989).

The Relationship of the Book of Jeremiah to the Torah

903

and 31–32.44 Among them, Deut 28 forms the peak, providing by far the most connections.45 I mention briefly some exclusive links, picking out only the correspondences of the texts: • Deut 4:29 // Jer 29:13: “and you will seek [. . .] and you will find, because you ask for him/me with all your heart” • Deut 5:29 // Jer 32:39: “a heart [. . .] to fear me [. . .] all days, for their benefit and that of their children” • Deut 5:33 // Jer 7:23: “and you shall go the whole way which YHWH has/I have commanded you [. . .], that it may be well with you” • Deut 28:48 // Jer 28:14: “an iron yoke on the neck” • Deut 28:49 // Jer 5:15: “a nation from afar above you whose tongue (language) you do not hear/know” • Deut 32:22 // Jer 15:14: “because a fire is kindled in my anger” The length of some of these links, their number, the precision of the equivalences and their concentration in some areas (like Deut 5:29–33 or Deut 28) point to a deliberate literary technique, in my view on the side of Jer, of picking up those very special expressions of the last book of the Torah and reusing them throughout Jer. 2.2 Evaluation of the Connections Overall, the links between the books of the Torah and Jer are too many and too strong to be accidental. They span the whole range, from the first chapter of Genesis to the final frame of Deuteronomy; some of these texts are generally deemed to be late.46 No book of the Torah is missing. The highest density of relationships and the closest parallels exist between Deuteronomy and Jer and suggest that Jer had great familiarity with Deuteronomy. The connections reveal a desire to choose important texts such as Gen 1, the first creation narrative; Gen 18 and 22, two key passages for Abraham; and Exod 19 and 34, cornerstones for Godʼs revelation at Mount Sinai. 44  For an overview of the major links between Deuteronomy and Jer, see G. Fischer, “Der Einfluss des Deuteronomiums auf das Jeremiabuch,” in Deuteronomium – Tora für eine neue Generation (ed. G. Fischer et al.; BZABR 17; Wiesbaden: Harrassowitz 2011), 247–269; for the relationship of the final chapters of Deuteronomy to Jer, see idem, “Das Ende von Deu­te­ ronomium (Dtn 26–34) im Spiegel des Jeremiabuches,” in “Gerechtigkeit und Recht zu üben” (Gen 18:19): Studien zur altorientalischen und biblischen Rechtsgeschichte, zur Religions­ geschichte Israels und zur Religionssoziologie; Festschrift für Eckart Otto zum 65. Geburtstag (ed. R. Achenbach and M. Arneth; BZABR 13; Wiesbaden: Harrassowitz, 2010), 281–292. 45  Fischer, “Fulfillment” (see n. 22). 46  Examples include Gen 1, normally attributed to the Priestly writer; Lev 26, with its outlook on the reversal of the exile at the end (vv. 40–45); and Deut 32, which covers Israelʼs history until the time of its postexilic restoration.

904

Georg Fischer

Comparison of the related texts shows some typical features: i. Similar blessings connect Gen 22 and Jer 4:2; Jer, however, extends it from Abraham and his descendants to all Israel, realizing the original promise. ii. The anticreation in Jer 4:23–25 signifies a reversal of Gen 1; normally such a turning back, as in Jer 4, presupposes an originally positive situation, chronologically and on a literary level. iii. Godʼs offer in Jer 5 to spare the whole city because of one just person looks like an intensification of Godʼs gracious dealings with Abraham in Gen 18 taken to the extreme. iv. Jeremiah 5:15 seems to clarify the strange expression “not to hear the tongue/language” in Deut 28:49 by changing the verb to “know.” v. In Deuteronomy, Moses is often the speaker, referring to God in the third person.47 The parallel passage in Jer 7:23 has God speaking in the first person; thus he confirms personally the announcements made by his prophet and gives added weight to them. All these traits can be observed more often. These and other indications (see 2.3 below) allow us to surmise that the direction of dependence runs from the Torah to Jer. The author48 of Jer was obviously very closely acquainted with the whole Torah.49 Our conference in Jerusalem was entitled “The Pentateuch within Biblical Literature: Formation and Interaction.” Regarding the aspect of the formation of the Torah, Jer has nothing to contribute. It is very different with the aspect of interaction. The first five books of the Torah are the source, and Jer uses them as basis for its expressions, ideas, and proclamations, but not without developing them further. The book of Jeremiah thus presents itself as a kind of prolonged reflection on and elaboration of the Torah. The next point will emphasize that even more. 2.3 The Book of Jeremiahʼs Differences with Respect to the Torah So far we have mainly concentrated on what is common to both the Torah and Jer. Now we have to go one step further and look at the differences. Some are obvious, like the presence or absence of divine law or the different genre of the book of Jeremiah. As a prophetic book, it is distinguished from the books of the Torah, which contain mainly narration and laws. By contrast, there are no laws to be found in Jer. The book of Jeremiah does, however, have some narratives  See above, 2.1, the example of Deut 5:33, with YHWH changed to “I” in Jer 7:23.  In my view, Jer is the work of one author, despite its composite nature, variety of language, and complex arrangement; for these reasons, see G. Fischer, Jeremia: Der Stand der theologischen Diskussion (Darmstadt: Wissenschaftliche Buchgesellschaft, 2007), 91–114. 49  The connections even with the last chapters of Deuteronomy suggest this; therefore, as a logical conclusion, Jer must be late, i.e., not earlier than 400 BCE. 47

48

The Relationship of the Book of Jeremiah to the Torah

905

about the prophet.50 On the other hand, the characteristic feature of prophetic books, to present divine speech and to address the people, is encountered not only in Jer but also in Deuteronomy. Furthermore, several texts in Deuteronomy portray Moses as prophet.51 We can observe three types of differences. The first type consists of aspects of the Torah not present in Jer. The second is marked by alterations made in Jer. The last type shows new themes within Jer. (a) Not Present in the Book of Jeremiah Whereas some commandments of the Torah, like the observance of the Sabbath, receive great attention (Jer 17:19–27), others are not even mentioned or play only a very minor role. This, however, must not be over interpreted, as there may be other causes for the absence of certain themes in Jer. The entire legislation for cultic offerings is completely absent in Jer. Jeremiah 7:22–23 relativizes the importance of holocausts and sacrifices. Whereas the Shema asks for visible signs like tefillin and mezuzot,52 Jer never speaks of them. The major feasts – Pesach, Sukkot, and the Feast of Booths – are not mentioned, nor are the pilgrimages prescribed on these occasions.53 Similarly, Yom Kippur does not occur in Jer. These and other observations in the same direction point toward a different orientation of Jer. Some religious practices do not seem to be considered relevant. (b) Differences and Changes in the Book of Jeremiah Several times Jer counters ideas and attitudes relevant for the Torah. This is a step beyond the previous point. It shows that Jer differs in some aspects from the Torah and even distances itself from it. The beginning of Jer 3 quotes the case presented in Deut 24:1–4: a woman sent away by her husband who has become another manʼs wife. According to that law, the former husband must not take her back later. In Jer 3:12, God himself invites unfaithful Israel to return to him, thus transgressing his proscription in Deut 24. Jeremiah 3 holds yet another change with regard to the Torah. Moshe Weinfeld has pointed to several instances of a “spiritual metamorphosis” in Jer that  Mainly in the second part, especially in Jer 26–44.  See especially Deut 18:15–18; 34:10–12; and D. M arkl , “Moses Prophetenrolle in Dtn 5; 18; 34: Strukturelle Wendepunkte von rechtshermeneutischem Gewicht,” in Deuteronomium – Tora für eine neue Generation (ed. G. Fischer et al.; BZABR 17; Wiesbaden: Harrassowitz 2011), 51–68; J.-P. Sonnet, “La construction narrative de la figure de Moïse comme prophète dans le Deutéronome,” RTP 142 (2010), 1–20. 52  Mentioned in Deut 6:4–9, esp. vv. 8–9; for the tefillin, see also Exod 13:9. 53  The pilgrimage of eighty men from Shiloh and Samaria to the “house of YHWH” in Jer 41:5, however, might allude to the “seventh month” (Jer 41:1) of the Feast of Booths. 50

51

906

Georg Fischer

involves important changes in the religious orientation of Israel.54 The first one of these passages is Jer 3:16–17: whereas the ark of the covenant is the primary object among those mentioned for the furnishing of the tent of meeting in Exod 25:10–22, Jer 3:16 stresses with five verbs that the ark will no longer have any relevance at all. The following verse (17) even replaces it with the new role attributed to Jerusalem. There are several similar cases for such religious changes. Jeremiah 7:22–23 lays all emphasis on “listening” to God, downplaying the significance of sacrifice. The “new exodus” from the countries of the exile in Jer 16:14–15, with a parallel in 23:7–8, will outweigh the old one out of Egypt. The new covenant in Jer 31:31–34 will surpass the old, broken one from Mount Sinai.55 God declares in Jer 21:5 that he will personally fight against the people in Jerusalem “with an outstretched hand and with a strong arm.” This is a reversal of Godʼs engagement for Israel, as depicted in Deut 4:34,56 underlined by the unique permutation of the adjectives. Unlike in Exod 14, God will take the side of Israelʼs foes in the battle. A veiled revocation of a law of the Torah is present in Jer 30:18, 21. The law about the apostasy of a whole city states that “evil men go out from its midst” and that it shall become a ‫תל‬, a “heap of ruins, tel” and “never be built up again.”57 Deuteronomy 13 aims at Jerusalemʼs fate and tries to explain what it had deserved. Jeremiah 30, picking up both expressions and creating a unique relationship between them, describes how God overturns this merited punishment, fails to abide by his own law, and acts graciously once more.58 A decisive difference, even a sharp opposition, lies in the contrasting movements of the Torah and Jer. Whereas the books of Exodus through Deuteronomy describe Israelʼs wandering toward the promised land, Jer reverses the direction and shows in Jer 42–44 how the people and their leaders choose to go back 54  M. Weinfeld, “Jeremiah and the Spiritual Metamorphosis of Israel,” ZAW 88 (1976), 17–56. 55  Another instance might be Jer 31:29–30, which contains the idea of individual retribution, in contrast to the idea that subsequent generations have to pay for the guilt of their forefathers. The idea of individual retribution is an opinion that could – falsely – arise on the basis of Exod 34:7, if one did not take into account “(in case) they hate me” in Exod 20:5, which places a condition on it. 56  G. Fischer, Jeremia 1–25 (HThKAT; Freiburg: Herder, 2005), 636; similarly, D. RomS hiloni , “Facing Destruction and Exile: Inner-Biblical Exegesis in Jeremiah and Ezekiel,” ZAW 117 (2005), 189–205, esp. 193–194. 57  These two expressions, referring to a city, occur for the first time in Deut 13:14, 17 and elsewhere only in Jer 30. 58  G. Fischer, Das Trostbüchlein: Text, Komposition und Theologie von Jer 30–31 (SBB 26; Stuttgart: Katholisches Bibelwerk, 1993), 191–192 and 207–208; E. Otto, Gottes Recht als Menschenrecht: Rechts- und literaturhistorische Studien zum Deuteronomium (BZABR 2; Wiesbaden: Harrassowitz, 2002), 76–78, has taken up this interpretation and defined it as “abrogation” of the Deuteronomic law in Jer.

The Relationship of the Book of Jeremiah to the Torah

907

down to Egypt,59 the land of their former slavery, despite Godʼs warnings through Jeremiah. These are some indications of Jerʼs deviation from prescriptions and concepts of the Torah. Whereas the construction of the sanctuary and the instructions for its functioning are a key element of the Torah (Exod 25–31; 35–40; etc.), in Jer the temple is destroyed and emptied.60 Whereas the Torah never mentions Jerusalem, in Jer it is besieged, captured, and burnt, yet for the future receives a new role as a center for the worship of YHWH. There are still some other, more general differences in accentuation, for example, with regard to Mosesʼs mediation and to prayer for others.61 Overall, Jer advocates at least in some points for an essentially different position from that of the Torah. (c) New Aspects In contrast to the books of Isaiah and Ezekiel, Jer employs ‫חדש‬, “new,” only twice:62 in Jer 31:22 and 31, both times within the scroll of consolation. In the first passage, God announces that he will “create (something) new in the country,” whereas in the latter he promises a “new covenant.” The small number of occurrences are countered by the importance of the contents. Jeremiah 31:22 envisions a changed relationship between the genders and an increased role for women as a basis for a more just and fruitful society. Jeremiah 31:31 is the beginning of the unit presenting a fundamental change for Israelʼs faith; three aspects of it have already been dealt with above.63 In comparison to the other books of the Hebrew Bible, Jer offers a deeper analysis of the causes for the disaster. The Torah, being fictionally situated in a time before the peopleʼs entering the land, could only sporadically refer to infidelity, idolatry, and other reasons for Israelʼs and Judahʼs downfall. The Former and the Latter Prophets adopt another stance, dealing more extensively with the communityʼs guilt throughout its history. The book of Jeremiah, however, goes 59  For the important role of Egypt in Jer, see M. P. M aier , Ägypten – Israels Herkunft und Geschick: Studie über einen theo-politischen Zentralbegriff im hebräischen Jeremiabuch (ÖBS 21; Frankfurt: Lang, 2002). 60  Jer 52:13, 17–27; for this motif and its significance in Jer, see G. Fischer, “Zur Relativierung des Tempels im Jeremiabuch,” in LʼEcrit et lʼEsprit (ed. D. Böhler et al.; OBO 214; Fribourg: Universitätsverlag, 2005), 87–99. 61  Knowledge of God in each individual and the group collectively (see Jer 31:34) makes teaching and religious exhortation superfluous; there is no more need for intermediaries. Additionally, the only mention of Moses in Jer appears in connection with his inability to obtain Godʼs forgiveness (Jer 15:1). Whereas Deut 23:6 taught not to seek the shalom of Ammonites and Moabites, Jer 29:7 exhorts the exiled Judeans to pray for the shalom of the city whose inhabitants took them captive. 62  Isaiah has sixteen occurrences of the Hebrew root for “new,” and Ezekiel has thirty-two. 63  These are vv. 31–34 as the last example for “spiritual metamorphosis,” the inscription of the Torah on the heart in v. 33, and the widespread knowledge of God in v. 34.

908

Georg Fischer

still further by concentrating almost relentlessly on the catastrophe of 587 BCE, penetrating into the roots of the destruction of Jerusalem at length, exceeding all other treatments of it by far. In a similar way, Jer goes beyond the Torah in describing the new salvation granted by God after 587. In this regard, Jer is much more restrained than Isaiah and Ezekiel, for instance. It can say more than is presented in the Torah, but it remains sober and realistic in its hope. The Torah announced Godʼs dealings with the people for the future, when they would enter the promised land. It could not show the realization of these assurances or of the threats. This, in turn, is one perspective of the prophetic books, starting with Joshua. The book of Jeremiah occupies a special place therein, insofar as it picks up the curses of Deut 28 more than any other scroll of the Bible. The book of Jeremiah demonstrates that those announcements have been fulfilled; thus the way is open for conversion and return to God, as foreseen in Deut 30 and other texts.64 2.4 Significance The foregoing observations have revealed an interesting, even contrasting picture. On the one hand, it is obvious that Jer uses the Torah to a great extent, Deuteronomy being its main source for allusions. On the other hand, Jer goes beyond the Torah. The book of Jeremiah thus uses the Torah as its foundation yet has also the freedom to put forward new ideas and changes in matters of faith and orientation, even to the extent of revoking laws.65 To be sure, this attitude does not touch key issues, like YHWH as the God of Israel, the general concept of covenant, worship, etc. There is much common ground between the Torah and Jer, and it is only on this basis that Jer differs from it. If we ask for a possible background for this contrasting picture, another aspect typical of Jer might give a clue: no other book of the Hebrew Bible describes so extensively its authorʼs conflicts with other priests and prophets.66 Both groups 64  The last three paragraphs contained more general remarks on the differences between Torah and the following books of the Bible, including Jer. The book of Jeremiah stands out among them only in some specific aspects. 65  The distance Jer displays with regard to Deuteronomy and some Dtr ideas asks for great caution in applying the label DtrJ, as is often done – besides, it is far from clear what is meant by “DtrJ.” Although one has to acknowledge Deuteronomic and Deuteronomistic elements in Jer, it employs many more forms of language; it has a different scope and covers a much larger field of interests than Deuteronomy and the related books. Therefore, it can hardly be adequately understood by the term DtrJ or by proposing a Dtr redaction for it. Jer is really beyond such categories, as the subtitle of our session, “Beyond DtrJ,” indicated. 66  The main texts are Jer 4:10; 5:13; 6:13–14 // 8:10–11; 14:13–18; 23:9–40; 26–29; for the latter, see the recent study by A. C. Osuji, Where Is the Truth? Narrative Exegesis and the Question of True and False Prophecy in Jer 26–29 (MT) (BETL 214; Leuven: Peeters, 2010).

The Relationship of the Book of Jeremiah to the Torah

909

were involved in the interpretation of the Torah. This would fit well with the initial statement on those handling the Torah, in Jer 2:8. Jeremiah 18:18 also mentions these two groups and presupposes them to be working in opposition to Jeremiah. In addition, Jer 8:8 introduces a group of scribes connected with ‫תורה‬ and accuses them of deceit. It may well be that disputes over the understanding, relevance, and application of the Torah were a key factor for the formation of Jer.

3 Results 3.1 Jeremiah 2:8 as a Key The first instance of ‫ תורה‬in Jer connects it with a group of responsible persons in Judah, alongside priests, shepherds/leaders, and prophets. This group is described negatively, as are the others, and accused of “not knowing” God, although they are “handling/grasping ‫התורה‬.” Both these themes pervade Jer. ‫ תורה‬is neglected, despised, and abandoned and thus receives a similar fate to its originator, God. The “knowledge” of YHWH is largely absent, and the people even refuse it (Jer 9:5). The dynamic of Jer, however, foresees also a healing, presented in the announcement of a new covenant in Jer 31:31–34, with deliberate reversals of the relationship with divine ‫ תורה‬and of the knowledge of God in vv. 33–34. These problems with the ‫ תורה‬and with the group seemingly responsible for dealing with it indicate a struggle in the background of Jer, most probably connected with the conflicts of the prophet Jeremiah with other priests and prophets. They represent two parties, characterized by different attitudes; the latter stand for a more traditional position, hold authority, have high esteem for the temple and its cult, and are convinced that God is always on their side. The figure of the prophet Jeremiah, on the other hand, represents another stance,67 open to Godʼs new, different dealings with his people, seeing even Nebuchadnezzar as Godʼs servant, and promoting a “spiritual metamorphosis” enabling a deeper understanding of what happened in the destruction of Jerusalem in 587 BCE. 3.2 The Book of Jeremiah Is Based on the Torah but Differs from It Part 2 demonstrated the undeniable links between Jer and all the books of the Torah. These links suggest that Jer depends on the books of the Torah, in their 67  The main title of the session in Jerusalem was “Torah and Prophets I.” In my view, Jer occupies a very late place in the process of forming the prophetic books, after all the Former Prophets, after more than half of the Book of the Twelve (namely Hosea, Amos, Obadiah, Micah, Nahum, Habakkuk and Zephaniah), and even after Isaiah (in its entirety) and Ezekiel. Jer thus belongs to “Prophets III” (or even IV).

910

Georg Fischer

integrity, as even late passages seem to be presupposed in Jer. The extent of their usage varies and increases toward the last book, Deuteronomy. In light of these manifold and sometimes strong relationships, an interpretation of ‫ התורה‬in Jer 2:8 as “the Torah” – namely, the first five books of the Hebrew Bible – receives additional weight and cannot be excluded. In fact, it seems to offer the most likely and reasonable understanding of the first and probably also of most other occurrences of ‫ תורה‬in Jer. Though Jer picks up expressions, ideas, and motifs from the Torah, it does not completely concur with it. In some instances, it shows other aspects, or it even runs counter to texts of the Torah – e.g., regarding the ark of the covenant68 or the law of the apostasy of a whole city.69 This raises the question of whose authority Jer dares use to proclaim such reversals of important Torah concepts.70 Besides these contrasts with the Torah, Jer also goes a stage further than the Torah. It deals extensively with the downfall of Judah and Jerusalem and depicts this catastrophe against the background of Deuteronomic curses. As it discusses the realization of former announcements, Jer must be later. Furthermore, the undoing of creation in Jer 4:23–26 presupposes Godʼs creative act described in Gen 1. All these aspects point toward a more extensive background and a broader horizon present in Jer. 3.3 The Book of Jeremiah Offers a Glance into (Late) Persian Period Judah and Second Temple Judaism The critique contained in Jer 2:8 and the contrasting positions with regard to the Torah that have come to the fore through this analysis of Jer point to disputes during the time of its production, probably in the fourth century BCE, shortly before the Hellenistic era.71 Several rival opinions struggle to be heard as authoritative voices of God, and Jer engages with some influential groups who are falsely convinced that they are already and always on the right track. The book  Contrasting texts are Exod 25 and Jer 3:16–17.  Jer 30:18, 21 stand in opposition to Deut 13. 70  Jeremiahʼs legitimization is stressed several times throughout the book: at his commissioning in Jer 1:9, as God touches his mouth; in 5:14, still intensifying the force of the divine words in his mouth through the additional “fire”; and in 15:19, with Godʼs offer to Jeremiah to “be like his mouth.” All these elements are unique for the Hebrew Bible. Indirectly, the criticism of the false prophets in Jer 23:9–32 also heightens Jeremiahʼs authority, allowing us to suppose that the positive qualifications (like v. 18, or “wheat, fire, hammer” in vv. 28–29) are applicable to him and his proclamation of Godʼs words. 71  This dating for Jer is a consequence of its intertextual relationships: If the author of the book of Jeremiah could use the whole Torah, all the Former Prophets, and a large part of the Latter Prophets, it cannot be earlier than the fourth century BCE. At the same time, Hellenistic influence can hardly be detected in Jeremiah. For a more extended reasoning see F ischer , Jeremia 1–25 (see n. 56), 65–75. I am grateful to Mrs. Felicity Stephens for improving the English of this article. 68 69

The Relationship of the Book of Jeremiah to the Torah

911

of Jeremiah claims against them that the primary aim must be “to know God,” who “creates new things in the country” (Jer 31:22) and is able to go beyond what is traditional and usual.

* Let me conclude with Jer 2:8: “Those handling/grasping ‫ התורה‬do not know me.” This is an early comment on the custodians, interpreters, and exegetes of the Torah, dating back more than two thousand years. I wonder what the author of Jer would say to Pentateuch research today. In any case, it raises the question of the relationship between studying divine Scripture and being familiar with God.

Compositional Harmonization Priestly and Deuteronomic References in the Book of Jeremiah – An Earlier Stage of a Recognized Interpretive Technique Dalit Rom-Shiloni I. Introduction1 This paper addresses a topic that has achieved a consensus in Hebrew Bible scholarship on Persian period and Hellenistic literature, that is, that the harmonization or coordination of diverse literary or legal materials of non-Priestly, Priestly, and Deuteronomic origins is a feature of Second Temple scribal activity and that, within the Hebrew Bible, this feature is widely found in late historiography and in the redactional layers of the Pentateuch. The present study focuses on harmonization as a literary technique in the book of Jeremiah. As a student of prophetic literature generally and the book of Jeremiah in particular, I have found myself struggling with how to interpret the phenomenon of harmonization as a literary feature of this bookʼs compositional process. The discussion below shows that this feature may be found in poetic passages (i.e., passages which are widely accepted as “authentic” Jeremiah). Thus, several questions arise: What characterizes compositional harmonizations in the book of Jeremiah? What functions do they serve in the prophetic proclamation? Should the harmonization of pentateuchal traditions necessarily be understood as an indication of late dating? That is, does the use of such a technique necessarily depend on the prior existence of a canonical, written Torah? Must this phenomenon necessarily be associated with a late dating of the formation of the Pentateuch, especially its P/HL (Holiness Legislation) materials, and accordingly be an indicator of the prophetic book? Might scholars instead consider the implications of this literary phenomenon as a feature integral to earlier layers of the book of Jeremiah? Hence, this paper continues my teasing

1  This study was written with the help of the Israel Science Foundation (grant 148/09). A shorter version was read at the Sixteenth World Congress of Jewish Studies (Jerusalem, 2013), in the session “Convergence and Divergence in Pentateuchal Theory.” I am grateful to Dr. Ruth Clements for being my first and most challenging reader.

914

Dalit Rom-Shiloni

out of the relations between these two corpora, the pentateuchal literature and the book of Jeremiah.2 The paper begins with a few methodological comments, presents a short discussion of several examples from poetic passages in the book of Jeremiah, and offers some concluding observations on the implications of these findings for the study of Jeremiah, the Pentateuch, and their interrelationships.

II. Harmonization: The Collation of P(/HL) and D References in Late Biblical Literature and in Late Historiography Harmonization (also called coordination, combination, collation, conflation, or blend) is usually treated as a scribal phenomenon in two different contexts – either as a technique of literary composition performed by authors and redactors or as part of the transmission process performed by scribes, tradents, and translators.3 Within these two contexts, harmonization is seen as one of those tools that causes the boundaries between author, redactor, and scribe to break down, since in the process of smoothing out the text in harmonistic ways, even redactors, translators, and scribes become active authors to at least a minimal degree.4 2  See D. Rom-Shiloni, “How Can You Say, ‘I Am Not Defiled’? (Jer 2:20–25): Allusions to Priestly Legal Traditions in the Poetry of Jeremiah,” JBL 133/4 (2014), 757–775; idem , “ ‘On the Day I Freed Them from the Land of Egypt’: A Non-Deuteronomic Phrase within Jeremiahʼs Covenant Conception,” VT 65 (2015), 621–647. 3  For the identification of harmonization as an aspect of redaction criticism, see J. Barton, “Redaction Criticism,” ABD 5:645–647; for harmonization in relation to textual criticism, see E. Tov, “The Nature and Background of Harmonizations in Biblical Manuscripts,” JSOT 31 (1985), 3–29. At times, textual and redaction criticism are used together to explain conflations within MT; see idem , “The Composition of 1 Samuel 16–18 in the Light of the Septuagint Version,” in Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 97–130. 4  On the difficulty of attributing harmonization to an author, an editor, or a scribe, see Y. Z akovitch , Inner-Biblical and Extra-Biblical Midrash and the Relationship between Them (Tel Aviv: Am Oved, 2009), 145–157 (Hebrew). For scribes as possible exegetes or as influenced by interpretive trends in the literary compositions of their times (such as rewritten Torah or pseudepigraphy), see Tov, “Nature and Background of Harmonizations” (see n. 3), 15–19. The active role played by redactors (compilers, etc.) has been the focus of redaction criticism, to the point that the involvement of the prophets (as historical figures) in the writing of their prophecies was challenged altogether. See R. G. Kratz, “Redaktionsgeschichte in Altes Testament,” TRE 28 (1997) 367–378; and see Kratzʼs specific study on the prophetic literature, where he followed in the footsteps of O. H. S teck , “Die Redaktion der Prophetenbücher,” in Rezeption und Auslegung im Alten Testament und in seinem Umfeld: Ein Symposion aus Anlass des 60. Geburtstags von O. H. Steck (ed. R. G. Kratz and T. Krüger. Freiburg, Schweiz: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht, 1997), 9–27; and more recently, idem , The Prophets of Israel (trans. A. C. Hagedorn and N. MacDonald; Critical Studies in the Hebrew Bible 2; Winona Lake, IN: Eisenbrauns, 2015), esp. 27–35. Great attention has

Compositional Harmonization

915

The common denominator shared by these authors, redactors, scribes, translators, and commentators in creating coordinations or harmonizations is the fact that they are working with earlier traditions or written materials that are themselves in a gradual process of literary (and textual) growth.5 Harmonization thus sends a clear signal of diachronic development, made possible when later authors (redactors, scribes, etc.) have at their disposal two or more versions of an oral/written literary tradition or law, which they wish to combine (conflate and bring into agreement) in a new literary context.6 When the harmonized materials are pentateuchal in nature, scholars have often tended to identify the conflation of Deuteronomic (D) and Priestly (P/HL) pentateuchal sources/traditions as a redactional procedure, involving textual sources, that could have taken place only in a very late period.7 In his influential study “Preliminary Midrash in Chronicles” (1980), I. L. Seeligmann discussed among his examples the Chroniclerʼs account of Jehoshaphatʼs appointment of judges (2 Chr 19:4–11) and its reliance on the

been (and still is) given to the role of R in pentateuchal studies; see for instance H. Donner, Aufsätze zum Alten Testament aus vier Jahrzehnten (BZAW 224; Berlin: de Gruyter, 1994), 259–285. 5  Carr has described scribal harmonization and coordination as “the most commonly documented forms of scribal revision across the board”; see D. M. Carr, “Scribal Processes of Coordination/Harmonization and the Formation of the First Hexateuch(s),” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman, et al.; Tübingen: Mohr Siebeck, 2011), 63–83, esp. 65–69, here 69. 6  Carr dedicated a detailed discussion to this phenomenon in D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011), 29–31, 34. He pointed out examples of harmonization in Proverbs that reflect oral or written transmission processes on the part of Hebrew tradents and translators alike. These processes may thus be traced both in qere and ketib differences within MT (cf. Prov 20:16; 27:13) and in differences between the LXX, the Syriac, and the Aramaic Targums. But of even more interest are Carrʼs discussions of harmonization as a scribal practice of coordination in ancient Near Eastern literature (40–48, 66–70), as well as in postbiblical literature of the Second Temple period; he noted, for example, the Temple Scroll, along with Proto-Samaritan and Samaritan pentateuchal traditions, which may go back to harmonizing tendencies within parallel biblical traditions (70–71, 91–98). These literary harmonizations are part of a trend toward expansion, which Carr (71) recognized in both Hebrew and Samaritan cultures as an accepted practice of ancient authors and redactors. Thus, harmonization actually functions as a technique of literary composition. 7  I find Carrʼs perspectives on the process of literary evolution and redaction to be illuminating. Note his concluding sentence on the development of the Hexateuch as “a long-range process [. . .] by which these compositions were gradually harmonized/coordinated (even while separate) before being combined and further coordinated in the Persian and later periods” (Carr, Formation of the Hebrew Bible [see n. 6], 303, and see also 347–349). Hence, harmonization does not require a written canonical Torah (see below).

916

Dalit Rom-Shiloni

conflation of two distinct passages from Deuteronomy, 16:18–20 and 17:8–9.8 Seeligmann was sensitive to the presence of Priestly (P/HL) phrases interwoven into this context: e.g., ‫ והזהרתם אתם‬in 2 Chr 19:10 echoes Lev 15:31 (as well as a non-P text, Exod 18:20); ‫ ולא יאשמו ליהוה‬in 2 Chr 19:10 echoes Lev 5:1–5, 19, etc.; and ‫ והיה קצף עליכם ועל אחיכם‬in 2 Chr 19:10 echoes Num 1:53; 18:5.9 Seeligmann pointed out that the Chroniclerʼs treatment of pentateuchal traditions was guided by his basic conception that the Torah – a written corpus of legal teachings (1 Chr 16:40), handed down from Moses (2 Chr 25:4) – was the highest authority that the people ought to have been following as of Solomonʼs time. Thus, an important rhetorical aim of the Chronicler is to present full obedience to “the Teaching of Moses, man of God” as an ancient custom (2 Chr 30:16; see also 2 Chr 23:18; 25:4; 33:8; 34:14).10 Two practical implications follow: (1) The Chronicler has no hesitation about transforming his pentateuchal sources. Seeligmann pointed out that, in bringing passages and phrases together, the Chronicler often changes their content, as is typical of later atomistic mid-

8  I. L. S eeligmann , “Preliminary Midrash in Chronicles,” Tarbiz 49 (1980), 14–32 (Hebrew), esp. 21–23, translated into German as “Anfänge der Midraschexegese in der Chronik,” in Gesammelte Studien zur Hebräischen Bibel (ed. E. Blum; FAT 41; Tübingen: Mohr Siebeck, 2004), 31–54. 9  Two of these phrases are distinctively Priestly (P/HL) and do not occur at all in Deuteronomy: ‫ והזהרתם אתם‬and ‫ אשם ל׳‬in the sense of “incur liability to God for reparation.” On the first phrase, see Ezek 3:17–21; 33:1–11 and note also the translation traditions for 2 Chr 19:10: Sam: ‫ ;והזהרתם את בני ישראל‬Peshitta: ‫ ;ותזהרון לבני יסריל‬LXX: Καὶ εὐλαβεῖς ποιήσετε τοὺς υἱοὺς Ισραηλ (“you shall make the sons of Israel cautious [concerning their uncleanness]”). On the second phrase, see J. Milgrom, Leviticus 1–16 (AB 3; New York: Doubleday, 1991), 334. Less distinctive, though still more frequent in P/HL than in other types of composition, is the third phrase, ‫( והיה קצף עליכם ועל אחיכם‬but note other contexts where ‫ קצף על‬occurs with God as agent: Lev 10:6; Num 16:22; 17:11; Deut 1:34; 9:19; compare Deut 9:7, 8 and Josh 9:20; 22:18, 20). Two peculiarities distinguish ‫( מן הלוים והכהנים‬2 Chr 19:8) from ‫הכהנים‬ ‫( הלוים‬Deut 17:8): the articulation, in 2 Chr 19:8, of two distinct groups, ‫( הכהנים והלוים‬attested frequently in Ezra-Nehemiah and Chronicles); and the reversed order of the phrase, which is otherwise attested only in Jer 33:21 and Neh 10:1. ‫( ושטרים הלוים‬2 Chr 19:11) is a hapax and a remarkable conflation of ‫( שפטים ושטרים‬Deut 16:18) and ‫ובאת אל הכהנים הלוים ואל השפט‬ (Deut 17:9). See J. H. T igay , Deuteronomy (JPS Torah Commentary; Philadelphia: Jewish Publication Society, 1996), 163–165. 10  A. R ofé , “The Law about the Organization of Justice in Deuteronomy (16:18–20; 17:8–13),” Beit Mikra 65 (1976), 199–210, esp. 204–209 (Hebrew), described this general mode of historiographical writing as “nomistic history.” He depicted a genre that seems historical but is actually halakic in its main focus and that reflects the late era of its authors and their conceptions of the Torah. Rofé mentioned that such halakic stories occur mostly in Joshua and Chronicles. He also pointed out the differences between 2 Chr 19:4–11 and the Deuteronomic passages 16:18–20; 17:8–13 and found these differences to be due to harmonization or actualization tendencies (206), but he did not discuss the Priestly echoes identified by Seeligmann, “Preliminary Midrash” (see n. 8), 22–24, 30–31.

Compositional Harmonization

917

rash.11 (2) One of the common techniques of such editorial change involves the conflation of Priestly and Deuteronomic sources; in Seeligmannʼs words: It is clear, and it was clarified in our study, that for him [i.e., the Chronicler] there are no distinctions between a Priestly source and a Deuteronomistic one. On the other hand, even where he is aware of the contradiction that appears between different pentateuchal pronouncements, it does not occur to him to change the pentateuchal phrasings; rather he only tries to bring them into accord by means of conflation.12

This phenomenon is not at all idiosyncratic to Chronicles and has also been recognized as typical of Ezra–Nehemiah.13 In fact, the same process has been identified in the Former Prophets (especially in Joshua) and as a factor in the formation of the Pentateuch.14 For instance, Alexander Rofé demonstrated that the account of the cities of refuge in Josh 20 was composed by means of the collation of two different (and at times contradictory) statements, written by two distinct authors from two perspectives: a Priestly law (Num 35:9–34) and a Deuteronomic legal tradition (Deut 4:41–43; 19:1–13), respectively.15  Seeligmann, “Preliminary Midrash” (see n. 8), 24.  Seeligmann, “Preliminary Midrash” (see n. 8), 32 (my translation). One such example is 2 Chr 19:4–11, discussed above. Other examples of minor additions taken from Priestly sources include 2 Chr 7:8–9 and 2 Chr 35:13. In 2 Chr 7:8–9, the phrase ‫ויעשו ביום השמיני עצרת‬ is taken from Lev 23:36 (compare to Num 29:35). This slight addition (‫ )עצרת‬clearly indicates the authorʼs knowledge of the festal calendar in Lev 23. 2 Chr 35:13, the account of Josiahʼs Passover, integrates Deut 16:6–7 and Exod 12:9. For a catalogue of numerous other instances of “content (thematic) harmonizations,” see I. K alimi , The Book of Chronicles: Historical Writing and Literary Devices (Biblical Encyclopaedia Library 18; Jerusalem: Bialik, 2000), 141–162 (Hebrew). 13  Among the many discussions of this phenomenon in Ezra–Nehemiah, see D. J. A. Clines, “Nehemiah 10 as a Example of Early Jewish Biblical Exegesis,” JSOT 21 (1981), 111–117; M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985), 109–113; S. J aphet , “Law and ‘the Law’ in Ezra–Nehemiah,” in Proceedings of the Ninth World Congress of Jewish Studies, Panel Sessions: Bible Studies and the Ancient Near East (ed. M. Goshen-Gottstein; Jerusalem: World Union of Jewish Studies, 1988), 99–115; K. W. Weyde, “ ‘And They Found It Written in the Law’: Exegetical Procedures Reflected in Nehemiah 8:13–18,” in Shai Le-Sara Japhet: Studies in the Bible, Its Exegesis and Its Language (ed. M. Bar-Asher et al.; Jerusalem: Bialik, 2007), *143–*163; with specific interest in the genre of penitential prayers and specifically of Neh 9, see M. J. Boda, Praying the Tradition: The Origin and Use of Tradition in Nehemiah 9 (BZAW 277; Berlin: de Gruyter, 1999), 43–73, 189–197. 14  See Carr, Formation of the Hebrew Bible (see n. 6), 132–137, who dated to the early Hellenistic period what he specified as the “Priestly revision of earlier materials” (195–201) and the “post-D Hexateuchal compositional layer” (257–267). For many more examples of harmonization in legal passages, mostly within Deuteronomy, see B. M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (New York: Oxford University Press, 1997), 37, n. 27; 60, n. 26; 77, n. 78; 80, n. 86; 83, n. 93. 15  A. Rofé, “Joshua 20: Historico-Literary Criticism Illustrated,” in Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 11

12

918

Dalit Rom-Shiloni

What is of most significance for the present discussion is Roféʼs observation that the two source texts (traditions) were utilized in different ways. That is, harmonization does not have to be “balanced,” and indeed, usually is not. In this case the primary passage is that of P (reflected in Josh 20:1–3, 6aβ, 7–9), whereas D is used to supplement the P account (reflected mostly in Josh 20:4–6).16 To put it in terms of intertextuality, Josh 20 utilizes literary allusions to the P passage intertwined with echoes of the D passages; the author borrows key legal terminology from both.17 Furthermore, harmonization may include phrases that appear in both sources. In this case, ‫( לנוס שמה‬Num 35:11, 15; Deut 4:42; 19:3) and ‫( גאל הדם‬Num 35:15, 24, 27; Deut 19:6, 12) function as connecting phrases.18 According to Rofé, the Deuteronomistic writer of Josh 20 was clearly aware of the P legal tradition to the extent that he could quote exact phrasing and legal terminology, just as he was able to quote freely the laws of D. He was motivated to harmonize the two accounts due to the discrepancy between them in specifying the governing authority that is to adjudicate the case of the accidental killer (Num 35:12: ‫ ;כל העדה‬Deut 19:12: ‫ ;זקני עירו‬compare Josh 20:4, 6: ‫זקני העיר ההיא‬, “the elders of the city of refuge”). Realizing that the Deuteronomistic author of Josh 20 “represents a late stage in the history of the Deuteronomistic school,” Rofé thus concluded: “Since he knows both D and P, he must be dated at a rather late period, toward the end of the period of the collation and closing of the canonical Torah.”19 131–147. For other studies of this chapter, see T. C. B utler , Joshua (WBC 7; Waco, TX: Word, 1983), 210–216. 16  Josh 20 has been of primary importance for the textual criticism of Joshua. Rofé, “Joshua 20” (see n. 15), 135, argued that the LXX account follows Num 35, whereas MT strives to use both Numbers and Deuteronomy, but M. Fishbane, “Biblical Colophons, Textual Criticism, and Legal Analogies,” CBQ 42 (1980), 443–446, esp. 445–446, posited two stages of growth even within the LXX, which shares the MTʼs tendency toward harmonization. R. D. Nelson, Joshua (OTL; Louisville: Westminster John Knox, 1997), 226–231, has presented the complicated multistage process that each of the textual versions has undergone. Compare the singular verb forms referring to Joshua in LXX Josh 20:7–8 to the plural forms of MT. The LXX of these verses does not match Num 35 (where plural verbs govern, see vv. 10, 11, 12, 13, 14, etc., referring to the people as a whole). In fact, the LXX draws here on Deut 4:41–43, which describes Moses allotting cities of refuge in Transjordan, to clarify Joshuaʼs role in completing this action now, on the west side of the Jordan, in the land. Nelson categorized this feature as indicating a revision within the LXX (231). Therefore, Roféʼs conclusion concerning MTʼs motivation (see below) seems applicable to the LXX as well. 17  Rofé, “Joshua 20” (see n. 15), 140, stressed the gradual diachronic growth of this passage and aptly suggested that it went through a secondary editorial stage. 18  Pace Rofé, “Joshua 20” (see n. 15), 137, who seems to have missed this important aspect of harmonization and to have understood ‫ לנוס שמה‬solely as Priestly but to have labeled as D the phrase ‫ונס אל אחת מהערים האלה‬. For examples of harmonization that build on or include similar phrases, see the passages from Jeremiah discussed below. 19  Rofé, “Joshua 20” (see n. 15), 141.

Compositional Harmonization

919

In the wake of Roféʼs study of Josh 20, four distinctive qualities of harmonization as a literary technique may be recognized (although Rofé himself noted only the first two): (a) Harmonization is motivated by the need to reconcile contradictory (or at least different) statements and seems even more essential when legal traditions are involved. (b) There is no necessary “balance” between the evoked traditions. One may be more explicitly or extensively referenced, the other(s) supplementary in nature. (c) The evoked texts may share common phrases that function as connecting ties. (d) The later author is free to combine his target texts in a way that enhances his own rhetorical aims, a procedure that may include coining his own phraseology.20 Though greater attention has been given to harmonization as an aspect of the process of textual transmission, it is clearly not confined to this arena, just as it is not solely a redactional practice. From the standpoint of function, the collation of two or more independent literary or legal traditions into one combined text may serve different goals. Two of these functions have been thoroughly analyzed in previous scholarship. Within narrative passages, harmonizations function to resolve contradictions, discords, and/or gaps within the plot.21 Within a legal text, they are used to clarify or blend legal regulations in order to adapt these effectively to the authorʼs time and circumstances (e.g., Josh 20:1–9; Neh 8:13–18).22 In each of these genres, both the preservation of received authoritative traditions (or texts) and innovations built

20  Rofé hardly noticed this last point, which to my mind is an essential mark of the creativity of the later author. Note, for example, the unique phraseology used in Josh 20 to describe the process of allotting the cities by sanctifying (‫ )קדש‬them (Josh 20:7); compare the phrases ‫ הקרה ערים‬or ‫ נתן ערים‬and ‫ היו הערים ל׳‬in Num 35:11–15; and ‫ הבדיל ערים‬in Deut 4:41; 19:2. 21  Examples of such harmonizations have been presented by scholars holding very different points of view on the structure and evolution of the narrative sections within the Pentateuch and on the roles of redactors (or compilers) in tying them together. See, e.g., J. S. B aden , J, E, and the Redaction of the Pentateuch (Tübingen: Mohr Siebeck, 2009), 255–286, esp. 258–260; compare T. B. Dozeman, “The Priestly Wilderness Itineraries and the Composition of the Pentateuch,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman, et al.; Tübingen: Mohr Siebeck, 2011), 257–288, esp. 280–284. On coordination and harmonization as essential techniques of redaction, see Carr, “Scribal Processes” (see n. 5), 70–82. On harmonization at the level of textual transmission, see Tov, “Nature and Background of Harmonizations” (see n. 3), 7–10. Harmonization may also function to resolve semantic and syntactical difficulties; see Tov, “Nature and Background of Harmonizations” (see n. 3), 6. 22  According to Fishbane, Biblical Interpretation (see n. 13), 164–165, this is the major motivation behind the legal exegesis of Deuteronomy in reference to what he considered to be “earlier JE and P traditions.” The blending of the earlier sources, according to Fishbane (220–221), reflects the Deuteronomic authorʼs attitude that they are equally authoritative, although they represent different traditions in relation to one and the same topic.

920

Dalit Rom-Shiloni

upon those traditions operate simultaneously to create distinctive combinations of literary or legal traditions.23 The prophetic literature suggests that a third, more sophisticated function may stand behind the harmonization of pentateuchal traditions. Prophetic combinations of diverse pentateuchal traditions develop neither out of a need to correct or smooth out narrative irregularities nor out of a need to establish or fine-tune legal regulations.24 Rather, prophets intentionally harmonize pentateuchal traditions for the sake of their historical or metaphorical value; for the additional moral or ethical lesson that the pentateuchal materials may be seen to convey; and to authorize their own messages via the evocation of familiar and authoritative traditions. I will illustrate this practice through three examples from the book of Jeremiah.

III. The Proximity of Priestly and Deuteronomic References in the Book of Jeremiah: Harmonization as a Literary Phenomenon The literary evolution of the book of Jeremiah has been a subject of scholarly debate for well over one hundred years. It seems necessary here to present my own views on this issue, as they serve in this study as a point of departure. The  Noting the variety of exegetical procedures utilized to harmonize, correct, or supplement pentateuchal ritual texts, F ishbane , Biblical Interpretation (see n. 13), 222, explained the importance of these techniques as witnesses to “ancient Israelite scholastic preoccupations with diverse rules believed to stem from one authority, and for their attestation of the diversity of hermeneutical procedures which had developed or were resorted to under such conditions.” In this respect, harmonizations of legal material within late historiography clearly differ from those found in the prophetic writings of the early and late sixth century BCE. 24  This seemingly obvious observation is not universally taken for granted by scholars. References to cultic issues in Zech 1–8 and in Haggai (as well as Malachi) gave rise to the suggestion that these late prophets had understood their role as instructors and interpreters of cultic regulations closely related to priestly (theocratic) circles. This notion goes back to J. W ellhausen , Prolegomena to the History of Israel (trans. J. S. Black and A. Menzies; Edinburgh: A&C Black, 1994), 393–404, 420–422; it was elaborated by P. D. Hanson, The Dawn of Apocalyptic (2nd ed.; Philadelphia: Fortress, 1979), 172–179, 240–262. See an apt critique of this view by R. Mason, “The Prophets of the Restoration,” in Israelʼs Prophetic Tradition: Essays in Honour of Peter R. Ackroyd (ed. R. J. Coggins et al.; Cambridge: Cambridge University Press, 1982), 137–154. I concur with the perspective that recognizes that these later prophets use pentateuchal traditions metaphorically, in keeping with the practice of their prophetic predecessors in the Neo-Babylonian period; see D. R om -S hiloni, “What Is ‘Persian’ in Late Sixth-Century BCE Prophetic Literature? Case Studies and Criteria,” in Discerning Criteria for Dating Persian Period Texts (ed. R. J. Bautch and J. D. Nogalski; Atlanta: Society of Biblical Literature, forthcoming), 1–45, esp. 17–32; idem , Exclusive Inclusivity: Identity Conflicts between the Exiles and the People who Remained (6th–5th Centuries BCE) (LHBOTS 543; London: Bloomsbury T&T Clark, 2013), 62–70. 23

Compositional Harmonization

921

book of Jeremiah began to take shape in the late seventh and early sixth centuries BCE, and the complicated evolution of the book stretches a few generations further, though still within the sixth century.25 This general framework does not exclude the possibility of minor additions until the end of the fifth century BCE. But the great bulk of material (by the prophet, his Judean tradents, and his Babylonian tradents and editors) seems to have originated during the Neo-Babylonian era; by the early Persian period, this material had already been compiled and was being edited and supplemented (presumably by both Babylonian tradents and Babylonian repatriates in Yehud).26 In recognition of this gradual growth of the book and the multiple participants in its authorship, I use the designation Jeremiah in this paper to stand for the prophet and for his followers (tradents, editors) within this sixth-century timeframe.27 25  This means that the book of Jeremiah had assumed more or less its final shape by the eras of Second Isaiah, Haggai, and Zech 1–8. On the use of the Pentateuch in Second Isaiah, see B. D. Sommer, A Prophet Reads Scripture: Allusion in Isaiah 40–66 (Stanford: Stanford University Press, 1998), 132–151; although Sommer noticed harmonization in only a few passages (e.g., Isa 56:1–5 [325]; 61:1–4 [328]). On allusions to and interpretations of Jeremiah in Zech 1–8, see M. R. Stead, The Intertextuality of Zechariah 1–8 (LHBOTS 506; New York: T&T Clark, 2009). 26  The sixth century as the time of literary evolution of the book of Jeremiah was suggested by several scholars, including E. W. Nicholson, Preaching to the Exiles: A Study of the Prose Tradition in the Book of Jeremiah (New York: Schocken, 1970), 116–136; C. Seitz, Theology in Conflict: Reactions to the Exile in the Book of Jeremiah (BZAW 176; Berlin: de Gruyter, 1989), 222–235, 282–296; A. Rofé, “Studies on the Composition of the Book of Jeremiah,” Tarbiz 44 (1974–1975), 1–29 (Hebrew); idem, “The Arrangement of the Book of Jeremiah,” ZAW 101 (1989), 390–398; J. R. Lundbom, Jeremiah 1–20 (AB 21A; New York: Doubleday, 1999), 92–102, esp. 100–101; M. Leuchter, The Polemics of Exile in Jeremiah 26–45 (Cambridge: Cambridge University Press, 2008), 1–24; and see this framework in his earlier book, idem , Josiahʼs Reform and Jeremiahʼs Scroll: Historical Calamity and Prophetic Response (Hebrew Bible Monographs 6; Sheffield: Sheffield Phoenix, 2006); and see D. Rom-Shiloni, “Group-Identities in Jeremiah: Is It the Persian Period Conflict?,” in A Palimpsest: Rhetoric, Stylistics, and Language in Biblical Texts from the Persian and Hellenistic Periods (ed. E. Ben-Zvi et al.; PHSC 5; Piscataway, NJ: Gorgias, 2009), 11–46. For a recent linguistic study of Jeremiah that further validates its sixth-century context, see A. Hornkohl, Ancient Hebrew Periodization and the Language of the Book of Jeremiah: The Case for a Sixth-Century Date of Composition (Leiden: Brill, 2014), 370–74. 27  Prominent, mostly in European scholarship, are suggestions for a much longer process of literary evolution that came to fruition only within the late Persian period, by the fifth and possibly the fourth centuries BCE. For Jeremiah as a literary tradition, and his book as serving theological and ideological needs of the Persian period exilic/postexilic golah community of the fifth century BCE, see R. C arroll , Jeremiah (OTL; Philadelphia: Westminster, 1986), 65–82. For the bookʼs evolution as a rolling corpus, see W. McKane, Jeremiah I–XXV (ICC; Edinburgh: T&T Clark, 1986), l–lxxxiii. Prolific studies on redaction criticism of the book have been written by German scholars; the list cannot be exhaustive, but see K-F. Pohlmann, Studien zum Jeremiabuch (FRLANT 118; Göttingen: Vandenhoeck & Ruprecht, 1978); G. F ischer , Jeremiah 1–25 (HThKAT; Freiburg: Herder, 2005), 75–93; idem , Jeremia:

922

Dalit Rom-Shiloni

Intensive scholarly attention has been given to similarities in phrasing and theme between Jeremiah and Deuteronomy (along with the Deuteronomistic historiographical literature). To a much lesser extent have scholars noted the affinities, references, and even literary allusions in Jeremiah to Priestly (P and HL) materials.28 The present study addresses the harmonization of Priestly (P/HL) and Deuteronomic references (by combination or collation) in three poetic passages in Jeremiah (Jer 2:3; 3:1–5; and 49:1–2). For each of these passages, I will identify characteristic P/HL and D vocabulary, conceptions, and phrasings used by the prophet and trace the transformations in meaning brought about by their combinations in their new contexts. These examples allow careful consideration of whether such harmonizations indicate actual literary dependence per se (i.e., allusion to written pentateuchal texts) or whether they simply indicate the knowledge of D and P/HL terminology and traditions, which the prophet blends together. Each of these examples presents serious methodological challenges, which I will articulate in each case. They challenge our presumptions about prophetic acquaintance with such pentateuchal materials, and they invite us to consider the rhetorical functions of these combinations of D and P/HL materials within the prophetic context. The suggestion I want to advance is that Jeremiah (the prophet and/or his tradents) is familiar with pentateuchal materials in their preliminary stages (clearly not with a written and redacted book of the Torah). On the basis of these specific case studies, it will be possible to ask more generally what are the features of the usage of compositional harmonizations as a literary technique within the earliest stages of this prophetic book.29 Der Stand der theologischen Diskussion (Darmstadt: Wissenschaftliche Buchgesellschaft, 2007), 131–148; E. O tto , “Jeremia und die Tora: Ein nachexilischer Diskurs,” in Tora in der He­bräischen Bibel: Studien zur Redaktionsgeschichte und synchronen Logik diachroner Trans­for­ma­tionen (ed. R. Achenbach et al.; BZABR 7; Wiesbaden: Harrassowitz, 2008), 134–182; R. Achenbach, “The Pentateuch, the Prophets, and the Torah in the Fifth and Fourth Centuries B.C.E.,” in Judah and the Judeans in the Fourth Century B.C.E. (ed. O. Lipschits et al.; Winona Lake, IN: Eisenbrauns, 2007), 253–285; K. Schmid, Buchgestalten des Jere­mia­ buches: Unter­suchungen zur Redaktions- und Rezeptionsgeschichte vor Jer 30–33 im Kontext des Buches (WMANT 72; Neukirchen-Vluyn: Neukirchener Verlag, 1996), 327–354; idem, “Jeremiah,” in T&T Clark Handbook of the Old Testament: An Introduction to the Literature, Religion and History of the Old Testament (ed. J. C. Gertz et al.; trans. J. Adams-Massmann; London: T&T Clark, 2012), 431–450. See further the papers of R. Achenbach and G. Fischer in this volume. 28  M. A. S weeney , Tanak: A Theological and Critical Introduction to the Jewish Bible (Minneapolis: Fortress, 2012), 294. For a few examples, see J. Milgrom, Cult and Conscience: The Asham and the Priestly Doctrine of Repentance (SJLA 18; Leiden: Brill, 1976), 70–74; Fishbane, Biblical Interpretation (see n. 13), 302–303; and see the discussion at n. 32 below and the articles referenced in n. 2 above. 29  A broader study of Priestly references (of various kinds) in Jeremiah, entitled “One of the Priests at Anathoth”: Priestly Traces in Jeremiah, is now in preparation. In that study, I will

Compositional Harmonization

923

1 Harmonization of P/HL and D: The People of God as Sacred Donation (Jer 2:2–3) Jeremiah 2:3 is asyndetically attached to v. 2, which describes (very positively!) the God-people relationship at its constitution in the desert. Utilizing two metaphorical worlds, the realm of marriage and the sphere of shepherding, v. 2 commemorates the peopleʼs loyalty to God and Godʼs role in leading his people safely through the desert.30 Verse 3 introduces a different set of metaphors using a cluster of legal and cultic terms, which focus on Israelʼs status before God and particularly on the constant, divine protection guaranteed to the nation.31 The search for markers that may reveal the sources for the imagery of Jer 2:3, and thus expose its literary dependencies on pentateuchal materials, yields three texts. Each makes a distinct contribution to the prophetic proclamation, and in fact there is a sort of hierarchical sequence to the Priestly (P/HL) and Deuteronomic allusions and echoes:32 Jeremiah 2:3

Deuteronomy 7:6

Israel was holy to the LORD (‫קדש ישראל‬ ‫)ליהוה‬

For you are a people consecrated to the LORD your God (‫)כי עם קדוש אתה ליהוה אלהיך‬ Numeri 18:8–12

the first fruits of His harvest (‫ראשית‬ ‫)תבואתה‬

8 The LORD spoke further to Aaron: I hereby give you charge of My gifts, all the sacred donations of the Israelites; I grant them to you and to your sons as a perquisite, a due for all time. [. . .] 10 You shall partake of them as most sacred donations (‫)בקדש הקדשים תאכלנו‬: only males may eat them; you shall treat them as consecrated (‫[ )קדש יהיה לך‬. . .] 12 All the best of the new oil, wine, and grain – the choice parts that they present to the LORD (‫)ראשיתם אשר יתנו ליהוה‬ – I give to you.

thoroughly analyze the phenomenon of harmonization. Among the prose passages that employ the same technique, see Jer 32:6–15 and 34:8–22. 30  For the use of marital metaphors in Jeremiah, see D. Rom-Shiloni, “The Covenant in the Book of Jeremiah: On the Employment of Marital and Political Metaphors,” in Covenant in the Persian Period: From Genesis to Chronicles (ed. R. J. Bautch and G. N. Knoppers; Winona Lake, IN: Eisenbrauns, 2015), 153–174. 31  The connection between vv. 2 and 3 has been discussed by commentators, usually with a focus on the lack of grammatical agreement between the two. See Lundbom, Jeremiah 1–20 (see n. 26), 250; W. L. H olladay , Jeremiah 1 (Hermeneia; Philadelphia: Fortress, 1986), 84–85; and Fishbane, Biblical Interpretation (see n. 13), 300–301. 32  Translations of biblical passages follow the NJPS edition, with occasional alterations. Italics stand for D phraseology, and underlined words designate P/HL references.

924

Dalit Rom-Shiloni

Leviticus 22:14–16

all who ate of it were held guilty (‫כל אכליו‬ ‫)יאשמו‬ disaster befell them – declares the LORD.

14 But if a man eats of a sacred donation (‫ואיש‬ ‫ )כי יאכל קדש‬unwittingly, he shall pay the priest for the sacred donation (‫)הקדש‬, adding one-fifth of its value. 15 But (the priests) must not allow the Israelites to profane the sacred donations that they set aside for the LORD, 16 or to incur guilt requiring a penalty payment, by eating such sacred donations (‫עון‬ ‫)אשמה באכלם את קדשיהם‬: for it is I the LORD who make them sacred.

The markers ‫קדש‬, ‫אכל‬, and ‫ אשם‬tie Jer 2:3 to the HL tradition concerning sacred donations (Lev 22:1–16) and specifically to its closing verses, 14–16. According to both P and HL, the eating of sacred donations, termed qodesh, is restricted to male priests33 and is to be carried out in a sacred place (Exod 29:33–34; Lev 6:7–11 [v. 9]; 6:17–23 [v. 22]; 7:1–10 [v. 6]; 24:9). Among those donations allotted to the priests, first fruits (of crops and herds) hold a special position (Num 18:8–20).34 Jacob Milgrom and Michael Fishbane defined Jer 2:3 respectively as “an aggadic midrash” and “an aggadic exegesis of an old priestly law.” They each emphasized the metaphorical analogy drawn by the prophet from the Priestly materials.35 Fishbane aptly argued for the idiosyncratic exegetical utilizations of each of the markers in their (new) prophetic context; though he did recognize the asymmetrical relationship between Jer 2:3 and Lev 22:14–16, he nevertheless found that the tensions between the two passages did not “subvert the rhetorical force and analogical power of the aggadic application. Since Lev 22:14–16 serves Jeremiahʼs oration, the several asymmetries are obscured and transcended by the more forceful hermeneutical analogies at play.”36 According to Fishbane, then, Lev 22:14–16 is the older “A” text, alluded to by Jer 2:3, the more recent “B” text. I will focus here on the exegetical techniques of analogy and transformation employed in this verse, because the use of such techniques holds important implications for the role of the prophet in adapting these pentateuchal materials, just as these techniques may shed light on the scope of the pentateuchal mate33  Although Num 18:19 specifies that qodashim, like certain other offerings (see Num 18:11), may be eaten by “your daughters” as well. 34  For the special status of “the most sacred precincts,” see B. A. Levine, Numbers 1–20 (AB 4A; New York: Doubleday, 1993), 435–437, 443–445. Outside P/HL sources, only Ezek 42:13 refers to priests eating sacred foods, and indeed the three major offerings (‫המנחה‬ ‫והחטאת והאשם‬, “the meal offerings, the sin offerings, and the guilt offerings”) are mentioned. 35  J. Milgrom, Cult and Conscience (see n. 28), 70–74; Fishbane, Biblical Interpretation (see n. 13), 301, 303. 36  Fishbane, Biblical Interpretation (see n. 13), 303.

Compositional Harmonization

925

rials at the prophetʼs disposal. To begin, however, a general question should be raised as to whether Jer 2:3 may actually be said to be working with a text of Lev 22:14–16. A look at the prophetʼs use of the root ‫ קדש‬provides an insight toward an answer to this question. Although Jeremiah uses the HL term qodesh as a metaphor for Israel, syntactically the prophet constructs this metaphor within the unique expression ‫קדש ישראל ליהוה‬, which in turn seems to be based on the Deuteronomic phrase ‫( עם קדוש אתה ליהוה אלהיך‬Deut 7:6). Quite remarkably, D itself appropriates the root ‫ קדש‬in its HL meaning of dynamic (and cultic) sanctity in order to give it a national (ethnic) meaning, designating the entire people of Israel as “belonging to,” or “a possession of,” God (cf. Deut 7:6; 14:2, 21; 26:19; 28:9).37 Hence, a relative chronology may be outlined: in P, the concept of sanctity is restricted to priests and sacred objects; HL broadens this conception to apply to the people as a whole; and D extends the concept to configure this general national conception of holiness as the unique and permanent status of the people of Israel before God.38 Jeremiah 2:3 utilizes Dʼs transformation of the HL term without giving up the dynamic HL meaning and the specific cultic context of sanctum.39 In fact, Jeremiah builds on and combines all three of these pentateuchal conceptions, invoking the holy status of Israel in the face of political enemies; thus he stands last in this relative ordering of the development of the concept. The phrase ‫“( כל אכליו יאשמו‬all who ate of it were held guilty”) illustrates this last point. The prophet merges the law concerning sacred donations with the conception that God acts as savior of his people from the enemies who threaten 37  B. J. Schwartz, The Holiness Legislation: Studies in the Priestly Code (Jerusalem: The Hebrew University Magnes Press, 1999), 255–258 (Hebrew), discussed the P usage of qodesh as sanctum, i.e., persons or items that are considered personal possessions of God; he noted that the theme of Israel (the people) as holy (sanctified) appears only in HL among the priestly materials (260–266). Schwartz drew an apt distinction between HLʼs conception of sanctification as a continuous, dynamic process, leading to the demand that the people be constantly sanctified (262), on the one hand, and non-P and D categorizations of Israel as permanently sanctified to God as result of their past election, on the other (e.g., Exod 19:6; Deut 7:6; 14:21; 26:16–19; see pp. 258–260, 261). 38  It would not make a difference to the present discussion to argue that HL is later than D (and P), as was suggested, for instance, by B. M. Levinson, “The Manumission of Hermeneutics: The Slave Laws of the Pentateuch as a Challenge to Contemporary Pentateuchal Theory,” in Congress Volume: Leiden, 2004 (ed. A. Lemaire; VTSup 109; Leiden: Brill, 2004), 281–324. These developments within the Pentateuch belong in any case to a stage prior to Jeremiah, who seems to be aware of all the various strands of tradition; see discussion below. 39  M ilgrom , Cult and Conscience (see n. 28), 70–74; F ishbane , Biblical Interpretation (see n. 13), 302–303. Both Milgrom and Fishbane argued that Jeremiahʼs transformation expanded the Deuteronomic conception, yet both scholars drew Ezra 9:2 into their discussion, a verse which, to my mind, is completely irrelevant both to this Jeremianic passage and to the Deuteronomic conception of the peopleʼs sanctity.

926

Dalit Rom-Shiloni

to devour them. The prophet uses the HL terminology that denotes eating the sacred donation (‫ )אכל קדש‬to describe the treatment of Israel by its enemies as the forbidden consumption of that which is sacred. Thus, the entire HL phrase takes on a metaphoric value in its new context (as Fishbane pointed out), and precisely this metaphoric usage illustrates the large gap between the law and the prophecy.40 The clear invocation of HLʼs ban on consumption of the qodesh by prohibited participants indicates that the prophet knows and retains the basic HL analogy of Israel as qodesh; he attracts to this context his acquaintance with the P traditions of first-fruit donations and their value (‫ )ראשית תבואתה‬as part of the priestly donations considered qodesh (cf. Num 18:12). But just as clearly, Jer 2:3 distances itself from the HL legal context: (1) The prophecy carries forward the transfer of the concept of qodesh from the ritual context of P/HL into Dʼs political realm, where the term denotes the status of Israel among the nations (Deut 7:6; 26:19; 28:9); the term is set within the framework of Godʼs commitment to Israelʼs safety in the face of its enemies (following Deut 7, among other passages). Thus, the needs of the prophecy both drive the selection of this pentateuchal legal tradition and control its use. The prophet has no interest in (and thus does not reflect) other components of the legal context of Lev 22:1–16.41 (2) The prophecy recasts ‫ אכל‬in a new metaphorical context to refer to military defeat and destruction (as in Deut 32:42 and elsewhere in prophetic literature: Jer 30:16; Amos 1:4, 7, 10, 12, etc.). (3) Finally, the prophecy strips ‫ אשם‬of its multiple HL cultic meanings – as referring to sin (Lev 4:13; 5:15); judgment (Ps 34:22; Hos 5:15); and the ‫אשם‬ reparation sacrifice (Lev 5:6, 7; cf. Lev 5:19–25, which brings all three senses together) – and gives it the single sense of the guilt incurred by Israelʼs enemies  Fishbane, Biblical Interpretation (see n. 13), 303.  Lev 22:1–16 (HL) addresses the priests concerning three specific matters that require special care: (1) the prohibition against profaning the sacred by eating it in a defiled state (22:1–9); (2) the prohibition against strangers (i.e., nonpriests) eating from the sacred offerings (vv. 10, 12), complemented by the list of those who are permitted to enjoy these offerings because they are included in the priestʼs social unit (vv. 11, 13); and (3) the responsibility of the priests to prevent the wrong person (‫איש‬, v. 14; compare to ‫זר‬, “stranger,” in v. 10) from eating the sacred offerings and the ritual reparations required when such a mistake is made (vv. 14–16). Verses 15–16 accentuate the priestsʼ potential role in causing such an offense: “But they (the priests) must not allow the Israelites to profane the sacred donations that they (the Israelites) set aside for the LORD; and they (the priests) will incur their (the Israelitesʼ) guilt, requiring a penalty payment, when they (the priests) eat their (the Israelitesʼ) sacred donations: for it is I the LORD who make them sacred.” The law focuses on the corrective procedures when such a mistake has been made. It is the priests who are obliged to correct the situation when their negligence has allowed a prohibited person to eat from the sacred offerings (see n. 42). See J. Milgrom, “The Cultic ‫ שגגה‬and Its Influence in Psalms and Job,” JQR 58 (1967), 115–125, reprinted as Studies in Cultic Theology and Terminology (SJLA 36; Leiden: Brill, 1983), 122–32, esp. 122–27. 40

41

Compositional Harmonization

927

for consuming the qodesh.42 Thus, the prophecy lifts all three terms from a semantic context describing inadvertent cultic error under priestly supervision and reapplies them in a political context to describe the deliberate destruction of the nation by its political enemies. Two summary points are in order. First, these observations challenge our perceptions of the prophecyʼs (or the prophetʼs) accountability to the HL law of Lev 22:14–16. The dissimilarities between Jer 2:3 and Lev 22:14–16 show that the prophecy atomizes the aforementioned HL terminology and legal conceptions; the prophet feels free to transform them and even to embed them in a totally different conceptual framework. Governed by Deuteronomic thought, the prophecyʼs exegetical intention is to illustrate the God-people relationship and Godʼs commitment to his people. These exegetical techniques anticipate the atomization techniques of later midrash and suggest a genuine harmonization of diverse sources under one uniting rubric, ‫קדש‬. Second, if we consider these verses to be “authentic,” we see Jeremiah, or his contemporary tradents, uniting Priestly terminology (and conceptions) with a Deuteronomic worldview and intertwining them in this prophecy to create entirely original meanings.43 This example suggests a much richer picture of the prophetic use of invoked traditions and texts. The prophecy (the B text) clearly alludes to Priestly sources (HL included), yet it does not engage in a literary dialogue with one main text (Lev 22:14–16, in this case) as its A text. In fact, the supposed A text seems rather to comprise a tapestry of different legal traditions, which are freely combined. Against Fishbaneʼs conclusions, one may in fact ask whether the prophet has the actual HL written text at his disposal or whether the prophecy instead makes use of Priestly (HL and P) terminology and legal knowledge based on either oral or disparate literary sources. This question deserves further investigation through the study of other utilizations of Priestly (P and HL) traditions in the book of Jeremiah. 42  For ‫ אשם‬in its different meanings, see Milgrom, Cult and Conscience (see n. 28), 3. This transformation explains the significant gap between ‫ כל אכליו יאשמו‬and the unique phraseology of Lev 22:16: ‫והשיאו אותם עון אשמה באכלם את קדשיהם‬. Martin Noth translated this phrase: “and so cause them to bear iniquity and guilt, by eating their holy things” (M. Noth, Leviticus [OTL; London: SCM Press, 1965], 158); he explained that the priestsʼ violation of this instruction (v. 15) “would involve all Israel in blame” (162); see also J. Milgrom, Leviticus 17–22 (AB 3A; New York: Doubleday, 2000), 1866–1868: “they (the priests) shall not permit the sacred donations of the Israelites [. . .] to become profane by causing them (the Israelites) eat their (own) sacred donations.” Both the Aramaic Targums and medieval Jewish interpreters struggled over the identification of the subject and the object in this verse (see Rashi and Rashbam), but none of these hermeneutical difficulties are reflected in Jer 2:3. 43  There is a widespread consensus among scholars of Jeremiah that chapters 2–10, and certainly the poetic passages within these chapters, are “authentic”; see Y. Hoffman, Jeremiah 1–25 (Mikra Leyisra‍ʾel; Jerusalem: The Hebrew University Magnes Press; Tel Aviv: Am Oved, 2004), 122–123 (Hebrew).

928

Dalit Rom-Shiloni

2 Collation of D and P: Divorce and Defilement of the Land (Jer 3:1–5) Jeremiah 3:1–5 1 as follows: If a man divorces his wife, and she leaves him and marries another man, can he ever go back to her? (‫הן ישלח איש‬ ‫את אשתו והלכה מאתו והיתה לאיש אחר הישוב‬ ‫ )אליה עוד‬Would not such a land be defiled (‫ ?)הלוא חנוף תחנף הארץ ההיא‬Now you have whored with many lovers: can you return to me? (‫)ושוב אלי‬ – says the LORD. 2 Look up to the bare heights, and see: Where have they not lain with you? You waited for them on the roadside like a bandit in the wilderness. And you defiled the land with your whoring and your debauchery (‫)ותחניפי ארץ בזנותיך וברעתך‬. 3 And when showers were withheld and the late rains did not come, you had the brazenness of a street woman, you refused to be ashamed. 4 Just now you called to me, “Father! You are the companion of my youth. 5 Does one hate for all time? Does one rage forever?” That is how you spoke; you did wrong, and had your way.

Deuteronomy 24:1–4 1 A man takes a wife (‫)כי יקח איש אשה‬ and possesses her. She fails to please him because he finds something obnoxious about her, and he writes her a bill of divorcement, hands it to her, and sends her away (‫)ושלחה‬ from his house; 2 she leaves his household and becomes the wife of another man (‫והלכה‬ ‫;)והיתה לאיש אחר‬ 3 then this latter man rejects her, writes her a bill of divorcement, hands it to her, and sends her away (‫ )ושלחה‬from his house; or the man who married her last dies. 4 Then the first husband who divorced her shall not take her to wife again, since she has been defiled (‫לא יוכל בעלה הראשון אשר‬ ‫שלחה לשוב לקחתה להיות לו לאשה אחרי אשר‬ ‫)הטמאה‬ – for that would be abhorrent to the LORD. You must not bring sin upon the land that the LORD your God is giving you as a heritage (‫כי תועבה היא לפני יהוה ולא תחטיא‬ ‫)את הארץ אשר יהוה אלהיך נתן לך נחלה‬. Numbers 35:33–34 33 You shall not pollute the land in which you live (‫;)ולא תחניפו את הארץ אשר אתם בה‬ for blood pollutes the land, and the land can have no expiation for blood that is shed on it (‫)כי הדם יחניף את הארץ‬, except by the blood of him who shed it. 34 You shall not defile the land in which you live (‫ולא תטמא את‬ ‫)הארץ אשר אתם ישבים בה‬, in which I myself abide, for I the LORD abide among the Israelite people.

Among the prophetic units of chapters 2–3, Jer 3:1–5 takes an extremely negative position on the God-people relationship.44 The prophecy is constructed as a disputation speech: the peopleʼs plea, quoted in vv. 4–5, is refuted by means of an allusion to the Deuteronomic law of divorce (Deut 24:1–4). Yet, at a crucial point, the prophecy purposefully detached from its Deuteronomic source and  I have dealt with this passage in an earlier paper; see D. Rom-Shiloni, “Actualization of Pentateuchal Legal Traditions in Jeremiah: More on the Riddle of Authorship,” ZABR 15 (2009), 254–281, esp. 262–267. 44

Compositional Harmonization

929

chose Priestly terminology that opens up a whole different, and harsher, conceptual world. Hence, this unit suggests an intentional collation of D and P. The allusions to the law of divorce in Deut 24:1–4 are clearly apparent. The Deuteronomic law is seen as a precedent; note the terminology ‫ הן ישלח‬in the opening phrase of the Jeremiah passage (cf. D: ‫)כי יקח‬.45 Also allusive are the repeated phrases, including ‫( איש את אשתו‬cf. D: ‫ )איש אשה‬and the recurring verbs or verbal phrases ‫ישלח‬, ‫והיתה לאיש אחר‬, and ‫שוב‬. The major thematic connection is the prohibition against reactivating the first marriage after the former wife has been married to another man (Deut 24:4a; see the question as posed in Jer 3:1: ‫)הישוב אליה עוד‬.46 The choice to construct his prophecy on the Deuteronomic law of divorce is governed by Jeremiahʼs desire to refute the peopleʼs expression of obedience to God, which itself uses familial metaphors of adoption and marriage (e.g., “Father! You are the companion of my youth” in v. 4b).47 Adopting the same metaphorical framework, the prophet treats these assertions of faithfulness as hypocritical and deceptive in the face of what he considers to be the peopleʼs continuous disobedience against God (“That is how you spoke, you did wrong, and had your way,” Jer 3:5b).48 It is this disobedience that evokes the language of adultery and divorce. The legal formulation of Deuteronomy (Deut 24:1–4) is invoked in Jer 3:1 to frame the pivotal question, “Can he ever go back to her?” But, in fact, the Deuteronomic legal prohibition in Jer 3:1 leads to two rhetorical questions. The first follows the formulation of the law, which establishes both the marriage and the divorce as entirely the prerogative of the man, whereas the second reverses the direction of return, asking, “Can you return to me?” This reversal is not  On the function of ‫ הן‬as similar to that of ‫ כי‬in legal contexts, see Jer 2:10 and compare Lev 10:18; 25:20, etc. B. O. Long, “The Stylistic Components of Jeremiah 3,1–5,” ZAW 88 (1976), 386–390, suggested that ‫ הן‬introduces legal or cultic circumstances which are adduced as the subject of the following rhetorical question (as also in Hag 2:10–14). 46  For a comparative study of this law in biblical, postbiblical, and extrabiblical sources, consult R. Yaron, “The Restoration of Marriage,” JSS 17 (1966), 1–11; for the relationship between the Deuteronomic law and the marriage stories of David and Michal and the prophet Hosea, see J. D. Martin, “The Forensic Background of Jeremiah III 1,” VT 19 (1969), 82–92. In my earlier article, Rom-Shiloni, “Actualization” (see n. 44), 262–267, I have reconstructed the entire interpretive procedure based on these literary and thematic connections, so I include here only the necessary points for the current discussion. 47  ‫ אבי אלוף נערי אתה‬brings together adoption and marriage formulae; note ‫ אלוף נערי‬in Prov 2:16–17 and see other examples in S. M. Paul, “Adoption Formulae: A Study of Cuneiform and Biblical Legal Clauses,” Maarav 2/2 (1979–1980), 173–185. The resemblance between the quotation in Jer 3:4–5 and Ps 103:9 gives credence to the content of these words the prophet quotes, and possibly even to its actual wording. Compare this phrasing to the consolation expressed in Jer 3:12. 48  This, then, reinforces Longʼs definition of this passage as a “disputation” (Long, “Stylistic Components” [see n. 45], 386–390). 45

930

Dalit Rom-Shiloni

coincidental but is the prophetʼs deliberate reaction to the peopleʼs claims of obedience in vv. 4–5.49 The Deuteronomic law functions here as the basis of a direct analogy (qal vahomer); just as the Deuteronomic law forbids the man to remarry his former wife (literally: “to return and take her as his wife”), how much more improbable it is that God can or will return to his people – that is, the woman who has constantly and repeatedly been disloyal to him, with many men.50 Using this civil law, presumably known to the prophet and to his audience, the prophecy presents an extremely negative view of the God-people relationship, such that it portrays the covenant relationship as irredeemable because of the womanʼs/ the peopleʼs sins.51 But this is where Jeremiahʼs allusions to Deut 24:1–4 end. In Deut 24, the terms ‫טמא‬, ‫החטיא‬, and ‫“( תועבה‬to defile,” “to bring sin,” and “abhorrent”) designate the severity of the transgression in remarrying the former wife.52 In 49  The womanʼs (the peopleʼs) initiative to return to God as the former husband is also portrayed in Hos 2:9, where the possibility is likewise denied in a fierce prophecy of judgment up to v. 15 (and in distinction to Hos 2:16–25). Note the Septuagint on Jer 3:1: ‫הישוב אליה עוד‬: μὴ ἀνακάμπτουσα ἀνακάμψει πρὸς αὐτὸν ἔτι, which Holladay, Jeremiah 1 (see n. 31), 113, translated as “shall she really return again to him?” suggesting the Hebrew Vorlage ‫השוב תשוב‬ ‫( אליו עוד‬in line with Hos 2:9). Therefore, contra J. P. Hyatt, “Torah in the Book of Jeremiah,” JBL 60 (1941), 381–396, there is no reason to accept the Septuagint version of Jer 3:1 as original or better. In addition, there is no need to assume a third legal source earlier than both Jeremiah and Deuteronomy, pace T. R. Hobbs, “Jeremiah 3:1–5 and Deuteronomy 24:1–4,” ZAW 86 (1974), 23–29; note also the criticism in Long, “Stylistic Components” (see n. 45). 50  The understanding of Jer 3:1 as a qal vahomer and thus as adducing a break in the God-people relationship was suggested by both McKane, Jeremiah I–XXV (see n. 27), 63–64, and J. R. Lundbom, Jeremiah: A Study in Ancient Hebrew Rhetoric (Winona Lake, IN: Eisenbrauns, 1997), 54; idem, Jeremiah 1–20 (see n. 32), 301–302. Compare Yaron, “Restoration” (see n. 46), 3, and Fishbane , Biblical Interpretation (see n. 13), 307–312, who argued that Jeremiah reverses the law in order to create a theological tension between the law and the principle of repentance. Although I disagree with Fishbane on this overall interpretation of Jeremiahʼs usage of the Deuteronomic law, I do concur with his observation that Jeremiah “has no legal force with respect to civil procedures” (311). 51  Jeremiahʼs use of this imagery thus differs from that of Hosea (2:4–15; 3:1–5); see W. L. H olladay , Jeremiah 2 (Hermeneia; Minneapolis: Fortress, 1989), 113. This extreme view of the God-people relationship does not recur in other prophecies in the book of Jeremiah, though it does occur in the communal laments quoted in Jer 14:7–9, 19–22; in certain prophecies of consolation, as in Jer 31:35–37; and in passages where the prophet is trying to refute a prevailing perception, as in 33:23–26. This pronouncement in Jer 3 may therefore be understood as an authentic ad hoc reaction to the peopleʼs words, and there is no need to try to harmonize these diverse prophetic statements; compare, for instance, Martin, “Forensic Background” (see n. 46), 91–92. 52  Since both verbs and the noun do occur in Jeremiah, their absence from Jer 3:1–5 is clearly deliberate; note ‫( טמא‬Jer 7:30; 32:34; and in a quote in 2:23); ‫( החטיא‬32:35); and ‫תועבה‬ (6:15; 8:12; 32:35).

Compositional Harmonization

931

other contexts, the prophet indeed uses these terms, but here he deliberately chooses to use the verb ‫ חנף‬and the phrase ‫החניף) ארץ‬/(‫חנף‬: “Would not such a land be polluted?” (‫הלוא חנוף תחנף הארץ ההיא‬, v. 1); “and you polluted the land with your whoring and your debauchery” (‫ותחניפי ארץ בזנותיך וברעתך‬, v. 2).53 Beyond the semantic difference, in using the phrase ‫החניף) ארץ‬/( ‫חנף‬, Jeremiah transforms and expands the Deuteronomic law of divorce by echoing the Priestly legal phraseology and terminology of Num 35:33–34. Thematically, this transformation intensifies the sin, which is said to pollute the land (a transformation that is anchored in the phrase ‫ולא תחטיא את הארץ‬, Deut 24:4). Jeremiah expands this imagery to encompass the concept of the connection of the fate of the land with the people who sin within it (this is further developed in Jer 3:2–3).54 The literary relationship of Jer 3:1–5 to this P tradition/concept calls for clarification. Though there is a clear phraseological similarity between Jer 3:1–5 and Num 35:33–34 in the use of the verb ‫ חנף‬with ‫ ארץ‬as its object, the prophecy differs from the Priestly passage, which connects the landʼs pollution with bloodshed in the context of murder laws (“for blood pollutes the land, and the land can have no expiation for blood that is shed on it,” v. 33).55 The Priestly phrase ‫ חנף ארץ‬has its ideological roots in the P/HL conception of the land. According to HL, the two domains related to blood – cultic bloodshed (i.e., child sacrifice) and sexual sin – defile (‫ )טמא‬the land to the point that it vomits out its inhabitants (see Lev 18 and 20, specifically 18:24–30; 20:22–24). Jeremiah uses ‫ חנף‬five times out of its total of eleven occurrences in the entire Hebrew Bible. In three of these instances, the verb designates the pollution of the land as the consequence of idolatry, here labeled as adultery (Jer 3:1, 2, and 9); at Jer 23:11, prophet and priest ‫חנפו‬, and as a result “‫ חנפה‬has gone forth to the whole land” (23:15).56 In all these contexts, sexual misconduct (actual or metaphorical) is said to pollute the land (see 3:1–5, 6–10; 23:10–11, 13–15). Using the recurring prophetic imagery of the worship of other gods as adultery, the prophet further develops the marital metaphor by suggesting a second transformation in the understanding of the Deuteronomic law. Although the Deuteronomic law of divorce is clearly the prophetʼs central text, the law itself lacks any clear accusation of adultery. Note that defilement and abomination in  Holladay, Jeremiah 1 (see n. 31), 113, pointed out Jeremiahʼs intentional usage of ‫חנף‬ as a harsher verb that associates the transgression of adultery with bloodguilt (Num 35:33; Ps 106:38) and with blood shed by violence. Yet Holladay did not note the centrality of ‫ חנף‬in its double meaning in Jeremiahʼs prophecy (see the discussion below). 54  For a different suggestion, see M. F ishbane , “Torah and Tradition,” in Tradition and Theology in the Old Testament (ed. D. A. Knight; Philadelphia: Fortress, 1977), 285–286. 55  Beside Num 35 and Jer 3, ‫ חנף ארץ‬occurs only in Isa 24:5 and Ps 106:38 (which seems to allude to the bloodshed of Num 35:33). 56  Holladay translated ‫ חנף‬as “commit sacrilege”; see Holladay, Jeremiah 1 (see n. 31), 624 and the discussion on 627–628. The translation “godless” (so NJPS and others) gives only a vague sense of the meaning and loses the moral, and even sexual, aspect of the term. 53

932

Dalit Rom-Shiloni

Deut 24:4 refer to remarriage to the first husband after a perfectly legal second marriage.57 This may explain why Jeremiah added Priestly materials at this point, an addition that allowed him to step beyond the context of the Deuteronomic law. With ‫ חנף ארץ‬the prophet brings the law of divorce together with the realm of adultery; this, in turn, he connects with bloodshed and with the HL conception of the land, according to which sexual misconduct “pollutes the land.” The bringing together of these separate metaphorical realms underscores the severity of the peopleʼs sin and its inescapable consequences. The freedom in using legal terminology appears, furthermore, in the way that both meanings of ‫ חנף‬are activated in Jer 3:1–5.58 The verb is explicitly used in vv. 1–2 in reference to sexual sins as sources of pollution. Although ‫חנף‬ itself does not appear in vv. 4–5, its second meaning, “to deceive, to behave hypocritically,” is suggested by the words of the people quoted in these verses, which indeed give the taste of deception and hypocrisy on the part of those whose deeds have polluted the land. Thus I cautiously suggest that this double sense of ‫ חנף‬functions as an additional linguistic anchor for Jeremiahʼs choice of the Priestly legal phrase in contrast to the Deuteronomic legal phrasing of Deut 24:1–4.59 Hence, the Deuteronomic threat ‫ולא תחטיא את הארץ‬, the double meaning of ‫חנף‬, and especially its Priestly connotations in relation to ‫החניף) ארץ‬/( ‫חנף‬, serve as the linguistic pivots between the Deuteronomic law and the Priestly contexts. These clear terminological connections show Jeremiahʼs knowledge of P phrases, which he reframes in a legal context of sexual sin that seems much closer conceptually to the HL view of the land (Lev 18 and 20). Thus, the prophet freely adopts and adapts both P and HL phraseology and conceptions to characterize the crisis in the God-people relationship in the metaphorical framework of marriage and divorce. 57  I am grateful to Dr. Ruth Clements for pointing this out to me. It is questionable, at best, whether the vague accusation ‫ מצא בה ערות דבר‬in Deut 24 stands for any kind of sexual misbehavior; within the biblical framework, adultery would have been punished by execution, not divorce (Lev 20:10; Deut 22:22, 23–27). According to Deuteronomy, then, abomination results from remarriage following a legally permitted second sexual partner. For early and modern interpretations of ‫ערות דבר‬, see m. Gittin 9:10, and Tigay, Deuteronomy (see n. 9), 221–222. 58  K. S eybold , “‫חנף‬,” TDOT 5:36–44, distinguished two contexts that determine the meanings of ‫חנף‬: with ‫ ארץ‬as subject or object, the verb denotes “to pollute” and, figuratively, “to desecrate” (as in Ps 106:38); with human subjects, the verb designates antisocial behavior characterized by hypocrisy and deception (as in Prov 11:9). Indeed, Jeremiah seems to play on both senses in other contexts as well; see Jer 23:13–15, where ‫ חנפה‬is the inclusive term for the prophetsʼ sins, including ‫ נאף‬and ‫ ;הלך בשקר‬Jeremiah may even play on the alliteration of ‫ חנף‬and ‫( נאף‬see Jer 3:9; 23:10, 14). 59  Double meanings and wordplays as a central interpretive technique occur, for instance, with the use of ‫קדש‬, ‫אכל‬, and ‫( אשם‬Jer 2:3), as discussed above; ‫( השבע לשקר‬in 7:9); and ‫ברית‬ and ‫( קרא דרור‬34:8–22).

Compositional Harmonization

933

Nevertheless, we should not blur the differences between Jeremiahʼs uses of the two bodies of legal material. Although Jeremiah quotes the language of Deut 24:1–4, to the extent that we can safely assume that he refers to this specific Deuteronomic law of divorce in its present literary form, there is no way to presume that the prophet knows or uses the Priestly text of Num 35 or the HL passages of Lev 18 and 20. What appears very plainly throughout is the prophetʼs freedom in handling different legal traditions in their exact terminology and conceptions, which are for him as clay in the potterʼs hand. 3 Harmonization and Collation of P (or HL) and D: Inheritance and Possession (Jer 49:1–2) The prophecy against Ammon in Jer 49 opens with a three-part rhetorical question, a structure that the book of Jeremiah often applies (see, for instance, Jer 2:14; 8:19, 22).60 The two distinct meanings of ‫ירש‬, “take possession of” and “inherit,” which occur in altogether different contexts within P (or HL) and D materials,61 allow the prophet to construct a semantic pun in a Janus parallelism.62 60  For the three-part rhetorical question in Jeremiah, see W. L. H olladay , “The SoCalled ‘Deuteronomic Gloss’ in Jer. VIII 19b,” VT 12 (1962), 494–498, who considered this structure to be a genuine Jeremianic stylistic feature, whereas B. O. L ong , “Two Question and Answer Schemata in the Prophets,” JBL 90 (1971), 129–139, considered it typical of the editorial strand of the book; see further M. Held, “Rhetorical Questions in Ugaritic and Biblical Hebrew,” ErIsr 9 (1969), 71*–79*, and Y. Avishur, “The Pattern ‫ מדוע‬. . . ‫ אם‬. . . ‫ה‬ in the Book of Jeremiah,” Beit Mikra 16 (1971), 152–170 (Hebrew). For Janus parallelism, see S. M. Paul , “Polysemous Pivotal Punctuation: More Janus Double Entendres,” in Text, Temples, and Traditions: A Tribute to Menahem Haran (ed. M. V. Fox et al.; Winona Lake, IN: Eisenbrauns, 1996), 369–374. 61  Scholars of Numbers debate whether Num 27:1–11 should be categorized as P or HL. I. Knohl, The Sanctuary of Silence: A Study of the Priestly Strata in the Pentateuch (Minneapolis: Fortress, 1995), 95, argued that it represented a late strand within the Holiness School on the basis of the following arguments: (1) the mention of ‫( עדת קרח‬Num 27:3); (2) the expression ‫לשארו הקרוב אליו‬, which otherwise appears only in Lev 21:2; note that all other occurrences of ‫ שאר‬in HL likewise occur in Leviticus (Lev 18:6, 12, 13, 17; 20:18; 25:49); and (3) the expression ‫( חקת משפט‬in Num 27:11), which appears otherwise only in Num 35:30, considered by Knohl to be another HL passage. I would call attention to the fact that none of this characteristic HL vocabulary appears in Num 27:8–11a, which constitute the essential legal unit and which thus may belong to an older strand. 62  LXX did not translate the wordplay on ‫ ירש‬but used παραλαμβάνω (“to receive from another an office, thus a successive position,” “to take up, catch up”; LSJ 2:1315) as the equivalent throughout. The closing phrase was thus translated καὶ παραλήμψεται Ισραηλ τὴν ἀρχὴν αὐτοῦ, “and Israel will inherit its dominion”; see G. A. Walser, Jeremiah (Septuagint Commentary Series; Leiden: Brill, 2012), 120–121. The LXX reading seems to be behind the interpretations of P. V olz , Jeremia (KAT; Leipzig: Scholl, 1928), 414–415, and F. N orth , “The Oracle against the Ammonites in Jeremiah 49:1–6,” JBL 65 (1946), 37–43, both of whom translated ‫ ירש‬as “inherit” throughout and had no sensitivity to the importance of its double meaning in this context; see the discussion below. But παραλαμβάνω is clearly not

934

Dalit Rom-Shiloni

Hence, ‫ ירש‬is the core of this entire prophetic passage.63 Jeremiah 49:1–2 1 Concerning the Ammonites. Thus said the LORD: Has Israel no sons, has he no heir? Then why has Milcom dispossessed Gad, and why have his people settled in Gadʼs towns (‫הבנים אין לישראל אם יורש אין‬ ‫?)לו מדוע ירש מלכם את גד ועמו בעריו ישב‬ 2 Assuredly, days are coming – declares the LORD – when I will sound the alarm of war against Rabbah of the Ammonites; it shall become a desolate mound, and its villages shall be set on fire. And Israel shall dispossess those who dispossessed him (‫)וירש ישראל את ירשיו‬ – said the LORD.

Numbers 27:8–11 8 “Further, speak to the Israelite people as follows: ‘If a man dies without leaving a son (‫)ובן אין לו‬, you shall transfer his property to his daughter. 9 If he has no daughter, you shall assign his property to his brothers (‫)ונתתם את נחלתו‬. 10 If he has no brothers, you shall assign his property to his fatherʼs brothers (‫ונתתם‬ ‫)את נחלתו‬. 11 If his father had no brothers, you shall assign his property (‫ )ונתתם את נחלתו‬to his nearest relative in his own clan, and he shall inherit it (‫)וירש אתה‬.’ This shall be the law of procedure for the Israelites, in accordance with the LORDʼs command to Moses.” Deuteronomy 11:31 For you are about to cross the Jordan to enter and possess the land (‫לבא לרשת‬ ‫ )את הארץ‬that the LORD your God is assigning to you. When you have occupied it and are settled in it (‫וירשתם אתה וישבתם‬ ‫)בה‬.

The first two components of the rhetorical question, which introduce the question ad absurdum, use ‫ ירש‬in the sense of “legitimate inheritance of a property by the nearest kin,” a meaning known from the P (or HL) inheritance laws; in both cases, the usage evokes the legal situation of death in the absence of heirs (‫ נתן נחלה‬/ ‫העביר‬, cf. Num 27:8–11).64 The third component of the rhetorical the regular LXX equivalent for ‫ירש‬, which is either κληρονομέω, “to inherit” (LSJ 1:959) or κατακληρουχέω, “receive as oneʼs portion, esp. of a conquered country, divide among themselves, portion out,” (LSJ 1:894). For examples of the first, see, e.g., Gen 15:3; 22:17; and otherwise in Genesis; it is also the translational equivalent in Lev 20:24 and in the law of inheritance at Lev 27:11. Note also Deut 1:8, 21, 39, etc., representing the national sphere and including the frequent phrase ‫וירשתם את הארץ‬, translated regularly by κληρονομήσητε τὴν γῆν (Deut 4:1, etc.). For an example of the second, see Num 13:30 (but note also 14:24, where ‫ יורשנה‬is translated as κληρονομήσει αὐτήν). The use of παραλαμβάνω in Jer 49:1–2 and 32:7 in reference to inheritance of land is exceptional and forms a distinct contrast with its usual usage as an equivalent of ‫( לקח‬with human objects), as in Gen 22:3; Josh 4:2. 63  G. Fischer, Jeremiah 26–52 (HThKAT; Freiburg: Herder, 2005), 536–537, noted that the four occurrences of ‫ ירש‬in vv. 1–2 are the center of this prophecy, but he did not tie their occurrence here to any specific pentateuchal references. 64  ‫( העביר נחלה‬Num 27:7, 8) is a hapax legomenon, and ‫ נתן נחלה‬occurs in a non-P text (Num 16:14) that refers to giving the land to the people of Israel. In Deuteronomy, the repeated

Compositional Harmonization

935

question uses ‫ ירש‬in its more common meaning, which prevails in D: “to take possession of a territory by force/in war.”65 In fact, this last strophe constructs ‫ ירש‬as part of the pair ‫ ירש‬and ‫( ישב‬as occurs in Deut 2:21, 22; 11:31; 12:29; 17:14; 19:1; 26:1; Josh 21:43). Though this pairing is a hapax legomenon in the book of Jeremiah, its usage in 49:1–6 agrees with its usage in D, where it legitimizes ownership of a property obtained by means other than ordinary legal inheritance.66 In this Janus parallelism, ‫ מדוע ירש מלכם את גד‬serves as the pivot between the inheritance law, invoked in the two prior questions, and the context of military conquest and occupation evoked by the following phrase, ‫ישב בעריו‬, which clearly casts ‫ ירש‬as the consequence of military conquest.67 This double phrase is ‫( הארץ אשר יהוה אלהיך נתן לך נחלה‬Deut 4:21; 12:9; 15:4; 19:10, 14; 20:16; 21:23; 24:4; 25:19; 26:1; see also ‫ לתת לך את ארצם נחלה‬in 4:38); the theme is then developed in Josh 13:8, 14, 33; 14:3, 13; 17:14; 19:49; 21:3. But P has two specific contexts in which this phrase is used to refer to inheritance within kinship frameworks: one concerns the Levites (Num 18:21, 24; 26:54, 62; 35:2, 8); the other concerns the legal question about inheritance posed by the daughters of Zelophehad (Num 27:7, 9, 10, 11; 36:2; Josh 17:4). This latter notion of the legitimate transfer of land to another person is behind Nabothʼs refusal to give up his land for Ahab (1 Kgs 21:3, 4). 65  In all of its occurrences in Deuteronomy, ‫ ירש‬has the meaning “to take possession of a land which is not legally permitted”; see N. L ohfink , “‫ ירש‬yarash,” TDOT 6:368–396 (quotation from 378), and M. Weinfeld , Deuteronomy and the Deuteronomic School (2nd ed.; Winona Lake, IN: Eisenbrauns, 1992), 313–316. This meaning, however, is not confined to D. It occurs in non-P stories in Genesis–Numbers as well (see, e.g., Num 13:30; 21:24, 32, 35; 32:21; 33:52–53, 55 and also the HL passage, Lev 20:24). The opposite meaning of ‫ירש‬, “inherit,” occurs only in two HL passages: Lev 25:46, which does not refer to land yet uses the three terms (‫אחז‬, ‫ירש‬, ‫ )נחל‬transferred into the sphere of slavery, and Num 27:11; for the HL origin of the Zelophehad passages, Num 27:1–11; 36:1–12, see Knohl, Sanctuary of Silence (see n. 61), 95. There is an important distinction between ‫ נחל‬and ‫ירש‬: the former occurs in family transactions within discussions of inheritance laws (Num 18:23–24; Deut 21:16; and Judg 11:2), whereas the latter is rarely used in this context (Gen 15:3, 4; Num 27:11; 36:8). According to E. Lipiński, “‫ נחל‬nahal,” TDOT 9:320–321, the two verbs (‫ נחל‬and ‫ )ירש‬become synonymous in meaning within literature of the sixth century BCE; this takes place through the Aramaic influence of ‫ירת‬, which is the equivalent of the Hebrew ‫( נחל‬see Isa 57:13; Ps 69:36; and note ‫ ירש ארץ‬in Ps 37:11, 22, 28, 34). See also Lohfink, “‫ ירש‬yarash,” 384. 66  The pair ‫ ירש‬/ ‫ ישב‬also occurs with ‫ ירש‬in hiphil, as in Num 33:53 (‫והורשתם את הארץ‬ ‫וישבתם בה‬, ‫)כי לכם נתתי את הארץ לרשת אתה‬. Parallelism of ‫ ירש‬/ ‫ישב‬, where the two verbs refer to the same agent, occurs also in Isa 54:3 and in Ps 60:36. 67  Most commentators, as well as the Vulgate and Peshitta, have followed the LXX reading of this phrase, taking ‫ מלכם‬as a proper name: διὰ τί παρέλαβεν Μελχομ τὸν Γαδ, “Why has Milkom seized Gad?” and have found it to be a better text. See, e.g., P. V olz , Jeremia (see n. 62), 414–415; W. R udolph , Jeremia (HAT; Tübingen: Mohr Siebeck, 1947), 286; Holladay, Jeremiah 2 (see n. 51), 366, 368; W. McKane, Jeremiah XXVI–LXII (ICC; Edin­ burgh: T&T Clark, 1986), 1204; and J. R. Lundbom, Jeremiah 37–52 (AB 21C; New York: Doubleday, 2004), 315; and note Rashi on this verse, who glosses it as “the Ammonite god who is called ‘king.’ ” Compare, however, to Fischer, Jeremiah 26–52 (see n. 63), 536–537, who preferred MT because of the phrasing in v. 3 (“their king, his prophets, and his priests”);

936

Dalit Rom-Shiloni

meaning is thus constructed by way of the prophetʼs acquaintance with these two phraseological and thematic arenas that the prophet brings together (i.e., harmonizes) in order to craft his prophecy against Ammon. With this possibility in mind, it seems plausible to suggest that ‫וירש ישראל את‬ ‫ ירשיו‬puns upon the two meanings and collates the traditions of P and D into a single phrase in which the two contexts are genuinely intertwined, passing judgment on Ammon in the form of a lex talionis.68 Israel will once again become the possessor of the land, even its legitimate inheritor, and Ammon the dispossessed. Commentators have found it difficult to contextualize this prophecy due to the fierce tension between the prophetʼs message of judgment upon Israel on the eve of destruction (as within Jer 32:16–25) and v. 2bʼs prospect of salvation and restored hegemony over Transjordan.69 Recognizing the wordplay and its implications absolves us from the search for a historical context for the prophecy. We may understand it as a hopeful message for the future, driven by the double meaning of ‫ ירש‬and the two distinct pentateuchal contexts served by that term. This example thus shows that the prophet (or is it only the book?) is familiar with the terminology of both P and D. The prophecy is sensitive enough to play brilliantly with the two meanings of ‫ירש‬, taking into consideration the distinctive contexts in which they each function. However, there seem to be no grounds (and no need) for arguing that the prophecy here alludes to specific P or D texts. The prophecy simply utilizes individual terms and the conceptions that stand behind them, which the prophet finds to be of relevance to his current message, the proclamation against Ammon.

see also Amos 1:15, which McKane suggested had influenced both Jer 48:7b and the current passage (McKane, Jeremiah XXVI–LXII, 1204). 68  So Lundbom, Jeremiah 37–52 (see n. 67), 317. Holladay, Jeremiah 2 (see n. 51), 366, pointed out that the wordplay is similar to that in 30:16: ‫ והיו שאסיך‬. . . ‫לכן כל אכליך יאכלו‬ ‫“( למשסה וכל בזזיך אתן לבז‬Assuredly, all who wanted to devour you shall be devoured. [. . .] Those who despoiled you shall be despoiled, and all who pillaged you I will give up to pillage”). But LXX entirely misses this pun, translating all occurrences of the Hebrew term with a single Greek verb; see n. 62 above. 69  Qimhi in fact explained v. 2b as predicting the distant future. Modern commentators have considered non-Jeremian the hopeful prediction that Israel will take back dominion from Ammon; the two main arguments are (1) that the passage contradicts Jer 27:5–8 and (2) that it conflicts with the catastrophe the prophet has predicted for the last years of Judah. McKane, Jeremiah XXVI–LII (see n. 67), 1205, accepted the message of restoration but gave up Jeremianic authorship, arguing that “it is almost certainly a mistake to associate it with the prophet Jeremiah” (1212); rather, the prophecy has the quality of “the unreality of a dream” (1212). Compare to Lundbom, Jeremiah 37–52 (see n. 67), 315, 318, who suggested 594 BCE as a possible date for Jer 49:1–2; i.e., a time when Jeremiah still prophesied restoration for Israel (see Jer 27).

Compositional Harmonization

937

IV. Harmonization as a Literary Phenomenon in the Book of Jeremiah: Conclusions and Further Thoughts On the basis of the preceding analysis of these three poetic passages, it is now possible to ask more generally what characterizes the literary phenomenon of harmonization within such passages, which are considered to be among the early layers of the book of Jeremiah. As we have seen, harmonization operates in these three passages in different ways and for different rhetorical purposes. Jeremiah 2:3 unites the D conception of divine protection from Israelʼs enemies with the HL restrictions on eating sacred donations in order to present a forceful statement about Israelʼs holy status and its political consequences; in Jer 3:1–5, the invocation of ‫החניף את ה)ארץ‬/( ‫ חנף‬brings together the marital metaphor and the law of divorce (D) with the Priestly conception of sexual sin and defilement of the land (P and HL) to enforce the notion that the covenant relationship has been terminated and is beyond repair; Jer 49:1–2 collates the P/HL familial law of inheritance with the D conception of possession of land through conquest, drawn from the national sphere, in order to proclaim judgment upon Ammon, on the one hand, and to express hope for the repossession of the land by Israel, on the other. Notwithstanding the differences between these examples, we can draw from them seven shared characteristics of the use and function of harmonization in Jeremiah: First, each of these three poetic passages brings together diverse intertextual references from the two distinct strands of pentateuchal materials, P/HL and D. The echoes or the (literary) allusions represent unique or rare linguistic phenomena in Jeremiah, which are clearly characteristic of either Deuteronomic or Priestly (P/HL) idiom. The collation of the Deuteronomic and the Priestly (P/ HL) references shows the ease and readiness with which the prophecy draws on both, and the outcome is a tapestry centered on a major literary allusion coupled with minor echoes to either D or P/HL, or even to both. Second, harmonization is built upon lexical and thematic associations that lead the prophet to activate a similar word or a theme shared between the disparate pentateuchal traditions. The three examples discussed here suggest that harmonizations are based on a single semantic anchor (‫קדש‬, ‫ חנף‬/ ‫טמא‬, and ‫ ;)ירש‬the prophecy draws on the diverse meanings of the anchor term within the disparate contexts in which it appears in each of the pentateuchal traditions.70 Third, harmonization in these three passages is intentional and thoughtful, and we can trace the thematic lines used to tie the different pentateuchal tra-

 This, however, might not be an absolute requirement, as there are other examples in which such an anchor term does not appear; instead, a thematic connection draws the references together; see, e.g., Jer 32:6–15. 70

938

Dalit Rom-Shiloni

ditions together.71 These examples illustrate how the prophet (or the prophecy) combines otherwise distinct theological conceptions in such a way that each contributes its own strength to the harmonic literary structure. Fourth, the three passages show a hierarchy in the use of the invoked materials. In each passage, one set of pentateuchal traditions serves as the governing reference and has thematic (and lexical) dominance, while other traditions (from the same or different sources) may be echoed and at times contribute terminology as well. In these three cases, the D references dominate and are used to establish the primary allusion, either through citation of a central phrase (2:3 is based on Deut 7:6; ‫ הוריש את הארץ‬/ ‫ ירש‬in 49:1–2 derives from Deuteronomic idiom) or paraphrase of a Deuteronomic law (Jer 3:1 follows Deut 24:1–4). Fifth, the P/HL references in these three examples are tied into their new governing contexts to provide an additional element, a surprising twist, an extra forceful restatement of the theme. From the perspective of intertextuality, however, these references appear only as echoes of P/HL terminology, not as allusions to specific literary passages or laws.72 Sixth, in two of the examples, the book of Jeremiah creates new and unique phrases, true collations of D and P/HL phrases that serve as the core, or “punch line,” of his message, respectively: ‫( קדש ישראל ליהוה‬Jer 2:3) and ‫וירש ישראל‬ ‫( את ירשיו‬Jer 49:2). Finally, though all three passages contain references to pentateuchal legal traditions, these compositional harmonizations differ from their counterparts in later historiography, particularly in their function. The prophecies join P/HL and D phrases, themes, and conceptions in combinations conditioned not by legal or literary questions but solely by the rhetorical needs of his prophecy, thereby transforming them. Two crucial features of this rhetorical/literary technique within the book of Jeremiah (and prophecy in general) deserve special attention: the thoughtful 71  Other combinations of pentateuchal references in Jeremiah appear likewise to represent a calculated, intentional, and thoughtful procedure (see, e.g., Jer 2:20–25; 32:6–15; 34:8–22); but I concede that, in quite a few examples, the reasons behind the fusions are barely apparent. It may be that such harmonizations are not always burdened by specific linguistic or theological intent but reflect only the ability to pun across diverse traditions, an ability that arises from the prophetʼs command of diverse pentateuchal materials. This question awaits more detailed study. Referring to scribal harmonizations in both parallel biblical traditions and in extrabiblical texts, Carr cautiously concluded that “it often is impossible to determine whether a given harmonizing plus is the result of intentional or unintentional scribal coordination” (Carr, Formation of the Hebrew Bible [see n. 6], 98). 72  The categorization of a reference as an echo or as a literary allusion is determined by considering the accumulation of the references in one prophetic context and the function of each presumed reference in that new context. An example that might suggest a more extensive acquaintance with a Priestly text is Jer 2:20–25, which evokes the trial of the sotah in Num 5:11–31; see Rom-Shiloni, “How Can You Say, ‘I Am Not Defiled’ (Jer 2:20–25)?” (see n. 2).

Compositional Harmonization

939

intentionality behind the harmonizations and the prophetʼs freedom in creating harmonizations in what appear oftentimes to be virtuosic ways. The prophet clearly feels completely free to create these wordplays and thematic combinations purely to suit the context of his prophecy.73 Jeremiah utilizes these national narrative and legal traditions interpretively, in the most creative ways – making use of analogy, expansion, transformation, reversal, and many other techniques – to tap this treasury of traditions for his own message.

V. The Book of Jeremiah and the Torah The examples discussed are but select illustrations of a more widespread phenomenon that cuts across both poetry and prose passages in Jeremiah. The discussion here has aimed at showing that the prophet has mastery of both Deuteronomic and Priestly (P/HL) traditions, to the extent that he is able to utilize actual lexemes as well as broader conceptions, and that, in using them, the prophet pursues the double (and not easily cooperative) aims of preserving authoritative earlier traditions while creating new prophetic messages. At the compositional level, the harmonization of pentateuchal references plays an important role as a literary component of prophecy at a time when pentateuchal materials had not yet gained canonical status, though they had already gained a measure of authority. Michael Fishbane has labeled this stage the “proto-canonical level” of biblical composition.74 David Carr has suggested that harmonization illustrates the coordination of different traditions based at times on memorization and on oral performance, as part of an oral–written transmission of traditions and texts.75 Harmonization thus finds a place among the very common rhetorical and literary practices by which traditions gradually develop, are treasured, and are transmitted, whether orally or in written forms.76 As has been recognized, harmonizations based on memorized material may be

73  The prophet fuses diverse allusions and echoes of different sources, placed side by side or even interwoven, so as to create new phrases and concepts. See Fischer, Jeremia (see n. 27), 131–148, whose recognition of Jeremiahʼs wealth of literary sources caused him to argue for a relatively late evolution of the book, long after the prophetʼs own lifetime. 74  M. F ishbane , “Types of Biblical Intertextuality,” Congress Volume: Oslo, 1998 (ed. A. Lemaire and M. Sæbø; VTSup 80; Leiden: Brill, 2000), 39–44, used several such terms to subcategorize the biblical data; see “canon before the canon” or “canon within the canon” 75  Carr, Formation of the Hebrew Bible (see n. 6), 90–98. 76  On harmonization within literary sources, see Y. Z akovitch , “Assimilation in Biblical Narratives,” in Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 175–196. Though not all assimilations include harmonization, Zakovitch noted that 1 Chr 21:22 was “conforming the terminology” of 2 Sam 24:24 to that of Gen 23:9 (181). See Carr, Formation of the Hebrew Bible (see n. 6), 90, 96.

940

Dalit Rom-Shiloni

either intentional or unintentional.77 Intentional harmonizations demonstrate the creative role of authors who, working with earlier traditions of various origins, produce new texts with harmonistic elements that clearly add their own interpretive dimensions while forwarding the later authorʼs own rhetorical agenda.78 I suggest here that the methodological framework of recognizing innerbiblical allusion and exegesis can contribute to the investigation of this phenomenon by exposing the deliberately allusive and interpretive roles of harmonization at the earliest stages of literary composition. It is more difficult to answer the question, What might have been the shape of the pentateuchal materials at Jeremiahʼs disposal?79 The discussion here substantiates the idea that, by the early sixth century BCE, Deuteronomic and Priestly (P/HL) traditions were available either orally or in some written form and were part of the prophetʼs repertoire; presumably these materials were also familiar to at least parts of his audience (as assumedly also to his tradents). But the examples may indicate only knowledge of specific terminology and concepts, and we still lack evidence for Jeremiahʼs knowledge of actual texts and contexts.80 Although the present study has focused on Jeremiah, Risa Levitt Kohn has identified a similar compositional technique used by Ezekiel, his younger, contemporary colleague.81 The two prophets (and so we may assume other prophets of their time) could indeed create “personal syntheses” out of the Deuteronomic  See Tov, “Nature and Background of Harmonizations” (see n. 3), 5.  In the arena of redaction history, C arr , Formation of the Hebrew Bible (see n. 6), 132–137, put forth the scholarly definition of post-Priestly texts as those that mix Priestly and non-Priestly materials. Accordingly, the addition of Priestly elements to non-Priestly texts is seen as part of “a gradual process of scribal assimilation” (133), i.e., a secondary interpolative intervention into the texts. Carr held that HL materials exhibit the same coordinative intention he identified in what he termed “post-D Hexateuch materials,” and thus suggested that the two reflect “a similar exilic milieu” of the sixth century BCE (298–303, quote on 301; see also 302–303). 79  Sommer, A Prophet Reads Scripture (see n. 25), 132–151, pointed out that Second Isaiah has access to pentateuchal materials “in some form,” just as the prophet could utilize materials from Jeremiah, Psalms, and even the Isaianic traditions (132 and n. 1). Sommer, however, did not deal with the phenomenon of harmonization of pentateuchal traditions that is discussed here. 80  I am hesitantly willing to suggest that the prophet had knowledge of a very limited number of either Deuteronomic or Priestly texts. Such might be the case, for example, with Jer 3:1 (reflecting Deut 24:1–4); Jer 34:12 (paraphrasing Deut 15:1, 12, which might hint at a knowledge of a wider chunk within Deut 15); and Jer 2:20–25, which I have recently suggested might reflect knowledge of P traditions surrounding the trial of the sotah (Num 5:11–31); see Rom-Shiloni, “How Can You Say, ‘I Am Not Defiled?’ (Jer 2:20–25)” (see n. 2). 81  R. L evitt K ohn , A New Heart and a New Soul: Ezekiel, the Exile and the Torah (JSOTSup 358; New York: Sheffield Academic Press, 2002), 96–104; on the fusion of D and P conceptions beyond Ezek 20, see 106–118. 77

78

Compositional Harmonization

941

and Priestly (P/HL) materials with which they were acquainted.82 But, as the present study shows, and in distinction from Levitt Kohn, we may understand this phenomenon as more broadly attested across the Judean and Babylonian communities of the early sixth century BCE, not confined to either Ezekiel or the Babylonian exiles. Furthermore, there are differences across the board between the prophetic usage of pentateuchal materials and the practices of the pentateuchal redactors, the Priestly redactors, and the authors of Chronicles and Ezra–Nehemiah. One significant difference is apparent between prophecy and late historiography: the prophetic literature lacks any explicit reference to a “written Torah” or citation formulae such as ‫ככתוב‬.83 Prophets, Jeremiah among them, do not quote directly, but rather allude to, echo, or show the influence of pentateuchal materials. Hence, the extent to which sixth-century prophets in general and Jeremiah in particular were exposed to diverse pentateuchal traditions may only be located within conventions that reveal implicit usage of such traditions, and this has been the focus of this study. It seems, then, that the late seventh and early sixth centuries BCE, first in Judah and then in Babylon, represented a period when pentateuchal materials in oral and/or written traditional form were already at hand, already authoritative, and yet not off-limits for creative homiletic exegesis in prophetic contexts.84 My close reading of these passages has allowed us to see something of the vast spectrum of this prophetic phenomenon, which only indirectly reveals the prophetʼs exposure to diverse pentateuchal materials and his freedom to use them. Delving deep into these phrases and their invoked traditions is, to my mind, one of the keys to deciphering the prophecies of Jeremiah in their fullest strength.

 R. R. W ilson , Prophecy and Society in Ancient Israel (Philadelphia: Fortress, 1980), 284. References to legal terminology within the book of Amos (see ‫בגדים חבלים‬, Amos 2:7 and Deut 24:17; cf. Amos 4:6–11 and the curses of Deut 28) have already been set out by Fishbane, Biblical Interpretation (see n. 13), 295–296. I would, thus, advise reconsidering our perceptions of these literary connections within the prophetic corpus and allowing for a more flexible usage of specific terms and legal contexts by writers of the eighth and, even more so, the late seventh and early sixth centuries; but it is not necessary to posit formal and written legal texts or “codes” of laws as the basis for such compositional practices at these early points in time. 83  See Japhet, “Law and ‘the Law’ in Ezra–Nehemiah” (see n. 13), 99–115. 84  It may thus be assumed (though this deserves further study) that the same phenomenon may be found in sixth-century prophecies among the Twelve. Levitt Kohn, New Heart (see n. 81), 117, n. 57, discusses Second Isaiah in this connection, though she did not specify examples. 82

Patterns of Descriptive Curse Formulae in the Hebrew Bible, with Special Attention to Leviticus 26 and Amos 4:6–12 John Kessler Curses, imprecations, or maledictions1 occupy a highly significant place in both ancient Near Eastern texts and the Hebrew Bible.2 The present study proceeds from the conviction that – although much profit can be had from analyzing these curse materials in terms of the specific content of the curses, the grammatical formulations used to express them, or their relationship to broader ancient Near Eastern maledictions – a further level of consideration is required to deal adequately with these materials in the Hebrew Bible. The biblical curse formulations must be understood in terms of their significance within the broader literary contexts in which they appear, most especially in terms of the ideological assumptions within those larger contexts, rather than in abstraction from them. The present study will thus seek to demonstrate that the biblical curse materials, when analyzed within their broader literary and ideological (or theological) contexts, fall into certain specific patterns. Moreover, even within these specific patterns, diversity may appear. Individual texts may use these curse formulae in unique ways, reflecting the ideological underpinnings of the specific contexts in which they are found. This latter point will be demonstrated through a comparative analysis of Amos 4:6–12 (henceforth Amos 4)3 and Lev 26. Thanks are due to the numerous colleagues at the Jerusalem colloquium who offered their comments and suggestions on this paper and to Mark Boda for his insightful reflections on it. I also offer my very special thanks to Dalit Rom-Shiloni for her careful reading of the paper and several important suggestions. I also wish to express great appreciation to my research assistant, Mari Leesment, for her painstaking work on the editing and formatting of the text. 1  Although, strictly speaking, there may be certain differences in meaning between these English terms, I will use them interchangeably here. 2  For a general introduction to the subject, see J. S charbert , “‫אלה‬,” TDOT 1:261–266; idem , “‫ארר‬,” TDOT 1:405–418; R. P. G ordon , “Curse, Malediction,” NIDOTTE 4:491–493; W. R. Urbroch, “Blessing and Cursing,” ABD 1:755–759; W. Schottroff, Der altisraelitische Fluchspruch (Neukirchen-Vluyn: Neukirchener Verlag, 1969). 3  Although the curses are contained in vv. 6–11, they are closely related to the preceding and following sections (see below). I will include v. 12 in my discussion, since it is generally seen as reflecting the culmination of the curses in vv. 6–11. The issues surrounding Amos 4:13 cannot be dealt with in the present context.

944

John Kessler

At the outset, I wish to delimit the specific type of curse materials I will be examining. Numerous scholars have isolated various aspects of the biblical and ancient Near Eastern curse materials and used them as points of entry for their analyses. Some have focused on the various lexemes and formulae employed,4 others on the particular socioreligious matrices out of which such curses emerged and on the subsequent growth and development of the various curse formulae,5 and still others on the relationship between curses and oaths or curses and covenants. 6 Clearly a review and appraisal of this far-reaching work is beyond the scope of the present paper. My discussion here will be limited to one specific type or form of curse expression: curses that are accompanied by a description of the specific misfortunes that may be (or have been) visited upon an offender for the violation of the stipulations set forth by the speaker of the curse. For the purposes of the present study, I will designate these as “descriptive curse formulae.”7 They stand in contrast to other expressions of malediction in which the adverse effects or results of the curse are left unstated.8 In descriptive curse formulae, however, the con-

4  J. K. A itken , The Semantics of Blessing and Cursing in Ancient Hebrew (ANES 23; Leuven: Peeters, 2007); H. C. Brichto, The Problem of “Curse” in the Hebrew Bible (JBL Monograph Series 23; Philadelphia: Society of Biblical Literature and Exegesis, 1963). 5  J. S. Anderson, “The Social Function of Curses in the Hebrew Bible,” ZAW 110 (1998), 223–237; S. G evirtz , “West-Semitic Curses and the Problems of the Origins of Hebrew Law,” VT 11 (1961), 137–158; idem, “Curse,” IDB 1:749–750; J. Hempel, “Die israelitischen Anschauungen von Segen und Fluch im Lichte altorientalischer Parallelen,” ZDMG 69 (1915), 20–110; Schottroff, Der altisraelitische Fluchspruch (see n. 2). 6  K. B altzer , The Covenant Formulary in Old Testament, Jewish and Early Christian Writings (trans. D. E. Green; Philadelphia: Fortress, 1971); D. R. H illers , Treaty Curses and the Old Testament Prophets (BibOr 16; Rome: Pontifical Biblical Institute, 1964); G. E. Mendenhall, Law and Covenant in Israel and the Ancient Near East (Pittsburgh, PA: Biblical Colloquium, 1955); idem, “The Suzerainty Treaty Structure: Thirty Years Later,” in Religion and Law: Biblical-Judaic and Islamic Perspectives (ed. E. B. Firmage et al.; Winona Lake, IN: Eisenbrauns, 1990), 85–100. 7  Such maledictions were sometimes designated treaty curses. The term futility curse was seen to constitute one distinct type of treaty curse; see H illers , Treaty Curses (see n. 6), 28–29. However, as I will demonstrate below, though maledictions containing descriptions of the results of the curse do indeed appear in treaty texts, they also occur elsewhere. Thus, use of the term treaty already prejudices oneʼs conclusions regarding the nature and function of the maledictions involved. 8  This difference is not always considered in treatments of the curse material. It has, however, been noted by P. B uis , “Deutéronome 27:15–26: malédictions ou exigences de lʼalliance?,” VT 17 (1967), 478–479. Examples of curse formulae with no stated consequences occur with the passive participle ‫ ארור‬in Deut 27:15–26; 28:16–19; 1 Sam 14:21; Jer 11:3; 20:14–16; 48:10, Mal 1:14. The ‫ ארור‬form has been the object of significant investigation; see esp. B richto, Problem of “Curse” (see n. 4), 77–96; Schottroff, Der altisraelitische Fluchspruch (see n. 2), and the references to cursing in Gen 12:3; Num 22–24 (on which see

Patterns of Descriptive Curse Formulae in the Hebrew Bible

945

sequences arising from the malediction are specifically named and described.9 At times these formulae are very brief, involving a simple reference to one aspect or sphere of life experience.10 Most are far more detailed. Some appear in almost fixed forms, across various cultures and in varying literary contexts.11 The so-called futility curses of Amos 5:11; Mic 6:14–15; and Hag 1:5–11, with their parallels in ancient Near Eastern literature (see infra), are a case in point.12 Descriptive curse formulae are widely distributed throughout the Hebrew Bible. Within the prophetic literature they often appear in the context of oracles of doom.13 Very extensive curse lists are limited to relatively few passages, notably Deut 28:1–68 (ET 29:1) and Lev 26:1–39. Examples of more sizable lists in the Book of the Twelve include Amos 4:6–11; Mic 6:11–16; and Hag 1:3–11; 2:16–17.14 These curse lists are close in form to the maledictions in the epilogue to Hammurabiʼs Code (E19–32), Sefire 1A and 2A, the Vassal Treaties of Esarhaddon cols. 6 and 7, and the Tell Fekherye Inscription.15 Such Brichto, Problem of “Curse,” 99–100); Judg 5:23; 21:18; Job 3:1–9. The euphemistic use of ‫ ברך‬in Job 1:5, 11; 2:5, 9 should also be mentioned here. 9  Compare also the broader, more grammatically based discussion of curse formulae in S. H. B lank , “The Curse, Blasphemy, the Spell, and the Oath,” HUCA 23 (1951), 73–95. Although Blank does not make a formal distinction between instances where the effects of the curse are specifically mentioned as opposed to where they are not, he does note that, in certain contexts, full verbal expression of the results of a curse for oath violation are more frequently omitted than verbalized (90–92). He attributes this reticence to a belief in the self-fulfilling power of the spoken word. 10  So, e.g., Deut 28:16–19. 11  See the inventory in F. C. F ensham , “Common Trends in Curses of the Near Eastern Treaties and kudurru-Inscriptions Compared with Maledictions of Amos and Isaiah,” ZAW 75 (1963), 155–175. 12  See A. Abou-Assaf et al., La statue de Tell Fekherye et son inscription bilingue assyroaraméenne (Études assyriologiques; Paris: Éditions Recherche sur les civilisations, 1982); Hillers, Treaty Curses (see n. 6), 28–29. 13  See the discussion in Hillers, Treaty Curses (see n. 6), 77–78. 14  Other passages in the Book of the Twelve containing curse materials beyond single allusions to curses, yet not as extensive as the passages just cited, include Amos 7:17; 8:9–10; 9:2–4; Obad 15–16. See also Joel 1:4–12, 17–19, which contains extensive descriptions of curses already imposed. A full inventory of the curse materials in the prophets, not to mention the Hebrew Bible as a whole, would be beyond the scope of this essay. I have cited these examples from the Book of the Twelve owing to the fact that Amos 4:6–12 forms a major focus of the present study and because they are close in form and content to Lev 26, another key area of focus here. Many other relevant examples could be cited from prophetic materials outside the Book of the Twelve. Numerous specific examples from the Hebrew Bible will be given in the typology of curse patterns proposed below. 15  For the standard translation of these texts, see M. E. J. R ichardson , Hammurabiʼs Laws: Text, Translation, and Glossary (London: T&T Clark, 2004); D. J. W iseman (ed.), The Vassal-Treaties of Esarhaddon (London: British School of Archaeology in Iraq, 1958); J. A. F itzmyer , The Aramaic Inscriptions of Sefîre (BibOr 19; Rome: Pontificial Biblical Institute, 1967); A. L emaire and J.-M. D urand , Les inscriptions araméennes de Sfiré et

946

John Kessler

ancient Near Eastern materials thus provide a useful baseline for comparison with the biblical data. In the present paper, then, I will focus on the use of these descriptive curse formulae in the Hebrew Bible with special attention to Amos 4:6–12 and Lev 26.16 I will demonstrate that although these traditional descriptive curse formulae are very general, and may carry a great variety of connotations, it is only through careful attention to the broader ideology expressed in the contexts in which they are found, and the distinctive ways in which they are shaped, that one may draw judicious conclusions regarding their meaning and significance.17 Our study will proceed as follows. In an introductory section, I will survey some of the major themes in earlier studies of curses in the Hebrew Bible and ancient Near Eastern texts and suggest that the results of these studies demonstrate the need for a nuanced and textually integrated approach to these descriptive curse formulae. In the second section, I will set forth a preliminary typology of such formulae. In my third section, I will assess both the similarities and significant differences between Amos 4 and Lev 26, texts that I see as rooted in the same general type of descriptive curse formula. Then, in the fourth section, I will suggest possible implications of this comparison for two disputed issues: the literary relationship between Lev 26 and Amos 4 and the compositional history of Lev 26. In my fifth and final section, I will set forth my conclusions, and suggest some implications of this study, especially as it regards the relationship between prophetic and pentateuchal texts.

lʼAssyrie de Shamshi-ilu (Hautes études orientales; Geneva: Droz, 1984); Abou-Assaf et al., La statue de Tell Fekherye (see n. 12); F. I. Andersen and D. N. Freedman, “The Orthography of the Aramaic Portion of the Tell Fekherye Bilingual,” in Text and Context: Old Testament and Semitic Studies for F. C. Fensham (ed. W. Claasen; JSOTSup 48; Sheffield: JSOT Press, 1988), 9–49; J. C. G reenfield and A. S haffer , “Notes on the Curse Formulae of the Tell Fekherye Inscription,” RB 92 (1985), 46–59; E. L ipiński , “The Bilingual Inscription from Tell Fekherye,” in Studies in Aramaic Inscriptions and Onomastics (OLA 57; Leuven: Peeters, 1994), 19–81. 16  I will explain my reasons for comparing and contrasting these two texts below. 17  See the similar observations regarding the Gattung of penitential prayer in B. Becking, “Nehemiah 9 and the Problematic Concept of Context (Sitz im Leben),” in The Changing Face of Form Criticism in the Twenty-First Century (ed. M. A. Sweeney and E. Ben Zvi; Grand Rapids, MI: Eerdmans, 2003), 253–268. See also the insightful observations of K. R. Veenhof, review of Der altisraelitische Fluchspruch, by Willy Schottroff, VT 22 (1972), 375–382, esp. 379, who observes that the significance and implications of curses found in literary compositions may differ in form and content from the same curses found in an oral context. See also Aitken, Semantics (see n. 4), 4–5, who discusses the distinctive nature of the biblical materials (and the myriad meanings of the key lexemes for cursing) in literary texts, in contrast to inscriptions or graffiti.

Patterns of Descriptive Curse Formulae in the Hebrew Bible

947

1 Earlier Study of the Biblical Curse Materials Study of the curse material in the Hebrew Bible has followed several lines over the last hundred years. The early twentieth century witnessed significant discussion of curses and cursing in light of various emerging archaeological discoveries and in connection with the study of oaths and blessing both in Israel and the ancient Near East.18 By the mid-twentieth century, much scholarly discussion had become focused upon the question of the efficacy of the spoken word (apart from the action of any given deity) in blessing and cursing. More specifically, scholars wrestled with the question whether oaths, blessing, and cursing reflected an earlier stage of Israelite religion in which words had a quasi-magical power. The studies of S. Gevirtz, and S. H. Blank, (vigourously contested by H. C. Brichto) are early illustrations of this.19 This discussion was taken up by A. Thiselton in the context of speech-act theory. Emphasis was shifted from the words themselves to the speaker(s) of such words of blessing and cursing and to their setting.20 The study of blessing and cursing was similarly a significant element in discussions of the development of Israelite religion, specifically questions regarding the difference between religion and magic and whether one could meaningfully differentiate between the two.21 Pursuing a different track, some scholars examined the similarities and differences between the biblical curses and other ancient Near Eastern materials.22 Others sought to distinguish between East- and West-Semitic curse formulations.23 In a related development, biblical curses were frequently seen as indicating that certain biblical texts were modeled on various ancient Near Eastern covenant forms, and these texts were then dated or redated according to the curse patterns within  Hempel, Die israelitischen Anschauungen (see n. 5); S. Mowinckel, Segen und Fluch in Israels Kult und Psalmdichtung (Kristiania: Dybwad, 1924); J. Pedersen, Der Eid bei den Semiten in seinem Verhältnis zu verwandten Erscheinungen sowie die Stellung des Eides im Islam (Strassburg: Trübner, 1914). 19  B lank , “Curse” (see n. 9); idem , “Some Observations Concerning Biblical Prayer,” HUCA 32 (1961), 75–90; Brichto, Problem of “Curse” (see n. 4); Gevirtz, “West-Semitic Curses” (see n. 5); idem , “Curse” (see n. 5). Brichto strongly challenged the assertion of Gevirtz and Blank that blessings and cursings carried with them a kind of automatic efficacy. 20  A. C. T hiselton , “The Supposed Power of Words in the Biblical Writings,” JTS 25 (1974), 283–299. See the excellent summary of this scholarly discussion in B. F. B atto , “Curse,” DDD 211–214. See also the analysis of the question from the perspective of text linguistics in Aitken, Semantics (see n. 4), 17–22. 21  See the extensive discussion in Aitken, Semantics (see n. 4), 5–22. 22  On the relationship of the curse material in these texts to the prophets, see F ensham , “Common Trends” (see n. 11); Hillers, Treaty Curses (see n. 6). See also S. M. Paul, Amos (Hermeneia; Minneapolis: Augsburg Fortress, 1991), 141–149, with bibliography and illustrations. 23  Gevirtz, “Curse” (see n. 5), 750; idem, “West-Semitic Curses” (see n. 5). See also the critique of Gevirtz in Hillers, Treaty Curses (see n. 6), 27, n. 44. 18

948

John Kessler

them.24 This period also witnessed several compilations of allusions to ancient Near Eastern curse materials in the Hebrew Bible.25 As well, some studies provided more form-critical analyses of the ancient Near Eastern curse materials and suggested implications for the study of Deut 28, Lev 26, and the Prophets.26 Many of these diverse issues continue to be discussed, and most are still in dispute. However, something of a broad consensus exists around two points. First, formally, although descriptive curses in the ancient Near East sometimes occur in relatively shorter lists (as in some of the kudurru inscriptions),27 over time these lists became much longer and the curses more elaborate.28 These curse lists likely served a rhetorical function: to terrorize the potential offender with all kinds of possible disasters drawn from every realm of life. As Nelson notes, “[D]eterring infractions was the primary goal. [. . .] The sequence of the individual items seems to be almost entirely arbitrary.” Moreover, the repetitions in the list were “intended to reinforce the rhetorical impact” of the curses.29 Second, on a functional level, such curses served as threats to ensure compliance with the wishes of a ruler or with the terms of a law or an agreement in the absence of any broader social or political mechanisms of enforcement. Hillers rightly compared the ancient treaty to “an elaborate promise, in which curses served to guarantee that the promise would be kept.”30 At times, however, some scholars drew more far-reaching conclusions regarding the exegetical sense of a given text, based largely on the presence of curse materials within it. Thus, once one had detected the presence of a descriptive curse formula within a text (especially a “futility curse”), the significance of the curse and the meaning of the text were seen as clear: the existence of a covenant was to be implied and curses were either dire warnings against covenant violation or evidence of such. However, covenant violation was often ill-defined. Which covenant was in view? And what were its demands and sanctions? Did the concept of covenant violation imply a minor infraction, the transgression of a single covenantal obligation, or persistent violation of the core duties of the covenant resulting in its complete abrogation?31  Mendenhall, Law and Covenant (see n. 6); Baltzer, Covenant Formulary (see n. 6).  F. C. Fensham, “Malediction and Benediction in Ancient Near Eastern Vassal-Treaties and the Old Testament,” ZAW 74 (1962), 1–9; idem, “Common Trends” (see n. 11), 12. 26  Hillers, Treaty Curses (see n. 6). 27  W. J. H inke , Selected Babylonian kudurru Inscriptions (SSS 14; Leiden: Brill, 1911); idem (ed.), Babylonian Boundary Stones and Memorial Tablets in the British Museum (London: British Museum, 1912); Hillers, Treaty Curses (see n. 6). 28  Hillers, Treaty Curses (see n. 6), 6. 29  R. D. Nelson, Deuteronomy: A Commentary (OTL; Louisville: Westminster John Knox, 2002), 327. 30  Hillers, Treaty Curses (see n. 6), 6. 31  On the various possible understandings of covenant violation and its effects, see, recently, S. M. Olyan, “The Status of Covenant during the Exile,” in Berührungspunkte: Studien zur 24

25

Patterns of Descriptive Curse Formulae in the Hebrew Bible

949

For example, speaking of the futility curses in Hag 1:3–11, David Petersen suggests, “[R]econstruction of the temple is treated as a covenant duty that, because it has not been accomplished, has brought on the futility curses of an abrogated covenant. [. . .] Not only are the people living an existence cursed because of the 587 disaster; their existence is also cursed because of their reaction to the result of that earlier cursing, the destruction of the temple.” 32 I believe Petersenʼs assessment to be inaccurate here and have argued to the contrary.33 Simply put, though it is certain that Hag 1:3–11 and 2:15–17 contain several verbal and thematic elements found in Deut 28 and Lev 26, can one move from this observation to the conclusion that the maledictions in Haggai imply the same covenantal arrangement and function in the same way as Deut 28 and Lev 26? This question is especially salient in that Deut 28 and Lev 26 clearly manifest points of striking discontinuity, especially with reference to the idea of covenant, and the specific covenants being referenced.34 What is more, temple building does not figure as a covenantal duty in either Deut 28 or Lev 26. How can one be sure that Haggai has subsumed temple reconstruction into the demands of the Sinai covenant?35 Only a careful reading of Haggai itself can determine the ideology that undergirds the descriptive curse formulae present in it. Numerous other examples of such an approach could be cited.36 There are three essential problems with such a broad and general approach. First, it does not take into account the varied ways in which these descriptive curse formulae function within the theological or ideological streams of tradition in the Hebrew Bible – that is, the differing ways in which these curse materials are employed to express the basis and dynamics of the divine-human relationSozial- und Religionsgeschichte Israels und seiner Umwelt; Festschrift R. Albertz (ed. R. Ebach et al.; Münster: Ugarit-Verlag, 2008), 333–344; J. Kessler, “Curse, Covenant, and Temple in the Book of Haggai,” in Covenant in the Persian Period: From Genesis to Chronicles (ed. R. J. Bautch and G. N. Knoppers; Winona Lake, IN: Eisenbrauns, 2015), 229–254. 32  D. L. Petersen, Haggai and Zechariah 1–8 (OTL; London: SCM, 1985), 50. 33  J. Kessler, “Curse, Covenant and Temple” (see n. 31). I will summarize my objection to his conclusion below. 34  These points will be taken up further below. 35  Note the development of my own thought on this point in J. K essler , The Book of Haggai: Prophecy and Society in Early Persian Yehud (VTSup 91; Leiden: Brill, 2002); idem, “Curse, Covenant and Temple” (see n. 31). 36  Examples of the notion that curse language automatically implies covenant are easy to come by. Other generalizations are more subtle. E. Gerstenberger, Leviticus: A Commentary (trans. D. W. Stott; OTL; Louisville: Westminster John Knox, 1996), 426, sees Deut 28 as “a sermon on the theme ‘God will punish the person who transgresses against the commandments.’ ” He then concludes that “Lev 26:14–33 also belongs, together with various discourses from Jeremiah and Ezekiel, to the genre of comminatory or admonitory sermon of the sort customarily used in the early Jewish communities for inculcating obedience to Yahweh.” This seems, to me, to level the differences between the various texts he cites and to subsume them under a single form, despite their distinctive use of the curse materials.

950

John Kessler

ship.37 Second, although it is widely recognized that various biblical texts use remarkably similar curse language, this homogenizing approach does not isolate and identify, on the micro level, one textʼs distinctive use of these curse formulae vis-à-vis other highly similar texts. Our study of Amos 4 in relationship to Lev 26 will examine this phenomenon in detail. Finally, as the curse typology to be presented below will demonstrate, the presence of curse language does not, in and of itself, presuppose covenant, nor may it be seen as evidence that a given text takes for granted the existence of either the Sinai or the ancestral covenants. In point of fact, at the most basic level, the presence of descriptive curse formulae in a text merely signifies that a speaker (within the world of the text) or writer (of the text) views certain real or potential actions negatively and invokes one or more specific disasters upon the offender(s) as deterrents or as punitive measures. Any inferences beyond this must be supported by arguments from the specific content and context of the curse.

2 A Typology of Descriptive Curse Formulae in the Hebrew Bible 2.1 Earlier Reflections: Two General Curse Patterns in the Hebrew Bible In earlier reflection on the curses in Hag 1:3–11 and 2:15–19 in the broader context of covenant in the Persian period, I suggested that there were two major curse patterns in the Hebrew Bible.38 The first of these might be called “violations of covenant.” Herein, a series of severe misfortunes is deployed for the violation of the core elements of the covenant in question. Such violations culminate in cataclysmic devastation and the destruction of the offender. Despite significant differences between them at a deeper level,39 Deut 28 and Lev 26 are biblical examples of this pattern. Reference to a covenant is explicitly stated or implicitly assumed (Lev 26:9, 15, 25, 42, 44–45; Deut 28:69 [ET 29:1]). Disobedience to the core demands of the covenant leads inexorably to ruin. 37  See O. H. S teck , “Theological Streams of Tradition,” in Tradition and Theology in the Old Testament (ed. D. A. Knight; Philadelphia: Fortress, 1977), 183–214; see also J. KESSLER, Old Testament Theology: Divine Call and Human Response (Waco, TX: Baylor University Press, 2013). 38  Kessler, “Curse, Covenant and Temple” (see n. 31). 39  On the various contrasts between Lev 26 and Deut 28, see, e.g., S. R. D river , Deuteronomy (ICC; New York: Scribnerʼs Sons, 1916), 304; J. Milgrom, Leviticus 23–27: A New Translation with Introduction and Commentary (AB 3B; New York: Doubleday, 2001), 2346–2348; G. Fischer, “A Need for Hope? A Comparison between the Dynamics in Leviticus 26 and Deuteronomy 28–30,” in Current Issues in Priestly and Related Literature: The Legacy of Jacob Milgrom and Beyond (ed. R. E. Gane and A. Taggar-Cohen; SBLRBS 82; Atlanta: Society of Biblical Literature, 2015), 369–385, esp. 380–385.

Patterns of Descriptive Curse Formulae in the Hebrew Bible

951

In a second, or more cultic, pattern, which may be designated “violations in covenant,” various misfortunes are visited upon an individual or community due to some specific legal or ritual violation. In such cases, there is no clear reference to violation of the core elements of a covenantal relationship (such as disobedience to Yahwehʼs “commandments, statutes and ordinances,” as is the case in Lev 26:15; Deut 5:31; 8:11; 26:17). Such texts often display only a very loose connection to the covenant idea. In these texts, one or several misfortunes are imposed, or blessings withheld, until such time as the offense is corrected. There is no sense of any progression of the maledictions toward ultimate disaster and obliteration. Examples of this pattern include Josh 7; 1 Sam 14:24–46; 2 Sam 21; and most especially 1 Kgs 8.40 Ongoing analysis convinced me that Hag 1:5–11 and 2:15–19 belonged to this second pattern. Thus, in contrast to Petersenʼs statement cited above, I maintained that, although on a form-critical level futility curses do indeed appear in Hag 1:5–11 and 2:15–19, one could not move from such a classification to the assumption that Hag 1:3–11 and 2:15–19 view the covenant between Yahweh and Israel as having been abrogated (so Petersen).41 The broader patterns of curse usage and the underlying ideological convictions of the literary context of Haggai had to be taken into account before any such judgment could be made. 2.2 A Typology of Descriptive Curse Formulae in the Hebrew Bible: Criteria and Classifications Further reflection persuaded me that these two patterns (“violation of covenant” versus “violation in covenant”) required still further nuancing and that it might be possible to establish a more complete typology of these descriptive curse formulae within the Hebrew Bible. The following typology represents the fruit of my subsequent investigation and involves the integration of a variety of criteria: (1) the presence of one or more descriptive curse formulae within a given text; (2) the literary form and setting in which such curses are presented (e.g., a single imprecation, shorter list, longer list, oracle of doom, etc.); (3) the purpose of the imprecations within the text in question (threat, warning, historical retrospective); and (4) the broader ideological/theological context, 40  On this point, see also P. R. B edford , “Discerning the Time: Haggai, Zechariah and the ‘Delay’ in the Rebuilding of the Jerusalem Temple,” in The Pitcher Is Broken: Memorial Essays for G. W. Ahlström (ed. S. W. Holloway and L. K. Handy; JSOTSup 190; Sheffield: JSOT, 1995), 71–94, here 74, who notes the “cultic” flavor of this pattern. Bedford insightfully observes, “[f]utility curses are not only to be found in the context of covenant breaking (as the demise of the kingdom, the exile and the absence of Yahweh are interpreted). They are also understood to be the result of cultic or ethical infractions (for example, 2 Sam 21:1–14; Amos 4:1–12; Hos 4:1–3; Isa 5:8–10; Joel 1–2).” Bedford places the futility curses in Haggai in the same category. 41  Petersen, Haggai and Zechariah 1–8 (see n. 32), 50.

952

John Kessler

purpose, and traditiohistorical rooting of the text as a whole (i.e., Priestly; Deuteronomic; wisdom, etc.). As a typology, it is therefore somewhat eclectic, integrating lexical, formal, and ideological considerations.42 The biblical materials containing descriptive curse formulae may be categorized as follows: a. Individual maledictions cited either alone or in series. These are generally drawn from the broader stock of ancient Near Eastern curses and are scattered in various contexts relating to divine judgment. The purpose and significance of these curses are highly context specific. Such maledictions occur frequently in prophetic oracles of doom and are creatively and distinctly configured (e.g., Isa 3:1–8; 3:16–4:1; 5:1–6; 13:16, 21–22; 14:20a; 16:10, etc.; Jer 3:3; 4:29; 5:15–17; 8:1–3, etc.; Ezek 7:10–13; 7:21–27).43 Occasionally they figure in legal contexts such as adjurations (Lev 5:1), trial by ordeal (Num 5:11–31), the issuing of judgments (Gen 3:14; 4:11; 9:25–26; Josh 9:22–23), or disputes between persons (1 Kgs 8:31–32).44 At times, such individual maledictions are pronounced to preclude the undertaking of specific acts (Josh 6:26).45 At other times, the consequences of the proscribed behavior are described using a metaphor such as the shrub in the wasteland in Jer 17:5–7. This pattern is in evidence in a wide variety of ancient Near Eastern texts (see infra). b. Curses used as a form of lex talionis or retribution in kind. Here afflictions are brought upon an offender that are fitting consequences of the evil committed. Exodus 22:22–24 is a prime example. In this text, the oppression of orphans and widows results in the death of the offenders, leaving their wives as widows and their children as orphans.46 A similar phenomenon exists in Amos 5:11. Other examples could be cited.47 c. Curses used as self-imprecations, invoked to demonstrate the speakerʼs innocence. Job 31:5–40 contains numerous such instances. 42  I wish to stress that the typology to be presented in this section is both preliminary and nonexhaustive. 43  See the fuller list in D. S tuart , Hosea–Jonah (WBC 31; Waco, TX: Word, 1987), xxxiii–xl. 44  Brichto, Problem of “Curse” (see n. 4), 42–55. 45  Following the excellent discussion and examples in Brichto, Problem of “Curse” (see n. 4), 79–80. 46  Note that none of the verbal forms normally associated with cursing appear here. Wives becoming widows and children orphans, however, clearly reflects traditional curse imagery as well as the implicit application of divine justice, with biting irony; see also Fensham, “Common Trends” (see n. 11), esp. 158, 170. On descriptions of lex talionis at work in narratives, see P. J. Nel, “The Talion Principle in Old Testament Narratives,” JNSL 20 (1994), 21–29. 47  See Prov 22:22–23. Note the analogous case in the ancient Near Eastern material where the effects of the curse relate to the primary activity of the deity who imposes them; see Hillers, Treaty Curses (see n. 6), 13–14.

Patterns of Descriptive Curse Formulae in the Hebrew Bible

953

d. Shorter allusions to a stock series of misfortunes (for example, “famine, sword, and pestilence”) within human or divine speech. Examples include Jer 14:12; 21:7, 9; 24:10; 27:8, 13; 29:17–18; 32:24, 36; 34:17; 38:2; 42:17, 22; 44:13; Ezek 5:12, 17; 6:11–12; 7:15; 12:16; 14:21; 2 Sam 24:13; 1 Chr 21:12; 2 Chr 20:9. Ezekiel 14:21 lists “four evil judgments” (‫)ארבעת שפתי הרעים‬. First Kings 8:37 enumerates a similar list.48 The significance and broader implications that may be drawn from such brief lists are highly context specific. e. Longer series of maledictions functioning as threats and deterrents. The curses in such lists function as penalties or deterrents with reference to violation of a textually proximate covenant and without reference to warning, remedial discipline, or divine long-suffering. Like much of the analogous ancient Near Eastern material, these maledictions are presented as a single list, without gaps for potential human response and divine forbearance. Although there may be elements of escalation in the punishments, these increments are not presented as having a restorative function or as creating opportunity for change or repentance. No provision is made for partial or remedial punishment (although such steps are not expressly precluded). Violation of the terms of the covenant grants the stronger party the right to terminate the covenant relationship altogether and to destroy the offender. This pattern appears in the epilogue of Hammurabi, Sefire I, Tell Fekherye, and the Vassal Treaties of Esarhaddon. Shorter enumerations are found in certain kudurru inscriptions. It is generally understood to be the outlook reflected in Deut 28.49 f. Longer series of incrementally applied maledictions intended to produce change and/or repentance. This pattern weaves together the ideas of punishment, corrective discipline, warning, and divine forbearance. It includes gaps or stages in the description of the imposition of the curses to allow for remediation in the divine-human relationship. Sometimes the descriptive curse formulae stop short of a full declaration of ultimate disaster, leaving open the possibility of future restoration.50 Nevertheless, if the curses are unheeded, the wrongdoerʼs offenses will result in great destruction (as above, e). This destruction and disaster need not be total, should the 48  F. I. Andersen and D. N. Freedman, Amos: A New Translation with Introduction and Commentary (AB 24A; New York: Doubleday, 1989), 440. 49  See M ilgrom , Leviticus 23–27 (see n. 39), 2347. There are differing opinions as to whether, in the D or Dtr perspective, Yahwehʼs relationship with Israel could be fully terminated due to covenant violation. Compare T. Römer and M. Z. Brettler, “Deuteronomy 34 and the Case for a Persian Hexateuch,” JBL 119 (2000), 401–419, here 412, n. 58, with G. E. Mendenhall and G. H erion , “Covenant,” ABD 1: 1179–1202. See also the nuanced discussion in C. Nihan, Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 542. 50  E.g., Amos 4 and Isa 5:25–29; 9:7[8]–10:4; see further discussion below.

954

John Kessler

stronger party choose to leave a remnant and restore the offender or the offenderʼs seed after a time of suffering and penitence (so Lev 26:40–45). This is generally understood to be the pattern in Lev 26; Amos 4; and Isa 5:25–29; 9:7–10:4.51 g. Various, nonincremental maledictions cited as divine responses to specific religious or ethical lapses on the part of the individual or community. These maledictions are meant to lead to remediation or correction of a specific fault. This pattern subdivides into two general types. The first is the imposition of maledictions as an incentive to prayer and repentance for unnamed faults, described only as “sins.” As such, this pattern is paradigmatic, serving as a model for dealing with individual sin in general.52 This pattern is in evidence in 1 Kgs 8, where the vocabulary used to describe these maledictions is shared with the standard curse motifs in ancient Near Eastern and other biblical materials (notably Lev 26 and Deut 28).53 A possible (but not necessary) implication of such terminology  I assume Lev 17–26 to be part of the work of one or more Holiness writers active at a time subsequent to the production of the earlier Priestly material. These tradents revised and recontextualized earlier material in light of the realities of their own age and their own theological preoccupations. On the priority of P to H, and the relationship of H to the Book of the Covenant and Deuteronomy, see the survey and discussion in J. Joosten, People and Land in the Holiness Code: An Exegetical Study of the Ideational Framework of the Law in Leviticus 17–26 (VTSup 67; Leiden: Brill, 1996), 10–16; idem , “Covenant Theology in the Holiness Code,” ZABR 4 (1998), 145–164, esp. 147; and the detailed discussion in B. M. L evinson , “The Manumission of Hermeneutics: The Slave Laws of the Pentateuch as a Challenge to Contemporary Pentateuchal Theory,” Congress Volume: Leiden, 2004 (ed. A. Lemaire; VTSup 109; Leiden: Brill, 2006), 281–324. Levinson utilizes this chronology in his numerous studies of legal development in the Hebrew Bible. Further arguments for the dating and sequence of the various sources, as well as surveys of the scholarly literature and alternative proposals, may be found in them. On Lev 17–26 in general, see the discussion in J. E. Hartley, Leviticus (WBC 4; Dallas: Word, 1992), 247–260; Joosten, “Covenant Theology”; idem, People and Land, 5–16; Milgrom, Leviticus 23–27 (see n. 39), 2363–2365; Nihan, Priestly Torah (see n. 49), 395–558. The specific dating of Lev 26 and its relationship to Lev 17–25 continues to be debated; see references above, especially Milgrom, Leviticus 23–27 (see n. 39), 2352–2363. On Isa 5:25–29; 9:7[8]–10:4, see H. W ildberger , Isaiah 1–12: A Commentary (trans. T. H. Trapp; CC; Minneapolis: Fortress, 1991), 218–245; E. Blum, “Jesaja und der ‫ דבר‬des Amos: Unzeitgemäße Überlegungen zu Jes 5,25; 9,7–20; 10,1–4,” DBAT 28 (1994), 75–95; J. L. Crenshaw, “ ‘A Liturgy of Wasted Opportunity’ (Am 4:6–12; Isa 9:7–10:4; 5:25–29),” Semitics 1 (1970): 27–37. 52  As M. W einfeld , Deuteronomy 1–11: A New Translation with Introduction and Commentary (AB 5; New York: Doubleday, 1991), 40, notes, Deuteronomy gives no explicit instructions for dealing with individual sin. He suggests, “[t]he authorʼs view seems to be that spiritual purification and repentance – consisting of confession and prayer – and not sacrificial offerings expiate sin.” 53  See Fensham, “Common Trends” (see n. 11); H. W. Wolff, Joel and Amos (Hermeneia; Philadelphia: Fortress, 1977), 213. 51

Patterns of Descriptive Curse Formulae in the Hebrew Bible

955

is that these infractions are seen to occur within a broader covenantal relationship.54 E. Talstra suggests that, in contrast to the lists of misfortunes in Deut 28 and Amos 4, 1 Kgs 8 describes various “cases” as part of an “affliction-prayer-forgiveness” schema. Thus the covenant curses of the broader tradition are shifted to a more focused case-by-case setting. They become “afflictions” (to use Talstraʼs term) sent to alert the individual or community of a specific fault requiring a concrete solution.55 A second form of this pattern occurs when one or more curses arise due to a specific, named fault. As above, the fault has created an impasse that cannot be removed until the fault is addressed. In the interim, the divinely sent malediction(s) will unrelentingly afflict the community. Achanʼs sin (Josh 7), Saulʼs breach of covenant with the Gibeonites (2 Sam 22), and the community in Yehudʼs failure to rebuild the temple (in Haggai)56 serve as examples of this pattern. In both variations of this pattern, once the specific transgression is remedied, the effects of the curse are removed. Until that time, the situation is at an impasse and the curseʼs effects remain. However, the fact that the maledictions are not presented as increasing in intensity suggests that the broader covenantal relationship is not threatened by them and that, left unresolved, the sins committed will result not in cataclysmic destruction of the individual or nation but simply a state of ongoing misery.  Of course, the fact that the prayer of 1 Kgs 8 is found within the so-called Deuteronomistic History would buttress such an assumption. On 1 Kgs 8 as Deuteronomic (or Deuteronomistic), see G. N. K noppers , Two Nations under God: The Deuteronomistic History of Solomon and the Dual Monarchies, Vol. 1: The Reign of Solomon and the Rise of Jeroboam (HSM 52; Atlanta: Scholars, 1993); idem, “Prayer and Propaganda: Solomonʼs Dedication of the Temple and the Deuteronomistʼs Program,” CBQ 57 (1995), 229–254. On broader issues related to the Deuteronomistic History, see the essays in T. Römer (ed.), The Future of the Deuteronomistic History (BETL 147; Leuven: Leuven University Press, 2000); G. N. Knoppers and J. G. McConville (eds.), Reconsidering Israel and Judah: Recent Studies on the Deuteronomistic History (Sources for Biblical and Theological Study; Winona Lake, IN: Eisenbrauns, 2000). 55  E. Talstra, Solomonʼs Prayer: Synchrony and Diachrony in the Composition of 1 Kings 8:14–61 (trans. G. Runia-Deenick; CBET; Kampen: Kok Pharos, 1993), 118–119. Referencing Talstraʼs analysis of 1 Kgs 8, M. J. B oda , A Severe Mercy: Sin and Its Remedy in the Old Testament (Siphrut; Winona Lake, IN: Eisenbrauns, 2010), 167, concludes, “Its categories of calamities are closely related to the curse materials found in Deuteronomy 28–30 and Leviticus 26 suggesting that the prayer is being conceptualized within the framework of the covenant.” 56  In this connection, it is crucial to distinguish any possible historical reasons for the communityʼs resistance to the templeʼs reconstruction (indeed, there may have been many factors involved) from the literary portrait of such resistance in Haggai. On this, see J. K essler , “Building the Second Temple: Questions of Time, Text, and History in Haggai 1.1–15,” JSOT 27 (2002), 243–256. Numerous interpreters fail to distinguish between the two and interject historical reasons for the peopleʼs reticence into the literary portrait of their reluctance in the book. This creates immense exegetical difficulties for the interpretation of the passage. 54

956

John Kessler

3 Amos 4:6–12 in the Context of the Curse Traditions of the Hebrew Bible 3.1 Amos 4:6–12: Closer and More Distant Parallels The foregoing (albeit preliminary) typology provides a backdrop for our study of the maledictions in Amos 4. Our focus in this section will be on the unique and distinctive aspects of Amos 4 when compared to texts closest to it in the typology suggested above. Even the most cursory reading of Amos 4 reveals that it bears a striking resemblance to several other texts involving descriptive curse formulae. H. W. Wolff has set forth an oft-cited chart comparing the elements in Amos 4, Lev 26, Deut 28, and 1 Kgs 8.57 Wolff sees Amos 4 as bearing the closest affinity to Lev 26. He states, The use of the first-person divine speech and direct address in the plural further associates Lev 26 with Am 4:6–11. [. . .] One must [. . .] conclude that Am 4:6–11 stands in proximity to the Holiness Code, which probably came into being in the latest period of the pre-exilic cultus. Lev 26 ‘presumably developed from a part of the agenda for the great autumn festival. Dtn 28, with its curses and, especially, with its references to Yahweh in the third person, is more distantly related.’58

Jeremias, for his part, sees Amos 4 as most closely connected to 1 Kgs 8.59 J. L. Crenshaw, E. Blum, and others have called attention to the connections between Amos 4 and Isa 5:25–29; 9:7–10:4,60 a text that Wolff omits from his list, likely because it lacks the extensive allusions to common ancient Near Eastern curse formulae contained in the others.61 3.2 Amos 4, Deuteronomy 28 and 1 Kings 8 As may be seen from the typology above, I view the curse pattern in evidence in Amos 4; Lev 26; and Isa 5:25–29; 9:7[8]–10:4 as distinct from the one found in Deut 28. Several features, among many, demonstrate this. For example, Lev 26; Amos 4; and Isa 5:25; 9:7[8]–10:4 (unlike Deut 28) use first-person singular verbs in the perfect and imperfect aspects, addressed to hearers in the  Wolff, Joel and Amos (see n. 53), 211.  W olff , Joel and Amos (see n. 53), 214. The embedded quote is from K. Elliger , Leviticus (HAT 1/4; Tübingen: Mohr Siebeck, 1966), 371. 59  J. J eremias , The Book of Amos: A Commentary (trans. D. W. Stott; OTL; Louisville: Westminster John Knox, 1998), 70–72; idem , Hosea und Amos (FAT 13; Tübingen: Mohr Siebeck, 1996), 198–213. 60  Blum, “Jesaja” (see n. 51); Crenshaw, “Liturgy” (see n. 51); Wildberger, Isaiah 1–12 (see n. 51), 225. 61  Due to the complexities involved in studying the specific relationship between Amos 4 and Isa 5:25–29; 9:7[8]–10:4, I can only address this issue in a general way in the present study. 57

58

Patterns of Descriptive Curse Formulae in the Hebrew Bible

957

second-person masculine plural. This construction, however, is absent from Deut 28. Furthermore, divine judgments are introduced by ‫ נתתי‬in Amos 4:6 and by ‫ ונתתי‬in Lev 26:4, 6, 11, 17, 19, 30–31. Deuteronomy 28, by contrast, uses ‫ נתן‬far less frequently62 and never in the first person.63 Conversely, the root ‫ ערר‬appears five times in Deut 28 (Deut 28:16–19) but not at all in Amos 4:6–11; Lev 26; or Isa 5:25; 9:7[8]–10:4.64 What is more, Yahweh never speaks in the first person in Deut 28, whereas Lev 26 consists entirely of Yahwehʼs first-person address to Israel. Thus the profoundly interpersonal “I and thou” ethos of Amos 4 is paralleled in Lev 26 but absent from Deut 28.65 However, and most important, the primary structuring device in Amos 4; Lev 26; and Isa 5:25; 9:7[8]–10:4 is the setting of the various maledictions as a series of incrementally stepped calamities, punctuated by intervals of reprieve designed to preempt the imposition of catastrophic divine judgment should the nation reverse its course. The maledictions function as disciplinary warnings.66 Deuteronomy 28, by contrast, lacks the notion of critical moments when repentance is possible. There, the maledictions function as threats. They are enumerated in the form of a list and structured rhetorically,67 utilizing various common descriptive curse formulae that range from simple futility curses to threats of loss of the land, descriptions of siege, deportation, exile, return to Egypt, and sale into slavery. The distance between Deut 28, on the one hand, and Amos 4; Lev 26; and Isa 5:25; 9:7[8]–10:4, on the other, is thus quite wide.68

 ‫ נתן‬occurs twice in the perfect (vv. 48, 65), twice in the imperfect (vv. 24–25), and once as a participle (v. 8, with reference to Yahwehʼs gift of the land). 63  This excludes 28:1, where the perfect of ‫ נתן‬makes reference to the giving of Yahwehʼs commandments. Also excluded here are the participial usages of ‫ נתן‬with reference to the land. 64  Deut 28 is generally seen as complex and composite, so the presence of ‫ ערר‬in vv. 16–19 should not be unduly pressed in this regard. Literary connections between the rest of Deut 27–28 and Amos 4; Lev 26; and Isa 5:25; 9:7[8]–10:4 might still be posited. Nevertheless, numerous scholars view Deut 28:15–19 as part of the earliest core of the chapter; see R. Nelson, Deuteronomy (see n. 29), 328–329; J. R. L undbom , Deuteronomy: A Commentary (ECC; Grand Rapids, MI: Eerdmans, 2013), 756–757. As such, the use of ‫ ערר‬in 16–19 is noteworthy and still suggests a distinction in outlook between Deut 28 and that of Amos 4; Lev 26; and Isa 5:25; 9:7[8]–10:4 at a certain level. See also the literature cited in n. 40, above. 65  W olff , Joel and Amos (see n. 53), 214, notes this distinction. It should, however, be noted that, from v. 36 onward, third-masculine plural forms are interspersed with the second-masculine plural forms; see P. Buis, “Comment au septième siècle envisageait-on lʼavenir de lʼalliance: Étude sur Lev 26, 3–45,” in Questions disputées dʼAncien Testament: Méthode et théologie (ed. C. Brekelmans; BETL 33; Leuven: Leuven University Press, 1974), 131–140, here 132. 66  The concept of corrective discipline is conveyed through the root ‫ יסר‬in Lev 26:18, 23, 28. 67  See Lundbom, Deuteronomy (see n. 64), 751–761. 68  See further Milgrom, Leviticus 23–27 (see n. 39), 2346–2348. 62

958

John Kessler

As we have seen, J. Jeremias suggests that Amos 4 is closer to 1 Kgs 8 than to Deut 28 or Lev 26. He maintains that Lev 26 served as a model for 1 Kgs 8, which in its turn provided the basis for Amos 4.69 However, if Amos 4; Lev 26; and Isa 5:25–29; 9:7[8]–10:4 stand at a significant distance from Deut 28, the distance between Amos 4; Lev 26; and Isa 5:25–29; 9:7[8]–10:4, on the one hand, and 1 Kgs 8, on the other, is even greater. True, all these texts utilize the motif of adversity leading to repentance. Moreover, several other parallels exist between all of these texts. Similar disasters occur whose purpose is to motivate the suffering individual or community to return (‫ )שוב‬to Yahweh (Amos 4:6, 8–11; 1 Kgs 8:35, 47–48).70 In Isa 5:25–29; 9:7[8]–10:4, ‫ שוב‬is utilized with reference to the removal of Yahwehʼs anger (Isa 5:25; 9:11, 16, 20; 10:4). Yet 1 Kgs 8 manifests significant differences from Lev 26; Amos 4; and Isa 5:25; 9:7[8]–10:4. These stem primarily from the function of the curse description formulae within them. Four major differences stand out. First, 1 Kgs 8 does not deal primarily with issues of national rejection of Yahwehʼs covenant through profound disobedience (cf. Deut 28) and its ultimate reconciliation and restoration (cf. Lev 26) or unheeded prophetic warnings leading to dramatic national disaster (cf. Amos 4; Isa 5:25; 9:7[8]–10:4). Rather, within the narrative flow of 1 Kings, the focus in 1 Kgs 8 is primarily upon the newly constructed temple as the culmination of Yahwehʼs purposes and its relationship to earlier cultic institutions.71 So, though 1 Kgs 8 does indeed deal with the remediation of transgression, its focus is

 Jeremias, Hosea und Amos (see n. 59), 198–213; idem, Book of Amos (see n. 59), 70–72, esp. 70, n. 16. Noting the numerous parallels and differences between these texts, he concludes that the latter is a penitential liturgy addressed to the exilic community. Jeremias understands Amos 4 to presuppose 1 Kgs 8 as well as Lev 26 and Deut 28. He rests his argument primarily on the reference to Sodom and Gomorrah in Amos 4:11 and, more broadly, on the fact that prayer in 1 Kgs 8 is addressed to Yahweh not only from within the temple but toward it from afar. Thus, he concludes, the words of Amos urge the exilic community not to miss their final remaining opportunity to avoid utter destruction by returning in repentance and praise to Yahweh. 70  W olff , Joel and Amos (see n. 53), 214, appropriately comments, “With each plague, according to Amos 4:6–11, Yahweh anticipated repentance. This combination of the plague series with the theme of ‘return’ is otherwise found only in a particular stratum of 1 Kgs 8 (a stratum which does not correspond to the final Deuteronomistic stratum), namely in 1 Kgs 8:33 and 35.” 71  Knoppers, Two Nations under God (see n. 54), 113–114, states, “[i]ndeed, it seems that in 1 Kings 8 the Deuteronomist is anxious to project an image of Solomon as a curator and guarantor of his nationʼs most sacred traditions. I would argue, however, that this stress upon Solomonʼs piety deliberately downplays the innovation involved in establishing the cultus of Solomonʼs royal shrine as normative for the whole people. [. . .] [T]he Deuteronomist not only integrates traditional institutions into the temple cultus, he weds these institutions, and the temple itself, to kingship. In doing so, he transforms older arrangements. Both king and temple are integral to national life.” 69

Patterns of Descriptive Curse Formulae in the Hebrew Bible

959

primarily on the role of prayer in or toward the temple.72 Second, both individual and collective cases of sin and forgiveness are in view in 1 Kgs 8. By contrast, Amos 4; Lev 26; and Isa 5:25; 9:7[8]–10:4 deal entirely with collective responsibility. Third, in 1 Kgs 8, the afflictions sent upon transgressors are presented as normal, even expected, occurrences. The parenthetical aside in v. 46, “for there is no one who does not sin,” underlines this point. The various instances of sin are introduced either by ‫( את אשר‬v. 31), ‫( אשר‬v. 33), or by ‫( כי‬v. 35), indicating that these are potential, even expected, occurrences. The iniquity of the nation in Amos 4; Lev 26; and Isa 5:25; 9:7[8]–10:4 is not of this nature. Rather, these texts concern Israelʼs inexcusable behavior within its relationship to Yahweh and its stubborn refusal to heed the providential warnings sent to it. Finally, in 1 Kgs 8, there is no sense of a series of punishments sent one after another, increasing in intensity, the neglect of which would lead to ultimate destruction. Taken together, these elements make the dissimilarities between 1 Kgs 8 on the one hand and Amos 4 on the other; Isa 5:25; 9:7[8]–10:4; and Lev 26 far greater than the similarities between them. Their classification within differing categories of curse materials is therefore wholly justifiable. 3.3 Amos 4:6–12 and Leviticus 26: Points of Commonality73 Amos 4 and Lev 26 share numerous points of commonality. As is frequently noted, their structure is strikingly similar. Amos 4 stands within the broader unit  As noted by Wolff, Joel and Amos (see n. 53), 214.  Beyond this point, I focus upon the points of commonality and dissimilarity between Amos 4 and Lev 26. The relationship of Isa 5:25–29; 9:7[8]–10:4 to the two former texts is highly significant but cannot be explored in detail here. The general similarity between Amos 4 and Isa 5:25–29; 9:7[8]–10:4 is beyond question (see our typology above). Both texts contain a refrain involving the verb ‫ שוב‬repeated five times (Amos 4:6, 8–11; Isa 5:25; 9:11–12, 16, 20; 10:4). Both relate repeated instances of Yahwehʼs affliction of his people, without positive effect: thus, “The people did not turn to him who struck them, or seek the LORD of hosts” (Isa 9:13) and “yet you did not return to me, says the LORD” (Amos 4:6, etc.). Both end on a note of indeterminacy and anticipate one (or several) further blows. Both leave open the possibility of some future hope. For these elements in Isaiah, see Wildberger, Isaiah 1–12 (see n. 51), 224–225, 233; H. G. M. Williamson, A Critical and Exegetical Commentary on Isaiah 1–27, Volume 1: Isaiah 1–5 (ICC; London: T&T Clark, 2006), 403–405. Discussion of the relationship between these texts is rendered particularly complex owing to issues surrounding the order of the material in Isaiah as it now appears; see Wildberger, Isaiah 1–12 (see n. 51), 224, and the survey in Williamson, Isaiah 1–5, 400–403; see also A. H. Bartelt, “Isaiah 5 and 9: In- or Interdependence?,” in Fortunate the Eyes That See: Essays in Honour of David Noel Freedman in Celebration of His Seventieth Birthday (ed. Astrid B. Beck et al.; Grand Rapids, MI: Eerdmans, 1995), 157–174; W. P. Brown, “The So-Called Refrain in Isaiah 5:25–30 and 9:7–10:4,” CBQ 52 (1990), 432–443; C. E. LʼHeureux, “The Redactional History of Isaiah 5:1–10:4,” in In the Shelter of Elyon: Essays on Ancient Palestinian Life and Literature in Honor of G. W. Ahlström (ed. W. B. Barrick and J. R. Spencer; JSOTSup 31; Sheffield: JSOT Press, 1984), 99–119. 72

73

960

John Kessler

of 3:1–6:14. There is a general consensus that 5:1–6:14 should be read as a unit consisting of parallel elements of devastating judgment and call to reform, with 5:1–17, itself in concentric pattern, at its core.74 There is far less unanimity of opinion regarding 3:1–4:13. Some, such as F. I. Andersen and D. N. Freedman, see it as closely connected to 1:1–2:16, whereas others see it as composed of several loosely connected sections: either 3:1–2, 3–8, 9–12, 13–15; 4:1–3, and 4–13 (thus S. M. Paul); or 3:1–2, 3–6, 7–8, 9–12, 13–15; 4:1–3, 4–13 (so R. B. Coote).75 Virtually all are agreed that the curse material in 4:6–11 is integrally connected to the condemnation of the official cult at Bethel in 4:4–5.76 The phrase ‫( וגם אני‬and I for my part) in v. 6a, following from the Gottesspruchformel in v. 5b, suggests as much.77 Further debate surrounds the meaning of v. 12 and its relationship to vv. 6–11. These words are read either as an announcement of imminent, inescapable doom78 or as the offer of a final chance for repentance.79

74  J. B arton , The Theology of the Book of Amos (Old Testament Theology; New York: Cambridge University Press, 2012), 23, n. 26. On 3–6 as a unit, see Andersen and Freedman, Amos (see n. 48), 461–471; Jeremias, Book of Amos (see n. 59), 83–85. 75  R. B. C oote , Amos among the Prophets: Composition and Theology (Philadelphia: Fortress, 1981), 69–75. 76  The arguments of M. Weiss, Amos (2 vols.; Jerusalem: Magnes Press, 1992) (Hebrew), constitute an important exception to this general trend. Weiss (Amos, 1:107) accepts vv. 4–5 as independent of the preceding unit, vv. 1–3, and of the following passage, vv. 6–12, in theme, style, and literary character. Dalit Rom-Shiloni has kindly provided the following summary of Weissʼs argument: “For Weiss, thematically, vv. 4–5 focus on worship at Bethel and Gilgal, whereas vv. 6–12 discuss the peopleʼs refusal to return to God without any reference to specific cultic sites; vv. 4–5 sarcastically call for activity, whereas vv. 6–12 confront the people with what they have failed to do. Verses 4–5 refer to the peopleʼs feelings as motivating their cultic behavior (v. 5b), whereas the accusation against the people in vv. 6–12 is set in the context of divinely imposed judgments. Stylistically, vv. 4–5 are built on rhythmic and laconic language, whereas vv. 6–12 feature a detailed and flowing style. The two units differ in tone, vv. 4–5 being much sharper, and whereas vv. 4–5 are satirical or even parodic, vv. 6–12 are words of admonition regarding past deeds ending with an ominous future prospect (v. 12). In their literary character, vv. 4–5 contain seven stitchs and close with the formula ‫נאם אדני יהוה‬, as is typical of Amos. However, vv. 6–12 form a closed unit, structured as 5+1 subunits defined by a repeated verset (vv. 6b, 7b, 9b, 10b, 11b). Weiss thus concludes that ‘there is no reason to assume that vv. 4–5 and vv. 6 and following were proclaimed in a single occasion’ (107).” 77  Paul, Amos (see n. 22), 141–142; W. R. Harper, A Critical and Exegetical Commentary on Amos and Hosea (ICC; Edinburgh: T&T Clark, 1905), 96; S. Amsler, “Amos,” in Osée, Joël, Amos, Abdias, Jonas (CAT 11A; Geneva: Labor et Fides, 1965), 157–247, esp. 201; Jeremias, Book of Amos (see n. 59), 69–74. 78  Paul, Amos (see n. 22), 151 and Weiss, Amos (see n. 76), 1:122–124. 79  Coote, Amos (see n. 74), 78; W. Brueggemann, “Amos IV 4–13 and Israelʼs Covenant Worship,” VT 15 (1965), 1–15; S. Amsler, “Amos: Prophète de la onzième heure,” TZ 318 (1965), 318–328; Wolff, Joel and Amos (see n. 53), 225; Jeremias, Book of Amos (see n. 59), 70–72.

Patterns of Descriptive Curse Formulae in the Hebrew Bible

961

This final call is sometimes seen as addressed either to the post-722 community at Bethel (so Wolff) or to the post-587 community in Judah (so Jeremias).80 Although assigned to different periods and/or tradents by the various interpreters of Amos, and despite certain problematic textual and verbal elements, the literary integrity of Amos 4:6–12 is generally admitted. It is widely agreed that vv. 6–11 consist of five strophes describing the imposition of divine judgment (6a, 7a, 9a, 10a, 11a), each concluding with the refrain ‫“( ולא שבתם עדי‬and yet you did not return to me”; Amos 4:6b, 8b, 9b, 10b, 11b).81 A secondary 7+1 structure focused around specific verbal forms followed by second-masculine plural pronominal suffixes has been suggested.82 More convincing is the observation that these five strophes begin with five first-person declarations in the perfect aspect, followed by a number of perfect83 and other nonperfective forms, and serving as a loose structure for the various maledictions visited upon Israel. In the initial first-person declaration (v. 6a) a single malediction is imposed; in the latter four, several are mentioned. The first (v. 6a) begins with ‫ וגם אני‬and the second (v. 7a) with ‫וגם אנכי‬. The first two consist of only a single perfect first-person singular verbal form, whereas the latter three begin with a perfect first-person singular verb and are followed by various verbal constructions.84 The various curses involved are famine (v. 6), drought at a critical time in the agricultural cycle (v. 7aα), selective withholding of rain (vv. 7aβ, 7b), overcrowded towns with insufficient water supply (v. 8a), crop disease (v. 9aα), locusts (v. 9aβ), pestilence (v. 10aα), death of choice young men (v. 10aβ), death and stench (v. 10bα), and devastation and blighting to near annihilation (v. 11a). The precise number of plagues involved is debated, but the organization of the ten different events mentioned into a broader structure of seven plagues is highly plausible.85  Wolff, Joel and Amos (see n. 53), 225, sees these words as intended to elicit a repentant confession for the communityʼs enthronement of fertility gods; see also J eremias , Book of Amos (see n. 59), 70–72. 81  See Paul, Amos (see n. 22), 143, on the importance of the pentad in Amos and in rabbinic literature. On the text-critical issues and various proposals for emendation of these maledictions, see Amsler, “Amos” (see n. 77), 198–200; Paul, Amos (see n. 22), 144–149. 82  J. L imburg , “Sevenfold Structures in the Book of Amos,” JBL 106 (1987), 217–222; See also A ndersen and F reedman , Amos (see n. 48), 440, who enumerate seven plagues. It should be noted that six of the verbs are in the perfect aspect, whereas the seventh is a waw-consecutive plus imperfect. 83  There are seven perfect first-person singular forms in the passage (or eight, if ‫ ַה ְרֹּבות‬is seen as a textual error for ‫)הרבתי‬. 84  The syntax of the verbal forms following these perfects involves numerous complexities. A ndersen and F reedman , Amos (see n. 48), 436–437, 441, acknowledge their anomalous nature but view all the various forms as referring to the past; W olff , Joel and Amos (see n. 53), 214, sees the imperfects as remnants of earlier curse formulations from which the Amosian tradents have drawn. 85  Crenshaw, “Liturgy” (see n. 51), 32, is uncertain as to the specific number of plagues here. J. A. S oggin , The Prophet Amos: A Translation and Commentary (trans. J. Bowden; 80

962

John Kessler

Leviticus 26, for its part, contains numerous structural elements that echo aspects of Amos 4. Verses 14–23 contain five groups of plagues introduced by ‫( ואם‬Lev 26:14, 15, 18, 21, 23, 27)86 and four references to a sevenfold punishment for rejection of Yahweh (Lev 26:18, 21, 24, 28). As is often noted, the maledictions in Lev 26:14–39 correspond inversely to the blessings of vv. 3–13.87 This is loosely analogous to the structure of Amos 4 with its five strophes, seven plagues, and seven first-person perfect forms.88 What is more, the sevenfold repetition of the phrase “to walk in hostility” (‫ ;הלך קרי‬Lev 26:21, 23, 24, 27, 28, 40, 41) is reminiscent of the Amosian fivefold refrain describing Israelʼs failure to repent (vv. 6, 8, 9, 10, 11).89 Divine judgments, introduced by ‫נתתי‬, appear in Amos 4:6 and by ‫ ונתתי‬in Lev 26:4, 6, 11, 17, 19, 30–31. Numerous verbal and thematic elements within the curse materials link the two passages. The famine of Amos 4:6 is echoed in Lev 26:20, 26, and the drought of 4:7 in 26:4 and 19. Yahwehʼs smiting in Amos 4:9 appears in Lev 26:24. The plague of pestilence (‫ )דבר‬in Amos 4:10 appears in Lev 26:25, and the curse of the sword (‫ )חרב‬of Amos 4:11 is echoed in Lev 26:6–8, 25, 33, 36–37.90 Both texts envisage a series of disasters, of increasing severity, visited upon the nation.91 And, in both texts, these curses are grouped into successive stages in anticipation of an even more devastating divine judgment in the future.92 This is achieved through the repeated use of ‫ ואם‬in Lev 26:14, 15, 18, 21, 23, 27;93 the motif of “walking in hostility” in Lev 26:21, 23–24, 27–28, 40–41; and the refrain “and yet you did not return to me” in Amos 4:6, 8, 9, 10, 11. These repetitions interrupt the OTL; London: SCM, 1987), 76, sees a series of six. A ndersen and F reedman , Amos (see n. 48), 440, suggest that there are seven. Paul , Amos (see n. 22), 143, also proposes the same number. Arriving at the number seven, however, involves grouping some of the plagues mentioned individually under a single heading. Thus, for example, Andersen and Freedman, Amos (see n. 48), 440, see all of the elements of v. 7 as part of a single plague. 86  So Milgrom, Leviticus 23–27 (see n. 39), 2289–2290; Hartley, Leviticus (see n. 51), 456. 87  B. A. Levine, Leviticus: The Traditional Hebrew Text with the New JPS Translation (JPS Torah Commentary; Philadelphia: Jewish Publication Society, 1989), 276; Milgrom, Leviticus 23–27 (see n. 39), 2288. 88  On the significance of the numbers five and seven, see Milgrom, Leviticus 23–27 (see n. 39), 2288, who similarly notes other key terms that occur seven times in Lev 26. 89  Moreover, this also corresponds to the fivefold refrain in Isa 5:20–25 and 9:7[8]–10:4. 90  For further discussion of the points of commonality between these texts, see S oggin , Prophet Amos (see n. 85), 78–79. 91  On the specific blessings and curses contained in Lev 26 and their number, see Hartley, Leviticus (see n. 51), 456–458; Milgrom, Leviticus 23–27 (see n. 39), 2286–2290. 92  As noted above, there are strong differences of opinion as to whether Amos 4:12 implies further hope of reprieve. See infra, pp. 969, and 973–974 n. 159, for a fuller discussion of the issue of gaps and increment in the judgments in Amos 4 and Lev 26. 93  Fischer, “Need for Hope” (see n. 39), 374, n. 18, comments, “[T]he beginnings with ‫ואם‬ in 26:14, 18, 21, 23, 27 [. . .] always indicate a new stage in Godʼs efforts to bring his people back to the right way.”

Patterns of Descriptive Curse Formulae in the Hebrew Bible

963

litany of curses and introduce the idea of critical moments when change for the better (or its opposite) may occur and wherein the nationʼs responsiveness (or lack thereof) to Yahwehʼs demands can be assessed. Furthermore, in both Lev 26 and Amos 4, Yahweh repeatedly addresses the nation or describes his activities toward it by means of several first-person singular verbs, addressed to hearers in the second-masculine plural. This sets up a profoundly interpersonal, or “I and thou,” tone in both passages.94 3.4 Amos 4 and Leviticus 26: Differences in Form, Function, and Worldview Despite the parallels just noted, several striking differences exist between Amos 4 and Lev 26, especially with reference to their underlying ideology and conceptual framework. It is to these differences that we now turn. My broader purpose here is to demonstrate that texts containing descriptive curse formulae and bearing great lexical, formal, and thematic similarities may nevertheless exhibit significant differences that stem from each textʼs underlying ideological or theological convictions. In that which follows, I will leave aside the obvious temporal difference between these texts – Amos 4 presents its maledictions as a historical retrospective, whereas the cursings in Lev 26 are set forth as threats and warnings – and examine eight areas of significant contrast between them.95 First, the two texts contain a very different mood and emotional tone. Leviticus 26 bears the marks of a bitter and disintegrating interpersonal relationship (see infra), a tone lacking in Amos 4:6–12. As noted above, both Amos 4 and Lev 26 contain numerous first-person singular verbs and suffixes and second-masculine plural verbs and suffixes, setting up an “I and thou” interaction – quite frequently a conflictual one. Nevertheless, the tone of the dispute in the two texts is markedly different. In Amos, there is a distinct absence of vitriol or venom in the portrait of both Yahwehʼs and Israelʼs responses.96 Israelʼs failure does not consist of overt hostility but is manifested through the absence of a positive response to Yahweh. Furthermore, Israelʼs failure to return is described without comment. No explicit reason is given for this failure – did it arise from folly or simple indifference? There is no clear reference to hostility or antipathy. What is more, the series of disasters in vv. 6–11 is seen as a disciplinary warning, and 94  As indicated above, this dynamic is absent in Deut 28; see also Wolff, Joel and Amos (see n. 53), 214, who notes this distinction. 95  In his careful discussion of these two passages, T. S. H adjiev , The Composition and Redaction of the Book of Amos (BZAW 393; Berlin: de Gruyter, 2009), esp. 149–155, also indicates a number of significant points of contrast between them. Hadjievʼs interest, however, is primarily directed toward the relative dating of Amos 4 vis-à-vis Lev 26. I will engage some of his observations below. The difference between the two passages is also noted by M. Graupner and H.-J. Fabry, “‫שׁוב‬,” TDOT 14:485. 96  Verse 12, “this I will do to you,” admittedly does take a sharper tone.

964

John Kessler

there is no evidence of increasing irascibility on Yahwehʼs part. In Lev 26, by contrast, the atmosphere is one of increasing mutual animosity. Israelʼs failure is seen primarily not as the absence of returning (‫ )שוב‬but rather as the presence of ongoing hostility. This is most clearly visible where Israelʼs determination to walk in hostility (‫ )קרי‬to Yahweh (vv. 21a, 23, 27, 40) is met with Yahwehʼs corresponding enmity (also ‫ קרי‬in vv. 24, 41a). Indeed, Yahwehʼs hostility of v. 24 becomes “furious hostility” (‫ )בחמת קרי‬in v. 28.97 Emotional language of animus abounds. When Israel rejects (‫ )מאס‬and abhors (‫ )געל‬Yahwehʼs ordinances (26:15, 42, cf. Jer 14:19; Ezek 14:25), and acts treacherously against him (26:40), Yahweh will turn and abhor Israel (‫געל‬, v. 30, albeit temporarily; cf. v. 44). He will set his face against them (‫נתן פנה ב׳‬, 26:17; cf. Lev 17:10; Ezek 14:8) and execute the vengeance of the covenant (‫ )נקם ברית‬upon them (26:25; cf. Deut 32:41, 43).98 Such emotionally charged language is entirely absent from Amos 4. Vehement language does appear elsewhere in Amos, but it is of a different order.99 Second, Amos 4 implies that Israel has the real ability to turn and repent before the final catastrophe arrives. The various disasters sent by Yahweh were corrective in nature, presupposing that these measures could have been effective had the people been willing to hear. As such, the text seems to betray a certain astonishment, signaled by its refrain, that the nation, smitten so grievously time after time, has learned nothing. Much discussion has surrounded the wisdom motifs in Amos.100 The description of Israelʼs failure to learn from Yahwehʼs  Note also the observation of H. Ringgren, ‫קרה‬, TDOT 13:162: “The divine reaction reflects the human provocation. [. . .] [T]here is an intensification of the punishment of hostility.” 98  Similar language of abhorrence and disgust abounds in Ezekiel, see B. J. S chwartz , “Ezekielʼs Dim View of Israelʼs Restoration,” in The Book of Ezekiel: Theological and Anthropological Perspectives (ed. M. S. Odell and J. T. Strong; Atlanta: Society of Biblical Literature, 2000), 43–67. On the question of literary dependence, see Nihan, Priestly Torah (see n. 49), 543–544 (Lev 26 presupposes Ezekiel or parts thereof); pace M. A. Lyons, From Law to Prophecy: Ezekielʼs use of the Holiness Code (LHBOTS 507; New York: T&T Clark, 2009); idem, “Transformation of Law: Ezekielʼs Use of the Holiness Code (Leviticus 17–26),” in Transforming Visions: Transformations of Text, Tradition and Theology in Ezekiel (ed. W. A. Tooman and M. A. Lyons; Eugene, OR: Pickwick, 2010), 1–32; R. Levitt Kohn, A New Heart and A New Soul: Ezekiel the Exile and the Torah (JSOTSup 358; London: Sheffield, 2002); Milgrom, Leviticus 23–27 (see n. 39), 2348–2352 (Ezekiel presupposes Lev 26). The issue cannot be discussed in the present context. 99  See Amos 5:21; 6:8. But there Yahwehʼs vitriol is directed against hypocritical cultic practices and pride. 100  See, e.g., H. W. Wolff, Amosʼ geistige Heimat (WMANT; Neukirchen-Vluyn: Neukir­chener Verlag, 1964); J. L. Crenshaw, “The Influence of the Wise upon Amos: The Doxologies of Amos and Job 5:9–16, 9:5–10,” ZAW 79 (1967), 42–52; J. A. Soggin, “Amos and Wisdom,” in Wisdom in Ancient Israel: Essays in Honor of J. A. Emerton (ed. J. Day et al.; Cambridge: Cambridge University Press, 1995), 119–123; S. L. Terrien, “Amos and Wisdom,” in Israelʼs Prophetic Heritage: Essays in Honor of James Muilenburg (ed. B. W. Anderson and W. J. Harrelson; New 97

Patterns of Descriptive Curse Formulae in the Hebrew Bible

965

disciplinary measures is reminiscent of the fool who is smitten a hundred times to no avail (Prov 17:10). Thus, if Israel has shown itself incapable of repentance, even out of enlightened self-interest, such a failure arose not from an innate inability to respond but from pure folly (in the biblical sense of refusal to heed reproof). Leviticus 26, by contrast, presents a very conflicted view of the nationʼs capacity to respond. Its heart is described as proud, uncircumcised, and in need of radical change (26:41; cf. Jer 9:25; Ezek 44:7, 9; cf. the opposite in Deut 10:16; 30:6; Jer 4:4).101 Thus, though the successive stages of cursings in vv. 14–39 clearly imply the theoretical possibility of repentance, it is only after the final catastrophes of vv. 31–33, 36–39, including the devastation of the land, exile to enemy land, living on the verge of total annihilation, and following a long period of abandonment (vv. 36–39) that change ensues. It is noteworthy that, at this point, a significant distinction exists between Lev 26 and Ezek 36. In Lev 26, no heart renewal through the action of the divine spirit is explicitly mentioned.102 Rather, the emphasis falls upon the role of suffering in Israelʼs recognition of its guilt,103 its repentance, and the confession of its own sin and the sin of the ancestors – stock motifs of penitential prayer.104 Thus, in contrast to Amos 4, it is only through such experiences of intense suffering that change arises in Lev 26. York: Harper & Brothers, 1962), 108–115; G. H. W ittenberg , “A Fresh Look at Amos and Wisdom,” OTE 4 (1991), 7–18; H. M. Barstad, The Religious Polemics of Amos: Studies in the Preaching of Am 2,7B–8; 4,1–13; 5,1–27; 6,4–7; 8,14 (VTSup 34; Leiden: Brill, 1984), 7–10. 101  See W. E. L emke , “Circumcision of the Heart: The Journey of a Biblical Metaphor,” in A God So Near: Essays on Old Testament Theology in Honor of Patrick D. Miller (ed. B. A. Strawn and N. R. Bowen; Winona Lake, IN: Eisenbrauns, 2003), 299–319. See also the discussion in F ischer , “Need for Hope” (see n. 39), 383. Fischer contrasts Lev 26:41, where Israelʼs uncircumcised heart must be humbled, with Deut 30:6, where Yahweh himself provides the nation with the ability to exhibit an appropriate response by effecting upon it the “circumcision of the heart.” 102  Nihan, Priestly Torah (see n. 49), 544–545. See also Lyons, “Transformation of Law” (see n. 98). 103  J. M ilgrom , “The Priestly Doctrine of Repentance,” RB 82 (1975), 186–205; B oda , Severe Mercy (see n. 55), 84–85. 104  See R. J. B autch , Developments in Genre between Post-Exilic Penitential Prayers and the Psalms of Communal Lament (AcBib; Leiden: Brill, 2003); M. J. Boda, Praying the Tradition: The Origin and Use of Tradition in Nehemiah 9 (BZAW 277; Berlin: de Gruyter, 1999); idem , “From Complaint to Contrition: Peering through the Liturgical Window of Jer 14,1–15,4,” ZAW 113 (2001), 186–197; idem , “The Priceless Gain of Penitence: From Communal Lament to Penitential Prayer in the ‘Exilic’ Liturgy of Israel,” HBT 25 (2003), 51–75; idem, “Renewal in Heart, Word, and Deed: Repentance in the Torah,” in Repentance in Christian Theology (ed. M. J. Boda and G. T. Smith; Collegeville, MN: Liturgical Press, 2006), 3–24; idem, Severe Mercy (see n. 55); M. J. Boda et al., Seeking the Favor of God, Volume 1: The Origins of Penitential Prayer in Second Temple Judaism (EJL 21; Atlanta: Society of Biblical Literature, 2006). As is often noted, Lev 26 and Ezek 36 differ markedly regarding the confession of the sins of the ancestors. In Ezek this idea is suppressed, likely because it conflicts with the overall perspective of individual responsibility in the book.

966

John Kessler

Third, and following from the previous observation, the terminology used to describe the response desired of Israel differs between the two texts. Essentially, both texts deal with the necessity of a transformation of life – turning away from things displeasing to Yahweh and returning to him – manifested by right attitudes and actions. Such a transformation is generally termed “repentance.”105 However, significant differences in the language and broader traditiohistorical framework used to describe this transformation appear in Amos 4, on the one hand, and Lev 26, on the other. In Amos, Israelʼs failure consists in refusing to return (‫ )שוב‬to Yahweh.106 Broadly speaking, one can say that ‫ שוב‬involves a reversal of course, turning from someone or something and returning to someone or something (Isa 1:27–28; 6:10; Jer 3:1, 10; Hos 2:7 [ET 2:5]; etc.).107 The language of ‫ שוב‬appears quite frequently in prophetic preaching. Calls to return to Yahweh are given in staccato appeals, without excessive explanation of what is involved.108 Frequently the emphasis falls on specific actions to be taken.109 Furthermore, these calls to ‫ שוב‬generally carry with them a sense of directness and certitude – should the nation return, Yahwehʼs acceptance of it is promised.110 Numerous suggestions have been made as to what such a return 105  On the subject of repentance more generally, see J. Milgrom, “Repentance in the OT,” IDBSup 736–738; T. M. R aitt , “The Prophetic Summons to Repentance,” ZAW 83 (1971), 30–49; J. Unterman, “Repentance and Redemption in Hosea,” SBLSP 21 (1982), 541–550; T. G. Smothers, “Preaching and Praying Repentance in Hosea,” RevExp 90 (1993), 239–246; B oda , “Renewal” (see n. 104); C. J. D empsey , “ ‘Turn Back, O People’: Repentance in the Latter Prophets,” in Repentance in Christian Theology (ed. M. J. Boda and G. T. Smith; Collegeville, MN: Liturgical Press, 2006), 47–66; T. E. F retheim , “Repentance in the Former Prophets,” in Repentance in Christian Theology (ed. M. J. Boda and G. T. Smith; Collegeville, MN: Liturgical Press, 2006), 36–41; K. M. OʼConnor, “Repentance in First-Person Plural,” Journal for Preachers 31 (2008), 9–13; M. J. Boda, “Repentance,” in Dictionary of the Old Testament: Prophets (ed. M. J. Boda and J. G. McConville; Downers Grove, IL, 2011) 664–671. 106  The bibliography regarding ‫ שוב‬is sizeable. See especially G raupner and Fabry, “‫”שׁוב‬ (see n. 95), 485; J. A. Thompson and E. A. Martens, “‫שׁוב‬,” NIDOTTE 55–59; J. A. Soggin, “‫שׁוב‬,” TLOT 3:1312–1317; and W. L. Holladay, The Root Šûbh in the Old Testament with Particular Reference to Its Usages in Covenantal Contexts (Leiden: Brill, 1958). There is no reason to suppose that the use of ‫ שוב‬as a term to describe religious reform is a late development; see Wolff, Joel and Amos (see n. 53), 220: “As far as we know, Hosea (2:9[7]; 3:4–5; 14:2[1]) and Isaiah (9:12[13]; 30:15) were the first to recognize that the purpose of judgment was to promote such a ‘return’; the Deuteronomist elevated the concept to the status of an admonition.” 107  Boda, “Repentance” (see n. 105), 664, defines repentance in the prophetic books as “the human experience of turning away from attitudes and activities opposed to God and his ways and turning to God and his ways.” 108  E.g., Isa 31:6; Jer 3:12, 14, 22; 18:11; 25:5; 31:21; 35:15; Ezek 14:6; 18:30; 33:11; Hos 14:2 (ET 14:1); Joel 2:12–13; Zech 1:3–4; Mal 3:7. 109  Isa 1:27–28; 55:7; 58:13; Jer 4:1–2; 18:11; 23:22; Ezek 13:22; 18:21. 110  This is most clearly seen in texts such as Zech 1:3–4, where an imperative followed by an imperfect or cohortative is best seen as having either a telic function or the force of a

Patterns of Descriptive Curse Formulae in the Hebrew Bible

967

specifically entails in Amos, although these cannot be evaluated in detail here.111 They include: (1) a return to worship at Jerusalem,112 (2) rejection of foreign deities,113 and (3) moral rectitude.114 Within the broader context of Amos 3–6, however, the concept conveyed by ‫ שוב‬most likely denotes a positive response to the prophetʼs preaching: a rejection of hypocritical religious activity and social and economic corruption (5:4–6, 14–15, 21–24).115 Most significantly, there is no explicit mention of prayer, cleansing, or rites of repentance in 4:6–12 or in Amos 3–6 more generally (cf. Joel 1:13–14; 2:12–17).116 The prophetic ‫ שוב‬does at times explicitly mention such inward motivations (Isa 6:10; 9:12 [ET 9:13]; 19:22; 31:6) or confession (Jer 3:12–23). However, such ideas do not appear in Amos. By contrast, in Lev 26 Israelʼs path to restoration is described apart from the use of ‫שוב‬.117 Furthermore, in Lev 26:40–45, much attention is given to more-internal matters such as remorse, confession, and purification (see infra). This is not, of course, to suggest that the tradents of Lev 26 did not share the ethical concerns expressed in the ‫ שוב‬ideology of the prophetic and Deuteronomic voices and a belief that devotion to Yahweh must pass from words to actions. Alongside their concerns regarding idolatry, cultic practices, and proscribed sexual relations, the Holiness writers also shared much of the same social ethic as is evident in Amos and Isaiah (cf. Lev 19:9–17; 33–35). However, in Lev 26 the restoration of the divine-human relationship does not have the immediacy seen in the texts involving ‫ שוב‬cited above. Far greater detail is given regarding the process involved in Israelʼs movement toward Yahweh and in Yahwehʼs reaction to it. Israel manifests several responses not explicitly named in Amos: the confession (hithpael of ‫ )ידה‬of its own sins (‫ )עון‬and treachery (‫ )מעל‬and the sins of its ancestors,118 the humbling (‫)כנע‬ promise; see Joüon 116b, f. 111  Outside 4:6–11, the verb occurs in Amos only in 1:3, 6, 8, 9, 11, 13; 2:1, 4, 6. 112  Coote, Amos (see n. 74), 50–53. 113  Wolff, Joel and Amos (see n. 53), 225. 114  Paul, Amos (see n. 22), 139. 115  A likely synonym of ‫ שוב‬in Amos is the verb ‫דרש‬, which appears in 5:4, 5, 6, 14 and 8:12. 116  Wolff, Joel and Amos (see n. 53), 225, suggests that prayer and repentance are in view in Amos: “The return to Yahweh, heretofore neglected, is now finally to become a reality through repentant confession to and fervent praise of him.” But where exactly are repentance and prayer to be seen in Amos? 117  Milgrom, Leviticus 23–27 (see n. 39), 2363–2365, notes this difference and sees it as a sure indication that Lev 26 precedes rather than follows Amos 4. On this diachronic issue see further below. 118  On the confession of sin, see Lev 5:5; 16:21; 26:40; Num 5:7; Dan 9:4, 20; Ezra 10:1; Neh 1:6; 9:2–3; 2 Chr 30:22. For ‫ואת עון אבתם‬, see Lev 16:21; 26:39–40; Isa 14:21; 65:7; Jer 11:10; 14:20; Ezek 18:17; Ps 109:14; Dan 9:16. On ‫מעל‬, see Lev 5:15, 21; 26:40; Num 5:6, 12, 27; 31:16.

968

John Kessler

of its uncircumcised heart (‫)לב ערל‬,119 and the making of amends (‫)רצה‬120 for its sins (26:41).121 What is more, in v. 34, even the land is purged (‫)רצה‬122 and enjoys the Sabbaths due to it. As well, Yahwehʼs response is portrayed less directly than in Amos.123 Yahwehʼs renewed acceptance of Israel is, in the final analysis, rooted not in Israelʼs repentance but in the covenantal relationship with the ancestors and the exodus generation.124 The difference between the two response patterns highlights the distinct traditiohistorical rootings of the two texts. As noted, the ideology of return (‫ )שוב‬is a core element of relational repair in the prophets (see references above) and to a lesser degree in certain Deuteronomic texts (Deut 4:30; 30:2, 10; 1 Kgs 8:48; 2 Kgs 17:13).125 The terms describing Israelʼs rebellion and restoration in Lev 26 are, by contrast, evocative of themes deeply rooted in Priestly traditions.126 This is not of course to deny any sense of interiority or confession in the Amosian ‫ שוב‬ideology 119  On ‫ ערל‬and ‫לב‬, see Deut 10:16; 30:6. On ‫ לב‬and ‫כנע‬, see 2 Chr 32:26; Jer 4:4; 9:26; Ezek 44:7, 9. 120  Compare Lev 26:34, 41, 43 with Isa 40:2; 2 Chr 36:21. Boda, Severe Mercy (see n. 55), 84 notes the Priestly use of this root (or its homonym) in connection with sacrifice in Lev 1:4; 7:18; 19:7; 22:23, 25, 27. He suggests that the idea of payment is assumed in all these instances and that “the Exile itself is functioning in the role of sacrifice” in Lev 26. 121  The syntax of v. 41 is disputed, see below and C. Nihan, “The Priestly Covenant, Its Reinterpretations, and the Composition of ‘P,’ ” in The Strata of the Priestly Writings (ed. S. Shectman and J. S. Baden; ATANT 95; Zürich: Theologischer Verlag, 2009), 87–134; Milgrom, Leviticus 23–27 (see n. 39). 122  M ilgrom , Leviticus 23–27 (see n. 39), 2323, comments regarding the translation of this verb in v. 34: “Scholars are at their witsʼ end to render this word.” He goes on to present several possibilities, and his conclusion that the overarching idea is one of restoration or repayment makes good sense in this context (compare Joel 2:25). He furthermore sees ‫רצה‬ here as a synonym for ‫כפר‬. 123  That is, in Amos 4 the repeated refrain implies that had Israel turned from its errant ways, Yahweh would doubtless have relented in sending disaster. 124  On which, see below. 125  On Amos 4 and Dtr ideology more generally, see Hadjiev, Composition and Redaction (see n. 95), 159–160; Barstad, Religious Polemics of Amos (see n. 100). See below on the questions related to the covenant in Amos. 126  Note especially the notion of purification and payment implicit in ‫ רצה‬in Lev 26:34, 41, 43 as well as the Priestly character of the various terms in vv. 40–45, noted above. For the general Priestly outlook of this section, see Nihan, Priestly Torah (see n. 49), 540–541. See also the numerous explorations of Priestly influence in the confession of sin, especially Boda, Praying the Tradition (see n. 104); idem , “The Priceless Gain of Penitence” (see n. 105); idem , “Renewal” (see n. 104); idem , Severe Mercy (see n. 55); B. A. L evine , “The Epilogue to the Holiness Code: A Priestly Statement on the Destiny of Israel,” in Judaic Perspectives on Ancient Israel (ed. J. Neusner et al.; Philadelphia: Fortress 1987), 9–34; Milgrom, “Priestly Doctrine of Repentance” (see n. 103); idem, Cult and Conscience: The Asham and the Priestly Doctrine of Repentance (SJLA 18; Leiden: Brill, 1976); B. J. Schwartz, “The Bearing of Sin in the Priestly Literature,” in Pomegranates and Golden Bells: Studies in Biblical, Jewish, and Near Eastern Ritual, Law, and Literature in Honor of Jacob Milgrom (ed. D. P. Wright et al.;

Patterns of Descriptive Curse Formulae in the Hebrew Bible

969

(cf. Hos 14:2–4 [ET 14:1–3]) but simply to note the priority given to such interiority in the Priestly perspective.127 Fourth, the calamities in Amos 4 are presented as occurring over a period of time, with gaps in between them. Although scholars dispute the precise referents of these catastrophes,128 few would see them as being imposed one directly on top of the other on the same group of people, with no intervals or relief between them. For example, rain falling upon one city but not upon another (4:7–8) implies some relief of drought in the land – even a surfeit of rain in some areas.129 Moreover, the repetition of the refrain lamenting Israelʼs refusal to return implies that, in the wake of each blow, Yahweh anticipated that the nation would pause to take stock, learn wisdom, and amend its ways. In Lev 26, by contrast, the maledictions of vv. 14–39 are presented in a more sustained and unrelenting manner, one on top of the other. This is evident in ‫ ויספתי‬in v. 21 and ‫ ואם באלה‬in v. 23, and implied in ‫ ואם עד אלה‬in v. 18 and ‫ ואם בזאת‬in v. 27. As Levine states, “The Execration is composed in an escalating scale with admonition heaped upon admonition. If the Israelites do not return to God after one series of tragic circumstances, then even more horrible punishments will ensue.”130 Israelʼs ongoing hostility to Yahweh continues in the midst of and despite its sufferings. Nor can the maledictions be seen as touching only some of the population (as in Amos). Rather, they engulf the whole nation. Fifth, the concept of covenant is mentioned explicitly in Lev 26 and occupies a central place, occurring in Lev 26:9, 15, 25, 42 (3×), 44–45.131 Much recent discussion has focused on which covenantal traditions are implied in the passage (uniquely Priestly, or Priestly and Deuteronomic)132 and whether it is seen as Winona Lake, IN: Eisenbrauns, 1995), 3–21; J. Sklar, Sin, Impurity, Sacrifice, Atonement: The Priestly Conceptions (Hebrew Bible Monographs 2; Sheffield: Sheffield Phoenix, 2005). 127  Hadjiev, Composition and Redaction (see n. 95), 159–160. 128  See, e.g., the discussion of A ndersen and F reedman , Amos (see n. 48), 436; Paul , Amos (see n. 22), 143; Soggin, Prophet Amos (see n. 85), 76–78. 129  Thus Soggin, Prophet Amos (see n. 85), 76. 130  Levine, Leviticus (see n. 87), 181. See also Hartley, Leviticus (see n. 51), 458. 131  The bibliography here is enormous. For an orientation to the major studies and positions taken, see H artley , Leviticus (see n. 51), 457–462; M ilgrom , Leviticus 23–27 (see n. 39); N ihan , Priestly Torah (see n. 49); idem , “Priestly Covenant” (see n. 121); O lyan , “Status of the Covenant” (see n. 31); J. S tackert , “Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction: Leviticus 26 as a Test Case,” in The Pentateuch: International Perspectives on Current Research (ed. K. Schmid et al.; Tübingen: Mohr Siebeck, 2011), 369–386; Joosten, People and Land (see n. 51); idem, “Covenant Theology” (see n. 51), all with extensive bibliography. 132  See J. S tackert , “The Holiness Legislation and Its Pentateuchal Sources: Revision, Supplementation, and Replacement,” in Strata of the Priestly Writings: Contemporary Debate and Future Directions (ed. S. Shectman and J. S. Baden; Zürich: Theologischer Verlag Zürich, 2009), 187–204; Nihan, Priestly Torah (see n. 49), 535–541; idem, “Priestly Covenant” (see n. 121), 104–115. Stackert sees the covenant in Lev 26 as an extension of the covenant idea in

970

John Kessler

conditional or unconditional.133 The position taken here is that, in agreement with Lohfink, Nihan, and others, the basic schema in the Priestly literature (introduced in its earliest stages and retained in its later growth) is that of an eternal covenant between Yahweh and the community as a whole, from which individuals can be cut off should they break it. The covenant itself, however, cannot be terminated but perdures so long as the seed of Israel remains in existence.134 In Lev 26, however, this perspective is extended to an entire generation or historical moment. The second- and third-masculine plural suffixes repeated throughout the chapter attest to such a perspective.135 As such, the cutting off of a recalcitrant individual (as is the case in the karet regulations elsewhere)136 is now transferred collectively to a recalcitrant generation that must suffer a period of exclusion from its relationship with Yahweh.137 This would seem to reflect the impact of the Deuteronomic and prophetic understanding of covenant in Lev 26. As such, a generation can lose the presence of Yahweh and its status as an active covenant partner through its disobedience. Only after a time of alienation and suffering can the relationship be restored.138 Such a perspective lies beneath the affirmations in Lev 26 that, if Israel lives in obedience, Yahweh will maintain his covenant with them (v. 9),139 place his dwelling140 in their midst (v. 11), the earlier Priestly material alone. Nihan, by contrast, sees Lev 26 as incorporating covenantal ideas found in other pentateuchal and prophetic literature. See the fuller discussion below. 133  It is widely acknowledged that these terms are reductionistic and set up a false antithesis. However this should not mask the real distinction between Priestly and non-Priestly conceptions of covenant. On the covenant in Lev 26, see the contrasting positions of S tackert , “Holiness Legislation” (see n. 132); M ilgrom , Leviticus 23–27 (see n. 39), 2352–2363; Nihan, “Priestly Covenant” (see n. 121); idem, Priestly Torah (see n. 49), 537–545. 134  N ihan , “Priestly Covenant” (see n. 121), 104–115, argues that, by incorporating the notion of a generational, collective cutting off (from Deuteronomic traditions) into the larger Priestly schema, Lev 26 harmonizes and reconciles the two traditions. See also the arguments in Joosten, “Covenant Theology” (see n. 51); Stackert, “Holiness Legislation” (see n. 132); idem , “Distinguishing Innerbiblical Exegesis” (see n. 133), who see no intermingling of traditions. 135  Second-masculine plural: Lev 26:1, 4–9, 11–13, 15–22, 24–26, 28–39. Third-masculine plural: Lev 26:36, 39–41, 43–45. 136  Gen 17:14; Exod 12:15, 19; 30:33, 38; 31:14; Lev 7:20–21, 25, 27; 17:4, 9; 18:29; 19:8; 20:18; 22:3; 23:29; Num 9:13; 15:30; 19:13, 20. 137  As observed by Stackert, “Holiness Legislation” (see n. 132), 383. Both Stackert and Nihan correctly observe that such a hiatus does not nullify the broader covenantal relationship between the two parties. 138  See Deut 4:25–31; 30:1–10. This same motif is highly important in the prophetic literature; see Hos 3:1–5; Mic 5:2 (ET 5:3); Isa 40:1–2; 49:14–50:3; 54:6–7; Zech 10:6 as well as in Lamentations, esp. 3:25–33. See also Boda, Severe Mercy (see n. 55), 82–85. 139  On this translation, see Hartley, Leviticus (see n. 51), 463; Milgrom, Leviticus 23–27 (see n. 39), 2343–2346. 140  Contrast Milgrom, Leviticus 23–27 (see n. 39), 2299–2303, who maintains that ‫משכן‬ in 26:11 should be translated “presence” with N. K iuchi , Leviticus (AOTC; Nottingham:

Patterns of Descriptive Curse Formulae in the Hebrew Bible

971

and be their God (v. 12). However, if the people steadfastly refuse to obey, Yahweh will execute covenant vengeance (v. 25)141 and unleash horrendous suffering (vv. 27–39). By contrast, should the people turn back to Yahweh in contrition, confession, and humility and commit themselves to renewed obedience (40–43), Yahweh will not break (‫ )פרר‬his covenant but remember it and restore his people (v. 44–45). Moreover, it is not ultimately Israelʼs confession, repentance, humility of heart, and making amends (vv. 40–41, 43)142 nor even the landʼs enjoyment of its Sabbaths (v. 43) that secures the restoration of Israelʼs relationship with Yahweh. Rather, it is Yahwehʼs remembering of the covenant with the ancestors (v. 42) and with the exodus generation (v. 45) that ultimately effects such restoration.143 In contrast to its central position in Lev 26, then, the concept of covenant in Amos is ambiguous. Indeed, the question whether, where, or to what degree the idea of covenant (the Sinai covenant in particular) is to be presupposed in this chapter has been one of the most divisive issues in the study of the book.144 For some, the mere use of curse vocabulary implies covenant and should be understood as implicit in the words of Amos.145 Others take a more skeptical view.146 It must be admitted that stock Deuteronomic phraseology is not abundant147 and that though some communal standards, perhaps reflective of those in the Covenant Code, may lie behind certain elements of the condemnations in Apollos, 2007) 477–478, and Hartley, Leviticus (see n. 51), 463–464, who assume it refers to the literal Priestly tabernacle, symbolic of Yahwehʼs presence. 141  ‫והבאתי עליכם חרב נקמת נקם ברית‬, i.e., the punishment due for covenant violation; thus H. G. L. Peels, The Vengeance of God: The Meaning of the Root NQM and the Function of the NQM-Texts in the Context of Divine Revelation in the Old Testament (trans. W. T. Koopmans; OtSt 31; Leiden: Brill, 1995), 103–109, who argues that the phrase here denotes “the fitting answer to contempt or damaging the covenant” (106). 142  The syntax of v. 41 is disputed. Nihan, “Priestly Covenant” (see n. 121), 107, n. 70, sees the particle ‫ או‬as introducing two alternate possibilities: either the people will confess their sin or their uncircumcised heart will be humbled. Milgrom, Leviticus 23–27 (see n. 39), 2232, sees it as an extension of the protasis “if they confess their sin,” adding the notion of sincere humility and repentance. Despite the syntactical difficulties involved in reading the particle ‫או‬ in this way (noted by Nihan), Milgromʼs rendering makes more sense in the context. However, with Nihan, and pace Milgrom, I see the lack of the conditional particle in v. 40 as indicating the certainty that all of these things will transpire. See further below. 143  Nihan, Priestly Torah (see n. 49), 540–542. 144  See the bibliography in M. D. C arroll R., Amos – The Prophet and His Oracles: Research on the Book of Amos (Louisville: Westminster John Knox, 2002), 126, and the survey of opinions in Barton, Theology of the Book of Amos (see n. 73), 103–06. 145  Thus Brueggemann, “Amos IV” (see n. 79). 146  Barstad, Religious Polemics of Amos (see n. 100), 66–67. 147  Even in those passages that are seen as the result of Dtr redactional activity, for which see Wolff, Joel and Amos (see n. 53), 112–113, typical Dtr vocabulary is not abundant; see M. Weinfeld, Deuteronomy and the Deuteronomic School (Oxford: Oxford University Press, 1972), and Driver, Deuteronomy (see n. 39), lxxvii–lxxxiv.

972

John Kessler

Amos,148 the term ‫ ברית‬is never mentioned in the book. Nevertheless, one should not be entirely skeptical. Covenantal ideas may indeed be present in Amos but in less explicit or developed forms.149 However the matter is decided, the role of covenant in Amos is veiled at best, in contrast to its preeminent position in Lev 26. Sixth, the degree and intensity of the peopleʼs suffering differ between the two texts. The plagues of Amos 4 involve serious suffering, but they nevertheless allude to the more common and repeated experiences of national life in the ancient Near East: famine, drought, pestilence, and military defeat. The plagues of Lev 26 go much further and reach a horrific crescendo in vv. 27–39: starvation, cannibalism, exile, dispersion among the nations, and ultimate destruction. Seventh, Amos 4 is open-ended, whereas Lev 26 moves toward an ultimate resolution. As we have seen above, Amos 4:12 portrays the nation at the cusp of a future choice – one that will determine its future before God. By contrast, Lev 26:40–45 (or according to some 39–45),150 describe Israelʼs future restoration in great detail. The degree of definiteness in this description is disputed due to the problematic syntax of vv. 40–41.151 The issue at this point involves the lack of the conditional particle ‫ אם‬before v. 40 and the terms ‫ או אז‬in v. 41. Some, such as Milgrom, take this as the long protasis of a condition continuing to the end of v. 41 (as in the NRSV “if they confess [. . .] if then their uncircumcised heart is humbled, and they make amends for their iniquity”).152 Nihan, by contrast, sees the passage as taking a more prophetic tone.153 Yahweh sets out Israelʼs future in  Wolff, Joel and Amos (see n. 53), 168, identifies the words of condemnation in Amos 2:8 as an ad hoc reflection of familiar abuses rather than an allusion to the Covenant Code. R. J. C oggins , Joel and Amos (NCB; Sheffield: Sheffield Academic, 2000), 118–120, esp. 121, urges caution regarding “a specifically covenantal reading in view of the lack of berith language in Amos. We have noted a number of points at which there are close similarities in outlook between Amos and the Deuteronomistic History. We need to bear in mind that there are important differences, and the lack of direct reference to the peopleʼs covenant status is one such.” See B. M. Levinson, “Is the Covenant Code an Exilic Composition? A Response to John Van Seters,” in In Search of Pre-Exilic Israel (ed. J. Day; JSOTSup 406; London: T&T Clark, 2004), 272–325. However the matter is decided, the practices condemned in Amos 2:8 seem analogous to the proscriptions in Exod 22:25–26 (ET 22:26–27); see A ndersen and Freedman, Amos (see n. 48), 319–320. 149  Even Barstad, Religious Polemics of Amos (see n. 100), 66–67, who doubts that covenant plays a major role in the preaching of Amos, would not deny that certain characteristically Deuteronomistic motifs were not created de novo by the Deuteronomists but were inherited and developed by them. 150  See Fischer, “Need for Hope” (see n. 39), 369–379. 151  Gerstenberger, Leviticus (see n. 36), 430, suggests that a scribal insertion has occurred here, disrupting the flow of the sentence. 152  Milgrom, Leviticus 23–27 (see n. 39), 2332. 153  N ihan , Priestly Torah (see n. 49), 545, views Lev 26:40–45 as “a prophetic oracle, stated by Yahweh himself” that “reveals to Israel all it needs to know about its future.” This 148

Patterns of Descriptive Curse Formulae in the Hebrew Bible

973

great detail, stating “they will confess.” Whichever position one takes, it is clear that Lev 26 presents a far more defined picture than Amos 4. The primary point in Lev 26:40–45 is that even if the nation walks in hostility to Yahweh, a way of return will surely be made available to it. Eighth, of the two texts only Lev 26 reflects upon the theological issues raised by Yahwehʼs judgment upon Israel and the possibility that that relationship might be severed altogether. For the Holiness writers, questions of theodicy, the divine character, and Yahwehʼs fidelity to his sworn covenant are of critical importance. Such questions, however, are absent in Amos 4:6–12.154 Despite their differences as to which covenant(s) are in view in Lev 26, both Stackert and Nihan (and the line of argumentation each represents) are agreed that, in Lev 26, the covenant can be broken from the human side and human offenders cut off from it but never, ultimately (or, perhaps better, only potentially), from the divine side.155 To do so would constitute a theological impossibility, at least from the perspective of Lev 26:44 and from the more general Priestly understanding of Yahwehʼs covenant.156 In Lev 26, the statement that Yahweh will not break (‫ )פרר‬his covenant is grounded in the statement “I am Yahweh their [or your]157 God” (26:44b–45).158 Thus, whereas both Amos 4 and Lev 26 foresee cataclysmic judgments for the nation, only in Lev 26 do we see sustained reflection on the conundrum posed by election and covenant, on the one hand, and devastating judgment, on the other. In sum, a comparison of these two passages demonstrates that even texts bearing close resemblances in their use of descriptive curse formulae may nevertheless display marked differences at a deeper level.159 position is also taken by Boda, Severe Mercy (see n. 55), 83; see especially his discussion of the syntax of vv. 40–41 in n. 23. Gerstenberger, Leviticus (see n. 36), 430, takes the same view, maintaining that v. 40 originally read “and they will confess” before a later scribal insertion. I find the position of Nihan and Boda to be more convincing than that of Milgrom. 154  To be sure, the hymnic elements in v. 13 dwell upon the divine character, but these cannot be discussed here. 155  Stackert, “The Holiness Legislation” (see n. 132); Nihan, Priestly Torah (see n. 49), 535–541; idem, “Priestly Covenant” (see n. 121), 104–115, 535–541. 156  On the reconciliation of the Priestly and non-Priestly perspectives in Lev 26, see Nihan, Priestly Torah (see n. 49), 542; idem , “Priestly Covenant” (see n. 123). On the concept of a ‫ברית עולם‬, see also O lyan , “Status of the Covenant” (see n. 31); S. D. M ason , “Eternal Covenant” in the Pentateuch: The Contours of an Elusive Phrase (LHBOTS 494; New York: T&T Clark, 2008). 157  See the commentaries on the text-critical issue here. 158  A Priestly idiom; see Exod 6:7; 16:12; 29:46; Lev 11:44; 20:7; 24:22; 25:17; 26:1, 44. 159  One further difference distinguishes the two texts, although it is only a slight one and involves a number of subtle exegetical decisions. The two texts display some dissimilarity regarding the increment in intensity with which the curses are portrayed. It is generally agreed that Lev 26 depicts a steep increment in the number and severity of the maledictions imposed. Indeed, the structure of the chapter is sometimes seen to reflect a Plagensteigerungsschema –

974

John Kessler

4 Amos 4 and Leviticus 26: Questions of Sequence and Ideological Relationship The preceding discussion, especially its traditiohistorical aspects, may shed light on the literary and ideological relationship between Lev 26 and Amos 4, especially the direction of influence (if any may be posited) between the two texts. It may also have implications for other issues such as the dating of various pentateuchal sources and traditions as well as the relationship between the prophetic and pentateuchal materials more generally. It would be beyond the scope of this paper to attempt an absolute dating of these two passages. Only a discussion of the relative dating and flow of influence between them is possible here. In this regard, four major positions emerge. First, there are those who see Amos 4 as subsequent to and dependent upon Lev 26. Included in this group are those suggesting an early Lev 26 and a late eighth-century Amos 4160 and others who suggest a late monarchic Lev 26 and an exilic (or later) Amos 4.161 A second position maintains that Amos 4 precedes Lev 26. According to this view, Amos 4 an increasing imposition of curses; on this see Nihan, Priestly Torah (see n. 49), 537, n. 561. This stands in contrast to Amos 4:6–12, where a degree of increment certainly exists but follows a far less uniform pattern. It is difficult to see much sense of increase in severity in the famine of v. 6, the partial drought of v. 7, the search for water in v. 8, the blight and mildew of v. 9a, and the locust decimation of v. 9b. In 4:10, we do encounter a significant increment. Here, for the first time, we read of military disaster. In 4:11, Yahweh acts more violently, signaled by the use of ‫הפך‬, “to overthrow.” However, Yahweh does not overthrow the nation as a whole, but only “some of you” (‫)בכם‬. Andersen and Freedman, Amos (see n. 48), 444, suggest that “the partitive use of b- is unusual, but it seems to be the only plausible explanation here. [. . .] As with the second plague the disaster was partial and selective. [. . .] From the almost total conflagration only a bit of charred wood was rescued.” Thus Yahweh leaves a remnant, “like a brand snatched from the fire” (4:11). See also Hadjiev, Composition and Redaction (see n. 95), 152–155. To be sure, the potential judgment implied in v. 12 may be severe and, as such, Amos 4 reaches a similar crescendo to that in Lev 26. In Amos 4, however, no precise description of that ultimate judgment is given. In sum, though admittedly not strictly incremental in every detail, the overall movement in Lev 26 is one of a steadily increasing degree of suffering to intolerable levels. G erstenberger , Leviticus (see n. 36), 413, sees “no ascending order” in Amos 4, as opposed to Lev 26. Others demur: A msler , “Amos” (see n. 77), 199, speaks of “une certaine gradation”; Paul , Amos (see n. 22), 143, following J. L. Mays, Amos: A Commentary (OTL; London: SCM, 1969), 80–81, speaks of a “ ‘bolerolike’ iteration.” I would thus suggest that a slight difference exists between the two passages at it relates to the degree of intensity with which the curses are imposed. 160  Thus Hartley, Leviticus (see n. 52), 461–462; Milgrom, Leviticus 23–27 (see n. 39), 2363–2364; Brueggemann, “Amos IV” (see n. 80). However, see below on Milgromʼs view regarding Lev 26:33b–35 and 43–44 as later interpolations. 161  Jeremias, Hosea und Amos (see n. 59), 198–213; idem, Book of Amos (see n. 59), 69–72; idem , “Amos 3–6: From the Oral Word to the Text,” in Canon, Theology, and Old Testament Interpretation: Essays in Honor of Brevard S. Childs (ed. G. M. Tucker et al.; Philadelphia: Fortress, 1988), 217–229.

Patterns of Descriptive Curse Formulae in the Hebrew Bible

975

is either a mid- to late monarchic redaction or composition, preceding a post-587 Lev 26,162 or an exilic text, whose form is reworked in Lev 26 in the postexilic period.163 A third view suggests that the two texts appeared at roughly the same time but each in a different milieu and containing a different ideology. Wolff, for example, opines that Amos 4 and Lev 26 emerged in the late monarchic period. Leviticus 26 grew out of the liturgy of the great autumn festival, whereas Amos 4 contains the word of a prophet in the Amosian tradition, decrying the worship of foreign deities at Bethel. Wolff appears to suggest that Amos 4 has been influenced by Lev 26. However, it is difficult to determine Wolffʼs understanding of the degree of this influence.164 Fourth and finally, there are those who view Lev 26 as a composite work, reflecting an ongoing process of development,165 leaving open the possibility that the relationship between it and Amos 4 may have evolved over time. For example, it might be argued that the two texts stood in a looser relationship to each other in an earlier period, and then a greater connection existed between the two at a later time. In response to these proposals, I would make the following preliminary observations. First, the lexical and formal elements common to these two texts are in and of themselves insufficient indicators of influence or dependence. The curse formulae in them belong to common ancient Near Eastern stock, so certainty cannot be based on form or vocabulary alone. However, the opposite position, that of a complete lack of any connection between the two (or only a very loose one) is equally problematic in light of their shared structural forms and motifs, the organization of the maledictions into five groups and their incremental progression, the numerical structure around five and seven, and their disciplinary function intended to produce return or repentance.166 Thus it seems beyond question that the two passages stand in some real literary relationship to each other.167 But what might that relationship be, and what implications may be drawn from it?  Thus Hadjiev, Composition and Redaction (see n. 95), 149–152, 161.  Gerstenberger, Leviticus (see n. 36), 413. So also Graupner and Fabry, “‫( ”שׁוב‬see note 93), 485, who state, “If some connection does remain, however, it is easier to see that the late text in Lev. 26 was influenced by Am. 4:6–11.” See also C. N ihan , “The Holiness Code between D and P: Some Comments on the Function and Significance of Leviticus 17–26 in the Composition of the Torah,” in Das Deuteronomium zwischen Pentateuch und deuteronomistischem Geschichtswerk (ed. E. Otto and R. Achenbach; FRLANT 206; Göttingen: Vandenhoeck & Ruprecht, 2004), 81–122, esp. 108. Nihan sees Amos 4 as exilic or postexilic yet nevertheless prior to Lev 26. 164  Wolff, Joel and Amos (see n. 53), 214, 244–245. 165  For various such approaches, see below. 166  See above for a more detailed demonstration of these features and the similar arguments of Blum, “Jesaja und der ‫ דבר‬des Amos” (see n. 51), 91, regarding the relationship between Amos 4 and Isa 5, 9–10. 167  See the discussion in Gerstenberger, Leviticus (see n. 36), 413. 162 163

976

John Kessler

Of the four positions just noted regarding the relationship between Amos 4 and Lev 26, various factors suggest that either the second proposal (Amos 4 precedes Lev 26) or the fourth one (Amos 4 is in a changing relationship to a growing Lev 26) is most likely. Let us examine these two possibilities in turn. If Lev 26 is viewed as a unified composition (for example as a Priestly epilogue to Lev 17–25168 or the book of Leviticus as a whole), various factors suggest that it is best dated after Amos 4.169 First, if Amos 4 is the later text, how should one account for the absence of any distinctly Priestly elements in it? Why would the Amosian tradents have taken up a manifestly Priestly text only to gut it of its most distinctive Priestly character?170 It is only with difficulty that Amos 4 could be read as a conscious revision or correction of, or alternative to, Lev 26. It is likewise difficult to see why Amos 4 would deliberately distance itself from an ideology of restoration through confession of sin, the humbling of oneʼs self, the making of amends, and the circumcision of the heart,171 all themes that are in evidence from the late sixth century onward, especially in the penitential-prayer settings of Ezra 9; Neh 9; Dan 9; and 2 Chron 36:20–21 (all later texts), leaving only a return-based (‫ )שוב‬focus, characteristic of prophetic preaching, particularly in evidence between the eighth and fiftth centuries.172 Milgrom sees this difference in response patterns between these two texts as decisively establishing the priority of Lev 26, but this difference seems rather to indicate movement in the reverse direction.173 Moreover, the ideology of returning to Yahweh is highly compatible with the Priestly notions of repentance and confession of sin and with the idea of the circumcision of the heart, as Deut 30:2, 6, 10 well attest. Second, Lev 26 reflects carefully on the relationship between Yahwehʼs  Joosten, People and Land (see n. 51), 6.  Thus G erstenberger , Leviticus (see n. 36). See further the insightful discussion in Hadjiev, Composition and Redaction (see n. 95), 148–150, 159–161. Many of his conclusions will be referenced in this section. 170  On Lev 26 as a distinctly priestly text, see N ihan , Priestly Torah (see n. 49); idem , “Priestly Covenant” (see n. 123); O lyan , “Status of the Covenant” (see n. 31); S tackert , “Holiness Legislation” (see n. 132); idem, “Distinguishing Innerbiblical Exegesis.” 171  For the vocabulary here, see nn. 116–118 above. Note especially its frequent use in exilic and postexilic material. 172  See, e.g., the discussion of Hos 14:2–4[1–3] in A. A. M acintosh , A Critical and Exegetical Commentary on Hosea (ICC; Edinburgh: T&T Clark, 1997), 558–564; See also Barstad, Religious Polemics of Amos (see n. 100), 65; Holladay, Root Šûbh (see n. 106). 173  Milgrom, Leviticus 23–27 (see n. 39), 2348, asserts that the notions of confession and repentance in Lev 26 are “embryonic [. . .] older [. . .] and inchoate.” He suggests that the prophets viewed such religious sentiments as “inadequate” when compared to concrete, active change. This, however, is to assume that the elements of contrition referred to in Lev 26 are superficial and do not lead to such change. This is certainly not the presentation of matters in Ezra and Nehemiah (Ezra 10:1–5; Neh 9:38–10:39; 13:1–30), where the community undertakes concrete remedial measures following declarations of penitence, no matter how problematic its determination to enact its repentance may seem to contemporary readers. 168

169

Patterns of Descriptive Curse Formulae in the Hebrew Bible

977

judgment of his covenant partner due to the latterʼs faithlessness and Yahwehʼs character as faithful to his own promises (26:44), a dimension completely absent in Amos 4. If Amos 4 were the later text, such an astonishing omission would require some explanation. Furthermore, why would the Amosian tradents efface the elements of covenant (Lev 26:9, 15, 25, 42, 44–45) and Sabbath (Lev 26:34–35, 43) from consideration in Amos 4, especially if Amos 4 is seen as a post-exilic composition? Finally, Amos 4 does not reflect those elements of Lev 26 that are most characteristic of an exilic setting and that feature prominently in Lev 26, such as war, enemies, captivity, and dispersion. If Amos 4:6–12 is exilic or later, why are the elements of wholesale defeat and exile, which are present in Lev 26, eliminated from it?174 It would appear, then, if Lev 26 is viewed as having undergone no literary development, Amos 4 is best understood as antedating it. However, if one admits the possibility of literary growth in Lev 26, the picture becomes a good deal more complex. One approach is to question whether Lev 26 originally included vv. 34–35 and 40–45.175 It is quite striking that, devoid of these verses, Amos 4 and Lev 26 are highly similar: they share the same type of curse pattern and a similar structure (the organization of the maledictions into five groups in an incremental progression, including moments of reprieve, leaving room for repentance), both use the broader numerical structure of five and seven, and both view Yahwehʼs maledictions as having a disciplinary function intended to produce change in the peopleʼs comportment.176 What is more, both envisage a steady descent into darkness as well as a decided refusal to learn from adversity and to recklessly pursue rebellion, and both end in disaster.177 Thus, when Lev 26 is viewed without vv. 34–35 and 40–45, it is difficult to detect any solid indications of the direction of influence between it and Amos 4. The plagues in both are drawn from the traditional biblical and ancient Near Eastern stock, and references to any specific events are difficult, if not impossible, to come by.178 Thus the direction of influence could flow either way. Moreover, despite the optimism of some regarding clear literary dependence between Lev

174  Hadjiev, Composition and Redaction (see n. 95), 150. Moreover, Hadjiev demonstrates that the bulk of the linguistic correspondences between Amos 4 and Lev 26 are centered in Lev 26:24–26, not in the chapter as a whole (151). 175  I will return, below, to the matter of whether these verses may simply be excised from the present text in order to recover its antecedent. 176  See above for a more detailed demonstration of these features and the similar arguments of Blum, “Jesaja” (see n. 51), 91, regarding the relationship between Amos 4 and Isa 5, 9–10. 177  See, however, the discussion of the meaning of Amos 4:12, above. 178  The various attempts to isolate certain elements in these texts and use them for purposes of dating have not achieved any wide consensus. Even the thorny questions of the ‫ מקדשיכם‬in Lev 26:31 or the reference to Sodom and Gomorrah in Amos 4:11 have elicited no agreement regarding dating.

978

John Kessler

26 and Amos 4 (however conceived),179 without vv. 34–35 and 40–45, it might equally be argued that both texts belong to a more generalized form or pattern concerning the perils of neglecting Yahwehʼs discipline.180 What is more, it is in Lev 26:34–35 and 40–45 that we encounter several elements not in evidence in either Lev 26:3–33, 35–39 or Amos 4. These elements include the following: (1) the Sabbath concept in vv. 34–35 and 43–44; (2) the Priestly vocabulary of sin, atonement, repentance, and requital in vv. 33–35 and 40–41; (3) reference to the covenants with the ancestors and the exodus generation in vv. 42, 45; (4) the fact that v. 40 presents Israelʼs return as certain (cf. Deut 4:29–30; 30:6–10);181 and (5) unlike Amos 4, Lev 26:3–33 and 36–39,182 and descriptive curse formulae in general, Lev 26:34–35 and 40–45 move away from the customary expressions of blessing and cursing and engage in theological reflection on subjects such as theodicy (including the Sabbath and the necessity for the landʼs desolation) and the nature of Yahwehʼs covenant (as ultimately indestructible). Turning now to position four, though numerous scholars see Lev 26 as a unified composition dating from some point during either the monarchic or Persian eras,183 many others see the text as containing one or more post-587 supplements.184 The principal arguments for the latter position are well known 179  Hadjiev, Composition and Redaction (see n. 95), 152, opines, “Amos 4 is probably the earlier text and the author of Lev 26 had it before him.” This may, however, overstate the case. 180  See the description of Amos 4 as a “negative historical retrospective” by C renshaw, “Liturgy” (see n. 51), 36, or as a parody of Ps 136, and thus a case of “Unheilsgeschichte” by Paul , Amos (see n. 22), 141. See also the rather apposite observation of Gerstenberger Leviticus (see n. 36), 413, who suggests that the “stepwise process of chastisement” of Amos 4 became a “popular schema in postexilic literature.” He concludes, “It may not be accidental that this litany, too, [i.e., Amos 4] exhibits five stages of chastisement [. . .] and that the individual motifs in part recur in Lev 26:14–33.” 181  See the discussion above, pp. 972–973. 182  In what follows, I will use Lev 26* to refer to an earlier version of Lev 26:3–45 and reserve Lev 26 for the final, reworked presentation of these verses, with the most distinctive elements of this reworking in vv. 34–35 and 40–41. 183  So Hartley, Leviticus (see n. 51), 456–462 (seeing Lev 26 as a monarchic text); Nihan, “Priestly Covenant” (see n. 121), 108; idem, Priestly Torah (see n. 49), 540, n. 573 (seeing it as postexilic); Fischer, “Need for Hope” (see n. 39), 378, also postexilic. 184  B. A. Levine, “Epilogue” (see n. 126), 19–20, following H. L. Ginsberg, sees the compositional history of the chapter as follows: (1) the primary epilogue in vv. 3–33a, 37b–38; (2) the first “post-catastrophe” addition in vv. 39–40b, 44–45, stressing the exile, confession of sin, and Yahwehʼs remembering of his covenant with the ancestors; (3) later interpolations in vv. 33b–37a, describing Israelʼs defeat and exile, the desolation of the land, and the landʼs Sabbaths. Levine views all three levels as exilic (25). Milgrom, Leviticus 23–27 (see n. 39), 2364, sees the core of Lev 26 as an eighth-century composition (including the themes of destruction and exile, here with reference to the Northern Kingdom) into which 33b–35 and 43–44 have been interpolated during the exile. B uis , “Deutéronome” (see n. 8), 139, sees vv. 36–45 as composed shortly before 587, at the same time as Jer 31 and Deut 30:1–10. Its

Patterns of Descriptive Curse Formulae in the Hebrew Bible

979

and cannot be rehearsed in detail here.185 They generally revolve around (1) the overall similarity of various curse texts that all end on a note of indeterminacy or uncertainty (Amos 4; Deut 28; Isa 5:25–29; 9:7[8]–10:4); (2) the fact that in Deut 28 and elsewhere in the prophets, texts that end in abject despair are attenuated to include elements of hope;186 (3) the introduction of the various motifs noted above, which are absent in Amos 4 and Lev 26:3–33, 36–39; (4) various allusions to the developing historical and ideological aspects of the exilic period;187 (5) the shift to the third person in vv. 36–37 and 40–45; and (6) the possibility that other prophetic texts may betray knowledge of a form of Lev 26 that does not include vv. 40–45.188 It is noteworthy that even those who maintain the unity of vv. 3–45 concede that v. 40 signals a change of perspective. Nihan, who views Lev 26 as a literary unity, nevertheless sees 40–45 as reflecting a “different, though not necessarily incompatible” perspective from the section that precedes it.189 Whether Lev 26 gives evidence of literary development is thus highly complex and perhaps not liable to a definitive solution. The present form of the text displays a high degree of literary cohesion, suggesting a careful and intentional composition. Thus the simple excision of vv. 33–34 and 40–45 on the basis of their content and ideology alone would appear somewhat arbitrary and tendentious. Nevertheless, the arguments presented above regarding the close similarities between Amos 4 and Lev 26 and the distinctive elements of vv. 33–34 and 40–45 suggest to me that the chapter may have had one or more literary antecedents that served as the basis of the present text. As such, Lev 26 may be seen as a reworking of certain earlier materials, and its relationship to addition to vv. 2–35 gave rise to the interpolation of the conditional elements in vv. 14 and 18. For other advocates of a more diachronic reading of Lev 26, see the bibliography in N ihan, Priestly Torah (see n. 49), 540, n. 573. See also Olyan, “Status of the Covenant” (see n. 31), for another diachronic perspective on vv. 40–45. 185  See especially the discussion in Levine, “Epilogue” (see n. 126). 186  On the Deuteronomic and Priestly traditions, see D. Rom-Shiloni, Exclusive Inclusivity: Identity Conflicts between the Exiles and the People who Remained (6th–5th Centuries BCE) (LHBOTS 543; New York: T&T Clark, 2013), 2. On Jeremiah and Ezekiel, see idem, “Deuteronomic Concepts of Exile Interpreted in Jeremiah and Ezekiel,” in Birkat Shalom 1: Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Judaism presented to Shalom M. Paul on the Occasion of his Seventieth Birthday (2 vols.; Winona Lake, IN: Eisenbrauns, 2008), 1:101–123. On the additional elements of hope in Amos, see B arton , Theology of the Book of Amos (see n. 73), 122–125. On Isaiah, see especially H. G. M. Williamson, The Book Called Isaiah: Deutero-Isaiahʼs Role in Composition and Redaction (Oxford: Oxford University Press, 1994); idem, “Isaiah, Book of,” in Dictionary of the Old Testament: Prophets (see n. 107), 364–378. 187  Levine, “Epilogue” (see n. 126). 188  See the excellent discussion in A. K opilovitz , “What Kind of Priestly Writings Did Ezekiel Know?,” in the present collection of colloquium papers, with bibliography. 189  Nihan, “Priestly Covenant” (see n. 121), 106; see also Fischer, “Need for Hope” (see n. 39), 372–375, and the bibliography cited there.

980

John Kessler

Amos 4 may be conceived of in two (or more) stages. In an earlier stage, the two texts may have shared a more general similarity, perhaps due to each oneʼs use of a more generally known literary form describing failure to learn from divinely sent adversity. The differences between them could be explained by the theological or ideological rooting of Amos and his school, on the one hand, and of the Priestly tradents, on the other.190 At a subsequent moment, another Priestly writer (or writers) reflected on the ideology of the specific descriptive curse pattern in evidence in Amos 4 and Lev 26*191 in light of later historical realities and ideological developments,192 reworking the text and incorporating their reflections into it. However the question of the composition history of Lev 26 is decided, I find it entirely plausible that within Lev 26 we may find reformulations or rereadings intended to provide a theological counterpoint193 or supplement to either Amos 4 alone or to both Amos 4 and Lev 26*.194 The Holiness writers responsible for its final form may have used the five-stanza structure and sevenfold repetitions of the Amos material as a model in Lev 26. Alternatively, such a general structure was more broadly utilized and may have shaped Lev 26* from its earliest stages, independent of Amos 4. Either way, the distinctive elements and features in Lev 26:34–35 and 40–45 noted above likely served to nuance and to add further considerations to those already present in the earlier material it presupposes. Two areas are of special significance in this regard. The first is the concept of covenant, implicit at best in Amos but central in Lev 26:3–45. As we have seen above, considerable debate exists regarding the covenant concept in Lev 26: Is it consistently Priestly (Stackert)? Does it involve the assimilation of certain Deuteronomic elements within a broader Priestly framework (Nihan)?195  This would account for the explicit mention of covenant in Leviticus and its absence from Amos. 191  I will use the designation Lev 26* to refer to any literary antecedents to the chapter in its present form that may have been used by those responsible for its current configuration. 192  Rom-Shiloni, Exclusive Inclusivity (see n. 186), 2, aptly comments, “Preexilic Priestly and Deuteronomic conceptions of exile do not recognize the notion of a partial deportation. [. . .] The ongoing existence of the Temple, the royal court, and daily life in Jerusalem stood in puzzling contrast to these understandings.” 193  On the concept of “theological counterpoint,” see J. L évêque , “Le contrepoint théologique apporté par la réflexion sapientielle,” in Questions disputées dʼAncien Testament: Méthode et théologie (ed. C. Brekelmans; BETL 33; Leuven: Leuven University Press, 1974), 183–202. 194  Thus Hadjiev, Composition and Redaction (see n. 95), 152. 195  See n. 155. I am persuaded by Nihanʼs arguments for a merging of covenantal concepts in Lev 26, rather than a uniquely Priestly perspective, yet this need not be an indication of an exilic or Persian-period date for Lev 26*. Dialogue between the two traditions could have begun at an earlier period; see the early dating of H by Joosten, People and Land (see n. 51), 200–207, and others. Alternatively, one could assert, as does Stackert (“Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction” [see n. 133]), that Lev 26* reflects Priestly 190

Patterns of Descriptive Curse Formulae in the Hebrew Bible

981

The notion of a covenantal bond between Yahweh and Israel lies at the heart of Lev 26 as a whole: Israel must obey Yahwehʼs statutes so that his ongoing presence might continue with them.196 Yet even if Israel should betray the terms of the covenant and despise (‫ )גאל‬Yahwehʼs statutes (26:15), engendering Yahwehʼs corresponding abhorrence of his people (26:30), this would not end the relationship. After a provisional time of rejection, marked by intense suffering but culminating in confession, repentance, and humility of heart,197 Yahweh would renew the relationship once again and not abhor his people to the point of ultimate destruction (26:44b). Thus 26:44 clarifies and determines the scope and extent of the abhorrence alluded to in 26:30.198 As such, in Lev 26 the covenant is “the single enduring reality.”199 Such an underlying bond, transcending even severe violations, is absent from (or at best hinted at in) Amos 4 but loudly proclaimed in Lev 26:40–45. The second area of counterpoint lies in the different approaches in evidence in Amos 4, Lev 26*, and Lev 26 (final) to issues of judgment, suffering, and national renewal. Shalom Paul has suggested that Amos 4 demonstrates “the futility of suffering.”200 Such a conclusion depends entirely upon reading 4:12 as implying no further hope. However, it is more likely that 4:12 is deliberately

conceptions of covenant which had always contained some element of conditionality, but that the idea of a (potential) definitive break (as in the Deuteronomic perspective) is introduced only in 40–45, esp. v. 44. Thus Lev 26* could envisage a time when Yahweh would abhor (‫)געל‬ Israel (26:30) and Israel would abhor (‫ )געל‬Yahwehʼs ordinances and spurn (‫ )מאס‬Yahwehʼs statutes (26:15, 43) yet remain within a consistently Priestly framework. The idea that such actions could lead to a definitive breach would be in evidence only in the later reflection and integration of disparate theological traditions contained in 26:44. 196  Significantly, although Lev 26:44b affirms that Yahweh will never break his covenant (‫)פרר‬, our chapter never uses the terminology of “eternal covenant” found elsewhere (Gen 9:16; 17:7, 13, 19; Exod 31:16; Lev 24:8; 2 Sam 23:5; Isa 55:3; 61:8; Jer 32:40; 50:5; Ezek 16:60; 37:26; Ps 105:10; 1 Chr 16:17). On the “eternal covenant,” see, among many others, M ason , Eternal Covenant (see n. 156); N ihan , “Priestly Covenant” (see n. 121); and especially S. M. O lyan , “An Eternal Covenant with Circumcision as Its Sign: How Useful a Criterion for Dating and Source Analysis?” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; Tübingen: Mohr Siebeck, 2011), 347–358, and idem , “Status of the Covenant” (see n. 31), and the literature cited therein. 197  See the texts cited in n. 137. 198  Such a clarification constitutes a further strong argument for the clarifying role of the later rereadings of Lev 26* reflected in 40–45. 199  The comments of Buis, “Lʼavenir de lʼalliance” (see n. 65), 134, on this point are highly insightful: “[A]u moment où le peuple est sur le point de disparaître, on voit quʼune réalité subsiste: lʼalliance.” 200  Paul , Amos (see n. 22), 143, although he may have overstated the case. If Amosʼs oracles have been edited in the south or have become the basis for the preaching of Isaiah in Isa 5, 9–10 (thus Blum, “Jesaja” [see n. 51]), then the reception of the Amosian material would indicate an ongoing belief in the capacity of suffering to produce change.

982

John Kessler

ambiguous and presents the future as open-ended.201 In the Amosian presentation (as in Isa 5:25–29; 9:7[8]–10:4), affliction has thus far had no effect. Failure to learn from it has left the nation on the brink of disaster. Yet, if it is accepted that Amos 4:12 may suggest the possibility of reprieve, the warning conveyed by past sufferings may still be heeded. In a sense, Amos 4 and Lev 26 engage related but distinct questions. Amos 4 underlines that rejection of divine discipline leads to doom and that, by implication, the Northern Kingdomʼs downfall was ultimately its own fault. It had been warned repeatedly, but to no avail. In Lev 26, however, the issue is not Israelʼs ultimate responsibility for its own calamities. Rather, Lev 26 focuses on the reality that sustained and intense suffering can bring about ultimate remediation and the basis on which it can do so. The two texts agree that suffering should lead to reformation of life. However, in contrast to the indeterminacy in Amos 4 (and the potential failure of suffering to avert disaster), in Lev 26, the Holiness writers affirm that suffering can – and indeed will – do so, provided that the suffering is sufficiently great and that it is allowed to do its work to the fullest.202 Indeed, it would appear that for the Holiness writers, it is only out of such affliction, including the landʼs complete devastation and abandonment, that true change can come.203 As noted, whereas in Amos, Israel must turn or return to Yahweh, for the Holiness writers, an additional, more profound and internal dimension was needed, one that would give rise to such a life reformation. In addition to a renewed commitment to Yahwehʼs statutes (implied in 26:40–41, 43, and stated in 46), the more internal dimensions of repentance, confession, humility, and circumcision of the heart were needed (26:40–41).204 Moreover, Lev 26 goes on to declare that even such repentance and confession do not in and of themselves secure the nationʼs future. Rather, Lev 26:42, 44–45, through reference to both the ancestral and Sinai covenants,205 provides an explanation of how it was that the nationʼs repeated high-handed rejections (and the consequences thereof), which are enumerated in Amos 4 and Lev 26* and normally unforgivable in Priestly thought (cf. Num 15), did not spell its total demise: Yahwehʼs covenant faithfulness is ultimately given the last word. To conclude this section, then, one can say that, however the matter of the composition history of Lev 26 is decided, in Amos 4 and Lev 26 we see both a  As noted above and by comparison with Isa 5:25–29; 9:7[8]–10:4.  The linking of intense suffering as the basis for renewal is apparent in the resumption of “their iniquities and the iniquities of their ancestors” (v. 39) in v. 40b. 203  Similar perspectives of total alienation and return are present in texts such as Hos 3:5; 5:25; Isa 40:1–2; 2 Chr 36:20–21. 204  See the excellent discussion in Fischer, “Need for Hope” (see n. 39), 381–385, regarding the alternative resolutions to the nationʼs obduracy presented in Lev 26 and Deut 28–30, especially around the matter of the circumcision of the heart. 205  On the conflation of these covenants, see N ihan , “Priestly Covenant” (see n. 121), 111–112; Hartley, Leviticus (see n. 51), 469–471. 201 202

Patterns of Descriptive Curse Formulae in the Hebrew Bible

983

particular ideological framework into which these descriptive curse formulae are set (adversities sent as incentives for change) as well as, in Lev 26:34–35 and 40–45, the theological reflections upon and refinements of that ideology. Both of these dimensions reveal significant aspects of thought concerning Israelʼs history and identity and most especially the dynamics involved in its relationship to Yahweh.

5 Conclusions and Implications Several implications emerge from the preceding analysis. First, if my arguments are sound, this study has demonstrated the variety of ways in which Israelite texts have creatively reconfigured the traditional stock of ancient Near Eastern curse vocabulary for use within various theological streams and traditions. This then underlines the need to ascertain the orientation and purpose of a given text before advancing broad hypotheses regarding the significance and function of any curse formula within it. Form-critical judgments alone are not sufficient when dealing with such curse materials. Second, as we have seen, there are several broader patterns into which descriptive curse formulations may fall. In the present instance, Amos 4; Lev 26; and Isa 5:20–25; 9:7[8]–10:4 historicize the more static pattern of sin-consequence or interdiction-deterrent found elsewhere. In doing so, they integrate the concepts of benediction and malediction with the idea of Israelʼs lived experience, stretched out over time, and the nationʼs sufferings as Yahwehʼs discipline and instruction.206 Moreover, this integration of blessing and cursing with lived experience enables the writers of these texts to view Yahwehʼs maledictions as challenges that put the nation to the test: Will it choose submission and blessing or rebellion and curse? Third, significant differences of perspective may appear even between texts belonging to the same general curse pattern. For example, in the present study, a careful analysis of the differences between Lev 26 and Amos 4 has revealed fundamentally different understandings at numerous key points, especially of the role of suffering in producing change, the way in which such change will be evidenced, and the basis of Israelʼs ultimate hope for restoration. Thus, when considering the significance of curse language in any given context, one must move beyond commonalities of forme and be attentive to differences of fond. Texts displaying similar formal elements may intentionally deepen, revise, or correct those upon which they have been patterned. Fourth, and finally, the fact that one or more prophetic texts (Amos 4; Isa 5:25–29; 9:7[8]–10:4) use a descriptive curse pattern strikingly similar to the one found in Lev 26 suggests that consideration of the literary growth of the 206

 See n. 180.

984

John Kessler

Pentateuch cannot be undertaken in isolation from the prophetic corpus. The prophetic materials, which so frequently display strong intertextual relationships with numerous pentateuchal texts, must play a significant role in pentateuchal analysis. Since the inception of modern biblical criticism, the prophetic materials have been seen as a foundational element in addressing issues of the literary development of the Pentateuch. The vitality of the scholarly literature addressing the relationship between these two bodies testifies to the continuing importance of the discussion. Failure to address ongoing developments in the study of the prophetic materials can only impoverish pentateuchal study, whereas attention to the interaction between the two bodies of literature can only enrich it.

Reading Zechariah 9–14 with the Law and the Prophets Sibling Rivalry and Prophetic Crisis Mark J. Boda The study of innerbiblical allusion in Zech 9–14 has focused most of its attention on the influence of other prophetic books, especially Isaiah, Jeremiah, Ezekiel, and at times Zech 1–8.1 Connections to the Torah have not been prominent in this area of research, except for some championing of the influence of Deuteronomy.2 The present article focuses attention on two places in Zech 9–14 (Zech 9:11 and 13:4–5) that suggest reliance on a similar motif within the book of Genesis, namely, the sibling-rivalry tradition. Accompanying these uses of the sibling-rivalry tradition in Genesis, and bolstering their status as a common strategy, is a parallel use of prophetic traditions drawn from what are now called 1  For example, B. S tade , “Deuterosacharja: Eine kritische Studie I,” ZAW 1 (1881), 1–96; idem, “Deuterosacharja: Eine kritische Studie II,” ZAW 2 (1882), 151–72; idem, “Deuterosacharja: Eine kritische Studie III,” ZAW 2 (1882), 275–309, focused on the prophetic traditions of Jeremiah and Ezekiel; M. D elcor , “Les sources du Deutero-Zacharie et ses procédés dʼemprunt,” RB 59 (1952), 385–411, on Ezekiel, Jeremiah, Trito-Isaiah, Job, and Joel; R. A. Mason, “The Use of Earlier Biblical Material in Zechariah 9–14: A Study in Inner Biblical Exegesis,” in Bringing Out the Treasure: Inner Biblical Allusion and Zechariah 9–14 (ed. M. J. Boda and M. H. Floyd; JSOTSup 304; Sheffield: Sheffield Academic, 2003), 1–208, on the major prophets, especially Deutero- and Trito-Isaiah and Zech 1–8; I. W illi -P lein , Prophetie am Ende: Untersuchungen zu Sacharja 9–14 (BBB 42; Köln: P. Hanstein, 1974), on the major prophets and Hosea; K. R. Schaefer, “Zechariah 14 and the Formation of the Book of Zechariah” (S.S.D. diss., École biblique et archéologique française, 1992), on Zech 1–13, Jeremiah, and Ezekiel; R. F. Person, Second Zechariah and the Deuteronomic School (JSOTSup 167; Sheffield: JSOT, 1993), on Deuteronomic literature; N. H. F. Tai , Prophetie als Schriftauslegung in Sacharja 9–14: Traditions- und kompositionsgeschichtliche Studien (Calwer theologische Monographien, Reihe A, Bibelwissenschaft 17; Stuttgart: Calwer Verlag, 1996), on Jeremiah, Ezekiel, Deuteronomic literature, and Hosea; R. Nurmela, Prophets in Dialogue: Inner-Biblical Allusions in Zechariah 1–8 and 9–14 (Åbo: Åbo Akademi University, 1996), mostly on Isaiah (1–11 and 29–31 but not 40–55), Jeremiah, Ezekiel, Zech 1–8. 2  For connections to Deuteronomy, see especially P erson , Second Zechariah and the Deuteronomic School (see n. 1). Those that have placed more emphasis on Genesis are K. J. L arkin , The Eschatology of Second Zechariah: A Study of the Formation of a Mantological Wisdom Anthology (CBET 6; Kampen: Kok Pharos, 1994), and S. Y. Lee , An Intertextual Analysis of Zechariah 9–10: The Earlier Restoration Expectations of Second Zechariah (LHBOTS 599; London: Bloomsbury, 2015).

986

Mark J. Boda

the Former and Latter Prophets, with attention given to a ninth-century (Elijah), eighth-century (Amos), and a seventh–sixth-century (Jeremiah) prophet. After laying out the evidence that Zechariah used these earlier traditions, the article will look at the appropriateness of allusion to this tradition within the literary horizon of both source (Genesis, Prophets) and host (Zech 9–14) texts.

I. Innerbiblical Allusion in Zechariah 13:4–5 and 9:11 A. Zechariah 13:4–5 Zechariah 13:4–5 ‫׃‬ ‫יאים ִאיׁש ֵמ ֶחזְ י ֹנֹו ְב ִהנָ ְבאֹתֹו וְ לֹא יִ ְל ְבׁשּו ַא ֶד ֶרת ֵש ָער ְל ַמ ַען ַכ ֵחׁש‬ ִ ‫וְ ָהיָ ה ַבּיֹום ַההּוא יֵ בֹׁשּו ַהנְ ִב‬ ‫׃‬ ‫עּורי‬ ָ ְ‫וְ ָא ַמר לֹא נָ ִביא ָאנ ִֹכי ִאיׁש־ ע ֵֹבד ֲא ָד ָמה ָאנ ִֹכי ִכי ָא ָדם ִה ְקנַ נִ י ִמנ‬ And it will happen on that day that the prophets will be ashamed, each of his vision when he prophesies, and they will not put on a garment of hair in order to deceive. And he will say, “I am not a prophet. I am a man who works the ground because a man has begotten/acquired me since the time of my youth.”3

Zechariah 9–14 has traditionally been divided into two major sections, beginning at 9:1 and 12:1, each introduced by the phrase ‫ ַמ ָשא ְד ַבר־יהוה‬followed by a preposition.4 The second of these sections, encompassing chapters 12–14, is usually divided into two major oracular units (12:2–13:6; 14:1–21), between which lies the pericope 13:7–9. Unlike 13:7–9, the two major oracular units share in common the general theme of battles at Jerusalem resulting in submission of people and renewal of holiness and a similar discourse style that punctuates the ַ ‫ב‬. ַ The two major oracular material at regular intervals with the phrase ‫ּיֹום־ההּוא‬ units contain possible evidence of redaction, relating to developments in the sociology of postexilic Yehud and especially the shifting relationship between Jerusalem and the outlying Judean regions.5 The first oracular unit in 12:2–13:6 ends with an emphasis on the cleansing of the land in 13:1–6, beginning with the provision of a cleansing fountain for sin and impurity in 13:1; followed by a removal of idols, prophets, and the unclean spirit in 13:2; and, finally, the eradication of illicit prophets from the land as parents turn on their child who dares to prophesy falsely (13:3). 3  For this translation with accompanying notes, see M. J. B oda , The Book of Zechariah (NICOT; Grand Rapids, MI: Eerdmans, 2016). 4  See further M. J. Boda, “Reading between the Lines: Zechariah 11:4–16 in Its Literary Contexts,” in Bringing Out the Treasure: Inner Biblical Allusion and Zechariah 9–14 (ed. M. J. Boda and M. H. Floyd; JSOTSup 304; Sheffield: Sheffield Academic, 2003), 277–91; idem , “Freeing the Burden of Prophecy: Maśśāʾ and the Legitimacy of Prophecy in Zech 9–14,” Bib 87 (2005), 338–57. 5  P. L. R edditt , “Israelʼs Shepherds: Hope and Pessimism in Zechariah 9–14,” CBQ 51 (1989), 631–42; R. Rhea, “Attack on Prophecy: Zechariah 13,1–6,” ZAW 107 (1995), 288–93.

Reading Zechariah 9–14 with the Law and the Prophets

987

These three verses orient the reader to the scenarios described in 13:4–6, each of which reveals false prophets seeking to hide their prophetic activity or identity, due to what is called “shame” in v. 4. According to this verse, those ַ words that appear who had identified as prophets will no longer don ‫א ֶד ֶרת ֵש ָער‬, within the biblical traditions about Elijah.6 ‫ ַא ֶד ֶרת‬is used in 1 Kgs 19:13 in a passage where the prophet encounters YHWH on Horeb.7 ‫ ֵש ָער‬is employed again in reference to Elijah in 2 Kgs 1:8 within the idiom ‫איׁש ַב ַעל ֵש ָער‬, ִ which most likely means that he regularly dressed in clothing made from animal hide.8 Most interpreters have concluded that Elijahʼs apparel was used to gain legitimacy for later prophets among the community. Nevertheless, although the individual words used in 13:4 appear in the Elijah traditions, the construct used in 13:4, ‫א ֶד ֶרת ֵש ָער‬, ַ occurs in only one other place in the Hebrew Bible, that is, Gen 25:25, where it refers to Esauʼs hairy body at the time of his birth.9 Rebekah and Jacob will later have to imitate this hairy condition in order to extract the firstborn blessing from Isaac (Gen 27). The importance of deceitfulness to the ‫ ַא ֶד ֶרת ֵש ָער‬tradition in Genesis suggests a relationship to Zech 13:4, where the ‫ ַא ֶד ֶרת ֵש ָער‬is used ‫ל ַמ ַען ַכ ֵחׁש‬.ְ 10 Zechariah 13:5 goes on to say that these false prophets will not only stop dressing like prophets but also verbally deny their prophetic status while pretending to be just common laborers who work the ground (‫ )ע ֵֹבד ֲא ָד ָמה‬and have ָ Again, one can discern the use of done so from their youth (‫עּורי‬ ָ ְ‫)א ָדם ִה ְקנַ נִ י ִמנ‬. earlier biblical traditions in this strategy for concealing the identity of a false prophet. First, Amos 7:14 bears striking resemblance to Zech 13:5:11 6  C. L. Meyers and E. M. Meyers, Zechariah 9–14: A New Translation with Introduction and Commentary (AB 25B; New York: Doubleday, 1993), 379; L arkin , Eschatology (see n. 2), 171; J. G ärtner , Jesaja 66 und Sacharja 14 als Summe der Prophetie: Eine traditions- und redaktionsgeschichtliche Untersuchung zum Abschluss des Jesaja- und des Zwölfprophetenbuches (WMANT 114; Neukirchen-Vluyn: Neukirchener Verlag, 2006), 296; M. A. S weeney , The Twelve Prophets (2 vols.; Berit Olam; Collegeville, MN: Liturgical, 2000), 2:694; P. L. Redditt, Zechariah 9–14 (International Exegetical Commentary on the Old Testament; Stuttgart: Kohlhammer, 2012), 117; see H. G. Reventlow, Die Propheten Haggai, Sacharja und Maleachi (ATD 25/2; Göttingen: Vandenhoeck & Ruprecht, 1993), 120; contra D. L. Petersen, Zechariah 9–14 and Malachi: A Commentary (OTL; Louisville: Westminster John Knox, 1995), 127. 7  See also 1 Kgs 19:19; 2 Kgs 2:2, 8. 8  Although it is possible that it means he was hairy, as suggested by Gen 25:25. 9  Meyers and Meyers, Zechariah 9–14 (see n. 6), 379; Petersen, Zechariah 9–14 (see n. 6), 127; P. L. Redditt, Haggai, Zechariah and Malachi (NCB; Grand Rapids, MI: Eerdmans, 1995), 135; E. W. Conrad, Zechariah (Readings; Sheffield: Sheffield Academic, 1999), 187; M. H. Floyd, Minor Prophets, Part 2 (FOTL 22; Grand Rapids, MI: Eerdmans, 2000), 532; Sweeney, Twelve (see n. 6), 2:694. 10  Meyers and Meyers, Zechariah 9–14 (see n. 6), 379. 11  M ason , “Use” (see n. 1), 171; M eyers and M eyers , Zechariah 9–14 (see n. 6), 380; R eventlow , Die Propheten Haggai, Sacharja und Maleachi (see n. 6), 120; Larkin ,

988 Amos 7:14 Zechariah 13:5

Mark J. Boda

‫ּובֹולס ִש ְק ִמים‬ ֵ ‫י־בֹוקר ָאנ ִֹכי‬ ֵ ‫לֹא־נָ ִביא ָאנ ִֹכי וְ לֹא ֶבן־נָ ִביא ָאנ ִֹכי ִכ‬ ‫עּורי‬ ָ ְ‫לֹא נָ ִביא ָאנ ִֹכי ִאיׁש־ ע ֵֹבד ֲא ָד ָמה ָאנ ִֹכי ִכי ָא ָדם ִה ְקנַ נִ י ִמנ‬

In both passages, the speaker begins by denying his prophetic status using the same words, ‫לֹא־נָ ִביא ָאנ ִֹכי‬, followed by reference to agricultural activity using different vocabulary. This difference in wording highlights a second allusion in Zech 13:5 to earlier biblical traditions. The phrase ‫ ע ֵֹבד ֲא ָד ָמה‬in Zech 13:5 also appears in Gen 4:2 in reference to Cain.12 In the previous verse in Genesis, one finds Adam (‫)ה ָא ָדם‬ ָ impregnating Eve and producing their firstborn son, Cain. ִ ִ‫קנ‬,ָ a phrase that is This is followed by the declaration of Eve, ‫יתי ִאיׁש ֶאת־יְ הוָ ה‬ echoed in the declaration of the one hiding his prophetic status in Zech 13:5: ‫ָא ָדם‬ ‫עּורי‬ ָ ְ‫ה ְקנַ נִ י ִמנ‬. ִ Once again we see a tradition connected with the legitimacy of a prophetic figure intertwined with a sibling-rivalry tradition in Genesis. This evidence strongly suggests that Zech 13:4 and 5 employ a common strategy in their allusion to earlier biblical traditions.13 Both allude to an earlier prophetic tradition related to the calling of a prophet whose commission is to attack pagan religion in the Northern Kingdom of Israel (Elijah and Elisha donning their cloaks; Amos defending his call). At the same time, both allude to an earlier Genesis sibling tradition involving deceit, and in the cases of both Esau/Jacob and Cain/Abel, a deception leads to the deceiver roaming the earth in fear of others. B. Zechariah 9:11 Zechariah 9:11

ֵ ‫ם־ב ִר‬ ְ ‫ם־א ְת ְב ַד‬ ַ ַ‫ג‬ ‫׃‬ ‫יתְך ִש ַל ְח ִתי ֲא ִס ַיריִ ְך ִמּבֹור ֵאין ַמיִ ם ּבֹו‬ Also you, because of the blood of the covenant which you made, I have set free your prisoners from the cistern in which there is no water.14

Most scholars have noted the presence of two major parallel oracular units within the first ‫ ַמ ָשא‬unit in Zech 9–14 as well, the units now encompassing Eschatology (see n. 2), 171; Gärtner, Jesaja 66 und Sacharja 14 (see n. 6), 296; Sweeney, Twelve (see n. 6), 2:694; J. D. Nogalski, The Book of the Twelve: Micah–Malachi (The Smyth & Helwys Bible Commentary; Macon, GA: Smyth & Helwys, 2011), 962; Redditt, Zechariah 9–14 (see n. 6), 117; A. Wolters, Zechariah (HCOT; Leuven: Peeters, 2014), 429. 12  F loyd , Minor Prophets (see n. 9), 533; R edditt , Zechariah 9–14 (see n. 6), 118; Sweeney, Twelve (see n. 6), 2:694, notes this but misses the connection to Eveʼs statement, instead linking it back to Zech 11:5, for which there is little justification. Sweeney also considers this connection between false prophecy and Cain as farmer as a prelude to the slaying of a shepherd (like Abel) in 13:7–9. M eyers and M eyers , Zechariah 9–14 (see n. 6), 381, note a connection to Gen 4 but then suggest an emendation that distances it from this passage. 13  For similar connections see Rhea, “Attack” (see n. 5), 291–92. 14  For this translation with accompanying notes, see Boda, Book of Zechariah (see n. 3), 721.

Reading Zechariah 9–14 with the Law and the Prophets

989

9:1b–17 and 10:4–12. As with chapters 12–14, a short pericope lies between these two units (10:1–3). Chapter 11 is distinguished from chapters 9–10 and serves as a central pericope comprising three units that lie between the two ‫ַמ ָשא‬ units. As with chapters 12–14, there is some evidence of redactional layering within chapters 9–10, possibly marked by the modulation of prophetic and divine voices throughout the oracles. The two main oracular units in 9:1b–17 and 10:4–12, however, share some key elements in common, both highlighting Godʼs plan to enact restoration for his people by restoring royal rule at Jerusalem, releasing both southern (Judah) and northern (Ephraim, Joseph) tribes from captivity, defeating the nations by using these released captives, and renewing the prosperity of the land. In the opening chapter, after defeating the nations and encamping at Jerusalem in 9:1–8, YHWH addresses the Daughter of Zion/Jerusalem, first introducing her king to her (vv. 9–10) before announcing the release of her inhabitants from captivity (v. 11). These same inhabitants are then addressed in v. 12, invited to return to Jerusalem, and identified in v. 13 as Judah and Ephraim, who are Zionʼs sons. Zechariah 9:11 contains a biblical allusion that is strikingly similar to what we observed in 13:4–5. The location from which YHWH promises to set free (‫שלח‬, piel) the prisoners related to Zion is ‫ּבֹור ֵאין ַמיִ ם ּבֹו‬. The word ‫ ּבֹור‬is used in relation to imprisonment several other times in the Hebrew Bible: Gen 37:20, 22, 24, 28–29; 40:15; 41:14; Isa 24:22; Jer 38:6–7, 9–11, 13; and Lam 3:53. ֵ The The designation for a prison in Exod 12:29 and Jer 37:16 is ‫בית ַהּבֹור‬. combination of the verb ‫ שלח‬in the piel with the phrase ‫ ּבֹור ֵאין ַמיִ ם‬also occurs in Jer 38:6, where Jeremiah is imprisoned in a cistern lacking water (‫וַ יַ ְש ִלכּו אֹתֹו‬ ַ ‫] וַ יְ ַש ְלחּו ֶאת־יִ ְר ְמיָ הּו ַב ֲח ָב ִלים‬. . .[ ‫ל־הּבֹור‬ ַ ‫)א‬. ֶ 15 One element in the phrase ‫ין־מיִ ם‬ ַ ‫ּובּבֹור ֵא‬ in Zech 9:11, however, does not occur in Jer 38:6: the prepositional morpheme ‫ּבֹו‬. The only place other than Zech 9:11 and Jer 38:6 in the Hebrew Bible where the combination ‫ ּבֹור ֵאין ַמיִ ם‬occurs is Gen 37:24, which describes Joseph being thrown into a cistern by his brothers.16 The term ‫ ּבֹור‬appears elsewhere in the Joseph narrative to describe his prison in Egypt (Gen 40:15; 41:14). Further 15  Mason, “Use” (see n. 1), 49–50; Meyers and Meyers, Zechariah 9–14 (see n. 6), 142; R edditt , Haggai (see n. 9), 54; P erson , Second Zechariah and the Deuteronomic School (see n. 1), 118; Larkin, Eschatology (see n. 2), 80; Tai, Prophetie als Schriftauslegung (see n. 1), 55; Wolters, Zechariah (see n. 11), 284. 16  M. S æbø , Sacharja 9–14: Untersuchungen von Text und Form (WMANT 34; Neukirchen-Vluyn: Neukirchener Verlag, 1969), 190; Mason, “Use” (see n. 1), 49–50; Meyers and M eyers , Zechariah 9–14 (see n. 6), 142; P etersen , Zechariah 9–14 (see n. 6), 60; Reventlow, Die Propheten Haggai, Sacharja und Maleachi (see n. 6), 98; Redditt, Haggai (see n. 9), 54; Larkin, Eschatology (see n. 2), 80; Tai, Prophetie als Schriftauslegung (see n. 1), 55; Wolters, Zechariah (see n. 11), 284; Lee, Intertextual Analysis (see n. 2), 122–125. Contra Sweeney, Twelve (see n. 6), 2:665, who links the reference to prisoners in a waterless pit to Isa 24.

990

Mark J. Boda

evidence of connectivity between Zech 9:11 and the Joseph texts in Genesis can be discerned in the use of ‫( ָא ִסיר‬prisoner, 39:20Q, 22; cf. 40:3, 5) to refer to Joseph, the same term that appears in Zech 9:11 to refer to those who are in the waterless pit. Thus Zech 9:11 contains elements that betray influence from both Jer 38 and the Joseph texts of Genesis. It appears, then, that innerbiblical allusions can once again be discerned that intertwine an earlier sibling-rivalry tradition from the book of Genesis (Joseph and his brothers) with an earlier prophetic-crisis tradition, this time related to the prophet Jeremiah.17 In the context of chapters 9–10, 9:11 is seeking to communicate the message of salvation to the audience, focusing especially on the restoration of the community from captivity to Jerusalem.

II. Innerbiblical Allusion and the Source and Host Texts The pattern of intertwining earlier sibling-rivalry traditions from the book of Genesis with earlier prophetic-crisis traditions in Zech 9–14 could be understood as an example of what Michael Stead has identified as “Sustained Allusion,” that is, “multiple scattered references to another text” or in this case multiple scattered references to similar types of texts.18 It is one thing to identify an innerbiblical allusion, but it is another to identify the reason for the allusion. This section of the article will build a foundation for reflection on the rhetorical strategy underlying these allusions by considering first the source texts (Genesis and Prophets) and then the host text (Zech 9–14). Consideration of both is important since, as Ziva Ben-Porat noted long ago, the purpose of innerbiblical connections like this is “the simultaneous activation of two texts.”19 A. Appropriateness of This Literary Allusion within the Source Texts: Genesis and Prophets 1 Genesis There is little question that the term ‫ּתֹולדֹות‬ ְ plays a significant function within the book of Genesis, whether as a structural signal at key intervals within the  Meyers and Meyers, Zechariah 9–14 (see n. 6), 141–42, also see here “an attempt to draw a connection between this passage and the Joseph and Jeremiah stories (Gen 37:24; Jer 38:6)”; similarly Mason, “Use” (see n. 1), 49–50. 18  M. R. Stead, “Sustained Allusion in Zechariah 1–2,” in Tradition in Transition: Haggai and Zechariah 1–8 in the Trajectory of Hebrew Theology (ed. M. J. Boda and M. H. Floyd; London: T&T Clark, 2008), 144–170, here 145; see M. R. S tead , The Intertextuality of Zechariah 1–8 (LHBOTS 506; London: T&T Clark, 2009). 19  Z. Ben-Porat, “The Poetics of Literary Allusion,” PTL: A Journal of Descriptive Poetics and Theory of Literature 1 (1976), 105–128, here 107; see Z. Ben-Porat, “Intertextuality,” Ha-Sifrut 34 (1985), 170–78 (Hebrew); with thanks to S tead , “Sustained Allusion” (see n. 18), 154. See a further example of this approach in Boda, “Reading” (see n. 4), 277–91. 17

Reading Zechariah 9–14 with the Law and the Prophets

991

book or simply a consistent lexical motif.20 This term sets the tone for the book, which emphasizes the origins and growth of the Abrahamic clan from the introduction of the image of seed and its multiplication in Gen 1 and 9 to the covenantal and promissory scenes in relation to Abraham, Isaac, and Jacob. This theme is articulated through certain leitmotifs and/or type scenes that appear throughout the book, including those related to female characters: the threat to the seedbearer (Abraham/Sarah, Isaac/Rebekah), barrenness (Eve, Sarah, Rachel), and maidservants used for bearing seed (Sarah/Hagar, Leah/ Rachel/Bilhah/Zilpah).21 Alongside these scenes involving female characters is a series of scenes involving mostly male figures, all of which focus on the dynamics of sibling rivalry: Cain/Abel, Shem/Ham/Japheth, Abram/Lot, Isaac/Ishmael, Jacob/ Esau, Rachel/Leah, Ephraim/Manasseh, and Jacobʼs sons. 22 These are key stories that drive the narrative of Genesis, accentuating the idea of ‫ּתֹולדֹות‬ ְ but doing so through crises that threaten the life of the promised seed. It is, then, not surprising that the one(s) responsible for Zech 9–14 have leveraged three of these stories from Genesis, where they play a key role in the final form of the book. The stories, however, are not all the same, and one can discern a development in the type scenes and/or motifs among the stories that have been chosen in Zech 9–14. The Cain/Abel story is the most severe in outcome, without positive resolution to the sibling rivalry, as Abel is killed. This emphasis on death is also apparent in the story of Jacob and Esau, as Jacob must flee from his brother lest he be killed (Gen 27:41–42; 32:11). Esau does not carry this threat out, however, and in the end there is a scene of reconciliation (ch. 33), admittedly with some ambiguity, since Jacob never does fulfill his promise to meet Esau at Seir (33:14; see 33:17–20). In the Joseph/brothers story, the brothers discuss killing Joseph and even fake his 20  See recently, M. A. T homas , These Are the Generations: Identity, Covenant, and the Toledot Formula (LHBOTS 551; New York: T&T Clark, 2011); J. S. DeRouchie, “The Blessing-Commission, the Promised Offspring, and the Toledot Structure of Genesis,” JETS 56 (2013), 219–47; and for bibliography, see C. M. Kaminski, Was Noah Good? Finding Favour in the Flood Narrative (LHBOTS 563; New York: Bloomsbury, 2014), 41–42. 21  On these female rivalry traditions, see I. Pardes , Countertraditions in the Bible: A Feminist Approach (Cambridge, MA: Harvard University Press, 1992), 63. Alterʼs classic treatment of type scenes draws on the female rival traditions; see R. Alter, “Biblical TypeScenes and the Uses of Conventions,” Critical Inquiry 5 (1978), 355–68. 22  On sibling rivalry, see M. J. Boda, “Sibling Rivalry,” in Dictionary of Biblical Imagery (ed. L. Ryken et al.; Downers Grove, IL: InterVarsity Press, 1997), 789, and K. W. Stuebing, “Sibling Rivalry,” in The IVP Womenʼs Bible Commentary (ed. C. C. Kroeger and M. J. Evans; Downers Grove, IL: InterVarsity Press, 2002), 17–18. It may be that the reason only Cain/Abel, Jacob/Esau, Joseph/brothers were chosen in Zech 9–14 was that they are the most similar, with Shem/Ham/Japheth involving a father figure, Abram/Lot involving an uncle/nephew relationship, and Isaac/Ishmael and Leah/Rachel reflecting wife rivalry traditions. The only other story that is similar is the Ephraim/Manasseh story.

992

Mark J. Boda

death by spilling blood on his coat, but in the end there is resolution by means of Josephʼs discussion with his brothers (50:15–21) and the reuniting of Jacobʼs family in Egypt. 2 Prophetic-Crisis Traditions Among the Hebrew canonical section called the Former and Latter Prophets, the three stories alluded to in Zech 9–14 are the three most intensely focused on the legitimacy of true prophecy. Within the Former Prophets, the Elijah/Elisha complex highlights the problem of false prophetic sources, showcased especially in Elijahʼs defeat of the prophets of Baal in 1 Kgs 18. Amosʼs encounter with the priest Amaziah in Amos 7 highlights the challenge that the prophet faces from the religious hierarchy. Jeremiah regularly expresses concern over false prophecy, showcased especially in his attack on those prophets at both Samaria and Jerusalem who speak a vision from their own imagination rather than from having stood in the council of YHWH (ch. 23). Of all the prophets, it is not surprising that these three were chosen, in that they share in common serious crisis in prophecy. Furthermore, these prophets have been plucked from three different eras of the history of Israel (Elijah in the ninth century, Amos in the eighth century, and Jeremiah in the seventh–sixth centuries BCE), and each represents a different configuration of north/south and prophet/audience relations: Elijah, a northern prophet to the north; Amos, a southern prophet to the north; and Jeremiah, a southern prophet to the south. 3 Summary This analysis shows that those responsible for Zech 9–14 are drawing on key motifs within at least one literary unit (Genesis) and possibly another (Prophets). The repeated use of these source texts within Zech 9–14 places these larger literary traditions and their associated themes in the background, which in turn affects our reading of Zech 9–14. B. Appropriateness of This Literary Allusion within the Host Text: Zechariah 9–14 Although these two passages in Zech 9–14 employ a similar combination of earlier biblical traditions, their respective contexts are quite different. How, then, do these instances of innerbiblical allusion function within their respective sections? 1 Orientation to the Redactional Structure of Zechariah 9–14 As already noted, Zech 9–14 comprises two major oracular sections, distinguished by the employment of the phrase ‫ ַמ ָשא ְד ַבר־יהוה‬at the outset of Zech 9

Reading Zechariah 9–14 with the Law and the Prophets

993

and 12.23 Each of these two sections is dominated by a particular literary style (historical/eschatological), prophetic tone (positive/negative), and sociological projection (Judah-Ephraim/Judah-Jerusalem), displayed in the major oracular material in 9:1–17; 10:4–12 in the first section and 12:1–13:6; 14:1–21 in the second. Throughout these contrasting oracular sections run a series of smaller negative prophetic units that all share shepherd-flock imagery and imperative expressions (10:1–3; 11:1–3; 11:17; 13:7–9).24 At the center of this complex lies the report of a prophetic sign-act in Zech 11:4–16, which also employs shepherd-flock imagery and imperative expressions. The depiction in this signact of the dissolution of the relationship between the shepherd and the nations and between Israel and Judah signals the shift in literary style, prophetic tone, and sociological projection between the two major oracular sections in Zech 9–14. The four short shepherd-flock units in 10:1–3; 11:1–3, 17; and 13:7–9 divide each of the two major oracular sections into two smaller sections that are nonetheless lexically and imagistically related to their surrounding sections and thus were created with those in view. ‫( ַמ ָשא ְד ַבר־יהוה‬9:1a)  Oracle for Jerusalem, Judah and Ephraim (9:1b–17)   Shepherd-flock unit: with vegetation imagery (10:1–3)  Oracle for Judah and Ephraim (10:4–12)   Shepherd-flock unit: with vegetation imagery (11:1–3)    Shepherd-flock unit: prophetic sign-act report (11:4–16)   Shepherd-flock unit: with sword imagery (11:17) ‫( ַמ ָשא ְד ַבר־יהוה‬12:1a)  Oracle for Jerusalem, Judah, house of David, and nations (12:2–13:6)   Shepherd-flock unit: with sword imagery (13:7–9)  Oracle for Jerusalem, Judah, and nations (14:1–21)25 2 Zechariah 9:11 within Zechariah 9–10 In light of the above discussion, it is apparent that the sibling rivalry/prophetic crisis traditions in Zech 9:11 are used in a section dominated by prophetic promise, focusing on the repopulation of Jerusalem by a community comprising Judah and Ephraim. The allusion to Josephʼs experience of imprisonment in 9:11 is important in the context of Zech 9–10, which envisions salvation for the house of Joseph/  For further argumentation, see Boda, “Reading” (see n. 4), 277–291.  Note E. R. W endland , The Discourse Analysis of Hebrew Prophetic Literature: Determining the Larger Textual Units of Hosea and Joel (Mellen Biblical Press Series 40; Lewiston, NY: Mellen, 1995), 42, who speaks of the use of forceful expression as a key discourse marker. 25  See Boda, Zechariah (see n. 3), 520–521. 23

24

994

Mark J. Boda

Ephraim (9:13; 10:6, 7; cf. 9:10). In Gen 37–50, it is Judah who saves the life of Joseph by suggesting that the brothers sell him to the Ishmaelites. Later in the Joseph story, Judah will play a key role in ensuring the survival of Jacobʼs family, assuming the role of spokesperson for the brothers as they converse first with Jacob and then with Joseph (Gen 43:3, 8–9; 44:14, 16, 18–34; cf. 46:28). By bringing the Joseph story into view, Zech 9:11 provides a precedent for Judahʼs role in relationship to Joseph and Ephraim. As the one who sold his brother into slavery, he will now play a role in releasing him from imprisonment in exile (see 10:6). As mentioned in the review of Genesis above, the Joseph sibling-rivalry story is the most positive of the various stories; it is not surprising that an allusion to this story would appear in Zech 9–10, which speaks mostly positively about the restoration of both northern and southern tribal entities. As for the role that allusion to the Jeremianic prophetic-crisis tradition plays in 9:11, there are a couple of possibilities. First, it may be that Jeremiahʼs experience is treated as an ironic foreshadowing of what would happen to the Judeans who refused to listen to his message (cf. Jer 18:19–23). In Zech 9:11, this would serve to remind the audience of the source of the captivity of the Judeans whose release is now announced. Second, it may be that this allusion to the prophetic crisis is subtly connected to the redactional strategy of the shepherd units, especially to the issue raised in the first shepherd unit at the centre of chapters 9–10, which focus on false prophecy, citing Jeremianic tradition (see esp. 10:1; cf. Jer 14:1–15:4).26 This could serve as a subtle reminder that what caused the captivity in the first place was concern over the relationship between people and prophecy, an enduring issue within the present community that threatened the success of the realization (see further below). 3 Zechariah 13:4–5 within Zechariah 12–14 In contrast to Zech 9:11, the sibling-rivalry/prophetic-crisis traditions in Zech 13:4–5 appear in a section dominated by prophetic judgment, focusing on the eradication of false prophecy from a community comprising Jerusalem and Judah. Though it was the sibling-rivalry tradition that seemed to fit most comfortably within the broader context of Zech 9:11, here it is the prophetic-crisis tradition that better suits the context, since Zech 13:1–6 focuses considerable attention on the cleansing of the community. By alluding to these earlier traditions, the one(s) responsible for Zech 13:1–6 highlight YHWHʼs rejection of the idolatrous stream highlighted in Zech 13:2 and also developed within the shepherd units throughout Zech 9–14.27  See Boda, “Freeing the Burden of Prophecy” (see n. 4), 355.  Notice also similar use of the language and diction of Deuteronomy and Jeremiah in both Zech 10:1–2 (which may explain use of the prophetic crisis tradition in 9:11) and Zech 13:3. Thus the development of what we see in Zech 13:1–6, or a portion of it, may be related to the stage when the shepherd units were used to draw together the collection. 26 27

Reading Zechariah 9–14 with the Law and the Prophets

995

The role that the sibling-rivalry traditions of Zech 13:4–5 play in Zech 12–14 is not as clear. As noted in the review of Genesis above, the sibling-rivalry traditions used in Zech 13:4–5 (Cain/Abel, Jacob/Esau) are not fully resolved, nor is the sibling rivalry tradition (Joseph/brothers) used in Zech 9:11. In the rhetorical flow of Zech 9–14, it is then not odd that these unresolved sibling-rivalry traditions are used in a text after the dissolution of the covenant between Israel and Judah in the shepherd unit of Zech 11:4–16.28 In addition, the use of the earlier Cain/Abel and Jacob/Esau traditions here in Zech 13, rather than Joseph/brothers with its focus on the various tribes of Israel, may reflect the focus of Zech 12–14 on a more limited definition of the community (Judah). One other possibility is that the allusion to the sibling-rivalry traditions alongside prophetic-crisis traditions focused on the north may be a subtle way of pointing to the source of the postexilic prophetic crisis: that is, figures in the province of Samaria who, according to the sociological worldview of Zech 12–14, are not even within the orbit of Jerusalem.

III. Conclusion Though there is no question that the biblical prophetic traditions and Deuteronomy exerted the most influence on the one(s) responsible for the texts now found in Zech 9–14, one can discern within both major oracular sections the influence of the opening book of the Torah. In Genesis, the one(s) responsible for Zech 9–14 found a narrative that echoed some of their own hopes and fears: an initial hope of estranged brothers saving one another in order to return to the land, only to be followed by later fear of an estrangement made necessary by threats to the very life of one brother. Intertwined with these connections to the sibling-rivalry traditions of Genesis are prophetic-crisis traditions that are part of the section of the Hebrew canon known today as the Prophets. If this strategy of intertwining biblical traditions from Genesis and the Prophets can be sustained, then this may suggest the role Zech 9–14 and possibly the Twelve as a collection played in an emerging canon consciousness. This has been suggested for the final verses of the book of Malachi, which intertwine Moses (Torah) and Elijah (Prophets),29 as well as Zech 1–8, which refers at three  See Boda, “Reading” (see n. 4), 277–291.  J. D. Nogalski, Redactional Processes in the Book of the Twelve (BZAW 218; Berlin: de Gruyter, 1993), 185; B. A. J ones , The Formation of the Book of the Twelve: A Study in Text and Canon (SBLDS 149; Atlanta: Scholars Press, 1995), 236–37; S. G. Dempster, “An ‘Extraordinary Fact’: Torah and Temple and the Contours of the Hebrew Canon,” TynBul 48 (1997), 23–56; A. Schart, Die Entstehung des Zwölfprophetenbuchs (BZAW 260; Berlin: de Gruyter, 1998), 302–3; S. B. Chapman, The Law and the Prophets: A Study in Old Testament Canon Formation (FAT 27; Tübingen: Mohr Siebeck, 2000), 112; S. G. D empster , “The Prophets, the Canon and a Canonical Approach,” in Canon and Biblical Interpretation (ed. 28

29

996

Mark J. Boda

places explicitly to the “earlier prophets” (Zech 1:3; 7:7, 12).30 The substantial innerbiblical allusion found in Zech 9–14 (beyond what has been discussed in this article) also heightens the possibility that this intertwining of biblical traditions from what we now know as Torah and Prophets reflects consciousness of a collection that includes the combined Torah and Prophets. These literary signals of a possible larger collection, if not canonical consciousness, suggest another important direction of research on the Torah that may be in danger of being overlooked or drowned out by other legitimate and helpful approaches focused either on the Torah alone or on underlying developmental layers within the Torah. To read the Torah in its final form and in connection with the Prophets provides keys to interpreting not only a text like Zech 9–14 but possibly also the Torah itself.

C. G. B artholomew et al.; The Scripture and Hermeneutics Series 7; Grand Rapids, MI: Zondervan, 2006), 293–325. 30  See Boda, Zechariah (see n. 3), 39–41.

Jacob, Moses, Levi Pentateuchal Figures in the Book of the Twelve Jakob Wöhrle Introduction The Book of the Twelve is a collection of twelve rather different prophetic books with a long-term redactional development.1 The individual books as well as the whole collection developed through several centuries, during which these books were re-worked – parts of them several times – and integrated into smaller and larger collections. The oldest traditions within the Book of the Twelve trace back to the eighth century. The latest literary strata probably stem from the second century. Within the Book of the Twelve, one finds several references to the Pentateuch. These references range from short adaptations of individual motifs, such as the exodus motif, up to comprehensive reformulations of pentateuchal traditions. Investigating the centuries-long history of formation of the Book of the Twelve thus gives important insights into the question of when the reception of such pentateuchal traditions began. In this way, the Book of the Twelve can shed some light on the emergence of pentateuchal traditions and hence on the formation of the Pentateuch itself. The following article will treat some comprehensive references to pentateuchal traditions within the Book of the Twelve. It deals with passages referring to pentateuchal figures and to certain issues connected with these figures: the adaptation of the Jacob account in Hos 12, the statements about Moses and the  For the formation of the Book of the Twelve, see the comprehensive studies of J. D. Nogalski, Literary Precursors to the Book of the Twelve (BZAW 217; Berlin: de Gruyter, 1993); idem, Redactional Processes in the Book of the Twelve (BZAW 218; Berlin: de Gruyter, 1993); A. Schart, Die Entstehung des Zwölfprophetenbuches: Neubearbeitungen von Amos im Rahmen schriftenübergreifender Redaktionsprozesse (BZAW 260; Berlin: de Gruyter, 1998); E. B osshard -N epustil , Rezeptionen von Jesaia 1–39 im Zwölfprophetenbuch: Untersuchungen zur literarischen Verbindung von Prophetenbüchern in babylonischer und persischer Zeit (OBO 154; Göttingen: Vandenhoeck & Ruprecht, 1997); J. W öhrle , Die frühen Sammlungen des Zwölfprophetenbuches: Entstehung und Komposition (BZAW 360; Berlin: de Gruyter, 2006); idem, Der Abschluss des Zwölfprophetenbuches: Buchübergreifende Redaktionsprozesse in den späten Sammlungen (BZAW 389; Berlin: de Gruyter, 2008). 1

998

Jakob Wöhrle

exodus in Mic 6, and the presentation of a covenant with Levi in Mal 2. The article will discuss the formation of these texts and the pentateuchal traditions presupposed in them, in order to gain insights into how and when pentateuchal traditions were adapted within the Book of the Twelve.

1 The Jacob Traditions in Hosea 12 The most comprehensive and most important adaptation of pentateuchal traditions within the Book of the Twelve, if not within all the prophetic writings of the Hebrew Bible, is to be found in Hos 12. This chapter offers a broad, but in many details rather difficult, judgment speech against the people. Under the recurring catchword “deceit” (‫מרמה‬, 12:1, 8; cf. ‫תמרורים‬, 12:15),2 it condemns the peopleʼs alliances with Assyria and Egypt or their economic misdeeds. The chapter ends with the announcement that YHWH will punish the peopleʼs behavior and return their deeds upon them. Hos 12 includes two passages referring to episodes from the life of the ancestor Jacob. After the opening words against deceit in political affairs in 12:1–2, a first reference to Jacob is documented in 12:3–7*:3 2  See E. Sellin , Hosea–Micha (2nd–3rd ed.; KAT 12/1; Leipzig: Deichertsche Verlags­ buch­handlung, 1929), 118; H. W. Wolff, Hosea (BKAT 14/1; Neukirchen-Vluyn: Neukir­che­ ner Verlag, 1961), 268; W. Rudolph, Hosea (KAT 13/1; Gütersloh: Gütersloher Verlagshaus, 1969), 223; H. Utzschneider, Hosea, Prophet vor dem Ende: Zum Verhältnis von Geschich­te und Institution in der alttestamentlichen Prophetie (OBO 31; Göttingen: Vandenhoeck & Ruprecht, 1980), 195–196; H.-C. S chmitt , “Der Kampf Jakobs mit Gott in Hos 12,3ff. und in Gen 32,23ff.,” in Theologie in Prophetie und Pentateuch: Gesammelte Schriften (ed. U. Schorn and M. Büttner; BZAW 310; Berlin: de Gruyter, 2001), 165–188, here 169; M. K öhlmoos , Bet-El – Erinnerungen an eine Stadt: Perspektiven der alttestamentlichen Bet-El-Überlieferung (FAT 49; Tübingen: Mohr Siebeck, 2006), 143. 3  Within Hos 12:3–7, v. 6 is doubtless a secondary addition. This verse is a kind of doxological statement interrupting the close connection between the introduction of a divine speech to Jacob at the end of 12:5 and the content of this speech in 12:7. For the secondary character of Hos 12:6, see, e.g., K. Marti, Das Dodekapropheton (KHC 13; Tübingen: Mohr Siebeck, 1904), 95; Sellin, Hosea–Micha (see n. 2), 122; Wolff, Hosea (see n. 2), 276; L. Ruppert, “Herkunft und Bedeutung der Jakob-Tradition bei Hosea,” Bib 52 (1971), 488–504, here 492; Utzschneider, Hosea (see n. 2), 210; J. Jeremias, Der Prophet Hosea (ATD 24/1; Göttingen: Vandenhoeck & Ruprecht, 1983), 150; H.-D. Neef, Die Heilstraditionen Israels in der Verkündigung des Propheten Hosea (BZAW 169; Berlin: de Gruyter, 1987), 24; A. de Pury, “Osée 12 et ses implications pour le débat actuel sur le Pentateuque,” in Le Pentateuque: Débats et recherches (ed. P. Haudebert; LD 151; Paris: Cerf, 1992), 175–207, here 188, n. 44; H. Pfeiffer , Das Heiligtum von Bethel im Spiegel des Hoseabuches (FRLANT 183; Göttingen: Vandenhoeck & Ruprecht, 1999), 72–73; Köhlmoos, Bet-El (see n. 2), 143, n. 245; E. Blum, “Hosea 12 und die Pentateuchüberlieferungen,” in Die Erzväter in der biblischen Tradition (ed. A. C. Hagedorn and H. Pfeiffer; BZAW 400; Berlin: de Gruyter, 2009), 291–321, here 300.

Jacob, Moses, Levi

999

Hosea 12:3–5, 7 YHWH has a suit against Judah to punish Jacob according to his ways, according to his deeds he will repay him. In the womb (‫ )בבטן‬he overreached (‫ )עקב‬his brother, in his strength he struggled with God (‫)שרה את אלהים‬. He struggled (‫)שרה‬4 against an angel5 and prevailed (‫)יכל‬, he wept and implored him. At Bethel he found him (‫)בית אל ימצאנו‬, there he spoke with ‘him’6 (‫)ושם ידבר עמנו‬:7 . . . 4  It is often assumed that, unlike ‫ שרה‬in Hos 12:4b, ‫ וישר‬in 12:5 derives not from ‫שרה‬, “to struggle,” but from ‫שרר‬, “to rule”; see M. Gertner, “The Masorah and the Levites: An Essay in the History of a Concept,” VT 10 (1960), 241–284, here 277; W olff , Hosea (see n. 2), 267–268; Ruppert, “Herkunft” (see n. 3), 496, n. 1; Utzschneider, Hosea (see n. 2), 189–190; Jeremias , Prophet Hosea (see n. 3), 153; W. D. Whitt , “The Jacob Traditions in Hosea and their Relation to Genesis,” ZAW 103 (1991), 18–43, here 32; de Pury, “Osée 12” (see n. 3), 187, n. 41; Blum, “Hosea 12” (see n. 3), 299–300, n. 41. However, because Hos 12:4–5 depends upon Gen 32:23–33 (see below), where the verb ‫ שרה‬is of central importance, it is much more likely that the term ‫ וישר‬in 12:5a also derives from the verb ‫ שרה‬and thus has to be vocalized ‫ ;וַ ּיִ ַׂשר‬see J. L. Mays, Hosea: A Commentary (OTL; London: SCM, 1969), 161; Neef, Heilstraditionen (see n. 3), 14, 17; Pfeiffer, Heiligtum (see n. 3), 85; Köhlmoos, Bet-El (see n. 2), 142, n. 241. 5  Given that Jacob struggled with ‫אלהים‬, “God,” in Hos 12:4b, the term ‫מלאך‬, “angel,” in Hos 12:5a is often seen as a later gloss and the preposition ‫ ֶאל‬in 12:5a understood to represent an original ‫אל‬, ֵ “God.” The beginning of Hos 12:5a, ‫וישר אל‬, is then interpreted in the sense of “God ruled,” based upon the view that ‫ וישר‬derives from the verb ‫שרר‬, “to rule” (see above, n. 4); see, e.g., Gertner, “Masorah” (see n. 4), 277–281; Wolff, Hosea (see n. 2), 267–268; Ruppert, “Herkunft” (see n. 3), 496; Utzschneider, Hosea (see n. 2), 189–190; Jeremias, Prophet Hosea (see n. 3), 153; W hitt , “Jacob Traditions” (see n. 4), 32; de P ury , “Osée 12” (see n. 3), 187, n. 41. However, since this assumption is rather speculative and since the present text of Hos 12:4b, 5a is easily understandable as speaking both of God and of an angel in order to leave open the question of which concrete heavenly being appeared to Jacob, such far-reaching text-critical procedures can and should be avoided; thus also E. M. Good, “Hosea and the Jacob Tradition,” VT 16 (1966), 137–151, here 141; Pfeiffer, Heiligtum (see n. 3), 86–87. 6  ‫עמנו‬, “with us,” seems to be a misspelling of an original ‫עמו‬, “with him,” caused by a misinterpretation of the third-person singular suffix with energic nun in the preceding verb, ‫ ;ימצאנו‬see Wellhausen , Die kleinen Propheten (see n. 6), 129; M arti , Dodekapropheton (see n. 3), 95; Sellin, Hosea–Micha (see n. 2), 122; Wolff, Hosea (see n. 2), 268; Rudolph, Hosea (see n. 2), 222; Mays, Hosea (see n. 4), 161; Köhlmoos, Bet-El (see n. 2), 142–143, n. 244; Blum, “Hosea 12” (see n. 3), 300, n. 43. 7  The imperfect forms ‫ ימצאנו‬and ‫ ידבר‬in Hos 12:5b can be understood only in a past sense; see J. W ellhausen , Die kleinen Propheten (Berlin: Reimer, 1892), 129; M arti , Dodekapropheton (see n. 3), 95; Köhlmoos, Bet-El (see n. 2), 142, n. 241. An iterative sense, as advocated by B lum , “Hosea 12” (see n. 3), 300, n. 42, is unlikely since it is not really understandable why God should “again and again” find Jacob in Bethel and talk to him. Even more unlikely is a present sense, as proposed by Pfeiffer, Heiligtum (see n. 3), 70, 79–80,

1000

Jakob Wöhrle

You, by the help of your God you shall return, maintain faithfulness and justice, and constantly wait on your God.

After the subsequent words against deceit in trade in Hos 12:8–14, a further reference to the ancestor Jacob is to be found in 12:13–14: Hosea 12:13–14 Jacob fled (‫ )ברח‬to the fields of Aram (‫)שדה ארם‬, Israel served (‫ )עבד‬for a wife, for a wife he kept sheep. But by a prophet (‫ )נביא‬YHWH brought Israel up from Egypt, by a prophet he was kept.

Hosea 12 thus mentions several occurrences in Jacobʼs life: a prenatal conflict with his brother, a fight with God or an angel, and a theophany at Bethel, as well as Jacobʼs escape to Aram and his service there for two wives. Beyond the life of the ancestor, it also mentions the exodus of Godʼs people led by a prophet. Owing to these references to certain episodes of Jacobʼs life and to the exodus, Hos 12 has become important in older as well as in more recent research on the formation of the Pentateuch. In pentateuchal studies, Hos 12 has been and is still taken as evidence that the Jacob account of the book of Genesis – or at least the oldest strata of this account – already existed in the time of Hosea, in other words, in the eighth century.8 Hosea 12, with its juxtaposition of the Jacob and exodus traditions, is also taken as evidence that the ancestor account and the exodus story were already connected in this early time rather than – as presumed in some current approaches9 – only in the Priestly version of the growing Pentasince the context of Hos 12:5b requires that it alludes to past events of the ancestorʼs life and not to the current situation of the addressees. 8  E. B lum , Die Komposition der Vätergeschichte (WMANT 57; Neukirchen-Vluyn: Neukirchener Verlag, 1984), 161–163; idem , “Hosea 12” (see n. 3), 312–319; A. de P ury , “The Jacob Story and the Beginning of the Formation of the Pentateuch,” in A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (ed. T. B. Dozeman and K. Schmid; Society of Biblical Literature Symposium Series 34; Atlanta: Society of Biblical Literature, 2006), 51–72, here 61–62; J. Van Seters, The Pentateuch: A Social-Science Commentary (Sheffield: Sheffield Academic, 1999), 108; D. M. C arr , The Formation of the Hebrew Bible: A New Reconstruction (Oxford: Oxford University Press, 2011), 474–475. 9  K. S chmid , Erzväter und Exodus: Untersuchungen zur doppelten Begründung der Ursprünge Israels innerhalb der Geschichtsbücher des Alten Testaments (WMANT 81; Neukirchen-Vluyn: Neukirchener Verlag, 1999), 358–359; J. C. Gertz, Tradition und Redaktion in der Exoduserzählung: Untersuchungen zur Endredaktion des Pentateuch (FRLANT 186; Göttingen: Vandenhoeck & Ruprecht, 2000), 380–388; E. Blum, “Die literarische Verbindung von Erzvätern und Exodus: Ein Gespräch mit neueren Endredaktionshypothesen,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 119–156.

Jacob, Moses, Levi

1001

teuch.10 But can Hos 12 indeed bear the burden of proof that a written precursor of the Jacob account existed in the eighth century? And can it indeed bear the burden of proof that a literary connection between the ancestorsʼ and the exodus accounts existed at this time?11 To answer these questions, the formation of Hos 12 shall be treated first. The aforementioned passages referring to the Jacob and exodus stories – i.e., Hos 12:3–7*, 13–14 – can be shown to be secondary additions to this chapter.12 This conclusion is supported, first, by the fact that Hos 12:3–7* interrupts the close connection of the surrounding judgment oracles in 12:1–2 and 12:8–12.13 After the first judgment oracle in Hos 12:1–2, the retrospect of Jacobʼs life is introduced in 12:3 with a new superscription. The subsequent judgment oracle in 12:8–12 against deceit in trade does not really connect with the end of 12:3–7*, where Jacob is called to return and do justice, but it easily connects with the word against political deceit in 12:1–2.14  Thus J. S. Baden, “The Continuity of the Non-Priestly Narrative from Genesis to Exodus,” Bib 93 (2012), 161–186, here 179. 11  See Schmid, Erzväter (see n. 9), 82–84; idem, “Genesis and Exodus as Two Formerly Independent Traditions of Origins for Ancient Israel,” Bib 93 (2012), 187–208, here 190–191. 12  It has often been presumed that Hos 12 is not a literary unity. According to B. D uhm , “Anmerkungen zu den Zwölf Propheten,” ZAW 31 (1911), 1–43, 81–110, 161–204, here 1–37, the primary layer of the chapter consists of the judgment oracles in Hos 12:1–2, 8–12, 15, whereas the retrospects of the Jacob and exodus accounts in 12:3–7, 13–14 are later additions; see Marti, Dodekapropheton (see n. 3), 91–98 (primary layer: 12:1a, 2aαb, 3b–4, 8–9a, 12, 15; later additions: 12:1b, 2aβ, 3a, 5–7, 9b–11, 13–14); J. Vollmer, Geschichtliche Rückblicke und Motive in der Prophetie des Amos, Hosea und Jesaja (BZAW 119; Berlin: de Gruyter, 1971), 111–113 (primary layer: 12:1–4, 8–10, 12, 15; later additions: 12:5–7, 11, 13–14). Other scholars see it the other way around; they take the adaptations of the Jacob and exodus stories as a primary layer and the judgment oracles as later additions; see Köhlmoos, Bet-El (see n. 2), 143 (primary layer: 12:3–5, 7, 13–14, [15]; secondary additions: 12:1–2, 8–12); R. Vielhauer, Das Werden des Buches Hosea: Eine redaktionsgeschichtliche Untersuchung (BZAW 349; Berlin: de Gruyter, 2007), 178 (primary layer: 12:3–5, 7, 13–14; secondary additions: 12:1–2, 6, 8–12, 15); see also Jeremias, Prophet Hosea (see n. 3), 150. According to Pfeiffer, Heiligtum (see n. 3), 68–70, two originally separate oracles (12:1–2, 8–9, 10, 12, 15 and 12:3–5, [6], 7, 11, 13–14) have been combined in this chapter. 13  Thus already Pfeiffer, Heiligtum (see n. 3), 69. 14  These observations speak strongly against Blumʼs new pleading for the literary unity of Hos 12; see Blum, “Hosea 12” (see n. 3), 292–299. Blum discusses and refuses only those approaches that take the retrospects of the Jacob and exodus stories in 12:3–7*, 13–14 as the primary layer of the chapter (see above, note 12). Against them, he rightly argues that these verses cannot stand on their own since the intention of these retrospects does not become apparent without the surrounding judgment oracles. However, this insight does not speak against the opposite view, according to which the judgment oracles in 12:1–2, 8–12, 15 are the primary layer of the chapter and the retrospects are later additions. And because of the above-mentioned observations – the new superscription in 12:3 and the fact that 12:8 does not really connect with 12:7 but easily connects with 12:2 – this approach seems to be more adequate than Blumʼs effort to adhere to the literary unity of Hos 12. 10

1002

Jakob Wöhrle

Several differences in form and content between the judgment oracles in Hos 12:1–2, 8–12, 15 and the retrospects of the Jacob and exodus stories in 12:3–7*, 13–14 also suggest the secondary character of the latter. First, the judgment oracles are written in divine speech. The retrospects of the Jacob and exodus stories, however, are written in prophetic speech; they mention God in the third person. Moreover, the judgment oracles in Hos 12:1–2, 8–12, 15 take up the catchword ‫מרמה‬, “deceit” (12:1, 8; cf. 12:15) several times. In particular, this catchword connects the oracle against political alliances in 12:1–2 and the oracle against economic misdeeds in 12:8–12. The intermediate adaptation of the Jacob story in Hos 12:3–7* as well as the retrospect of Jacob and the exodus in 12:13–14, in contrast, do not use the catchword “deceit.” Finally, the judgment oracles in Hos 12:1–2, 8–12, 15 designate Godʼs people with the term Ephraim (12:1, 2, 9, 15). The term Jacob, although it would be obvious in this context, is not used in these passages. Conversely, the Jacob and exodus retrospects in Hos 12:3–7*, 13–14 do not use the term Ephraim. All these observations taken together speak strongly for the conclusion that Hos 12:3–7*, 13–14 are secondary additions to the chapter. These words hardly trace back to the prophet Hosea. Rather, they stem from a later time. Against this background, the concrete references to the Jacob and exodus traditions within Hos 12:3–7*, 13–14 are noteworthy. As is well-known, these passages show close connections to the Jacob story of the book of Genesis. This is already true for the phrase ‫בבטן עקב את אחיו‬, “in the womb he overreached his brother,” in 12:4. This phrase takes up the term ‫בבטן‬, “in the womb,” from the story of Jacobʼs birth in Gen 25:24 as well as the verb ‫עקב‬, “overreach,” from the story about the loss of the birthright in Gen 27:36.15 The verb ‫ עקב‬establishes a further link to the birth story with its portrayal of Jacob holding the heel (‫)עקב‬ of his brother, Esau, in Gen 25:26.16 Hosea 12:4b thus takes up the portrayal of Jacobʼs overreaching of his brother from the book of Genesis and presents this as a behavior that determined the relationship between Jacob and Esau from the time before their birth. Hosea 12:4b, 5a then takes up, as commonly assumed, the story about Jacobʼs struggle at Penuel in Gen 32:23–33.17 In Hos 12:4b, the phrase ‫שרה את אלהים‬, “he struggled with God,” is nearly a word-for-word parallel to Gen 32:29 (‫שרית‬ 15  W olff , Hosea (see n. 2), 274; R udolph , Hosea (see n. 2), 228; J eremias , Prophet Hosea (see n. 3), 152; E. Bons, Das Buch Hosea (Neuer Stuttgarter Kommentar, Altes Testament 23/1; Stuttgart: Katholisches Bibelwerk, 1996), 150–151; Köhlmoos, Bet-El (see n. 2), 144; et al. 16  See G ertner , “Masorah” (see n. 4), 276; J eremias , Prophet Hosea (see n. 3), 152; Bons, Buch Hosea (see n. 15), 150–151; et al. 17  S ellin , Hosea–Micha (see n. 2), 122; W olff , Hosea (see n. 2), 274–275; R udolph , Hosea (see n. 2), 228; F. I. Andersen and D. N. Freedman, Hosea: A New Translation with Introduction and Commentary (AB 24; New York: Doubleday, 1980), 607–614; J eremias ,

Jacob, Moses, Levi

1003

‫עם אלהים‬, “you have struggled with God”). Hosea 12:5a is connected with Gen 32 by a second adaptation of the verb ‫ שרה‬and by the term ‫ויכל‬, “he prevailed” (cf. ‫ותוכל‬, “you prevailed”; Gen 32:29). Since the verb ‫ שרה‬is attested in the Hebrew Bible only in Gen 32:29 and Hos 12:4–5, it becomes obvious here that Hos 12, with its references to the Jacob story, presupposes a written form and not just, as is sometimes assumed, an oral version of the Jacob story, which differs from the Jacob story of the book of Genesis.18 The subsequent phrase in Hos 12:5b – ‫בית אל ימצאו ושם ידבר עמנו‬, “at Bethel he found him, there he spoke with him” – is remarkable. In previous research, it has commonly been assumed that this phrase refers to the story about the theophany at Bethel in Gen 28:10–22.19 But this assumption is problematic. First, the preceding text of Hos 12:4b, 5a already refers to the story about Jacobʼs struggle at Penuel from Gen 32:23–33. Thus, a reference in Hos 12:5b to the theophany from Gen 28 would come too late.20 Additionally, the presentation Prophet Hosea (see n. 3), 153; Bons, Buch Hosea (see n. 15), 151; Köhlmoos, Bet-El (see n. 2), 144; et al. 18  Also remarkable is the fact that the Penuel story of Gen 32:23–33 uses the verb ‫ שרה‬only in Gen 32:29 in the course of the etiological explanation of the name Israel. In the preceding portrayal of Jacobʼs struggle with an unknown man, the story uses the verb ‫( אבק‬32:25, 26). Thus, it seems that an older Penuel story using the verb ‫( אבק‬32:23–27, 30b, 31–33) has here been secondarily combined with an explanation of the ancestorʼs name via the term ‫שרה‬ (32:28–29, 30a); see J. Wöhrle, Fremdlinge im eigenen Land: Zur Entstehung und Intention der priesterlichen Passagen der Vätergeschichte (FRLANT 246; Göttingen: Vandenhoeck & Ruprecht, 2012), 89–90. The verb ‫ שרה‬is therefore a rather specific feature of the Penuel story in the current version of Gen 32:23–33. Hos 12:3–7ʼs use of the verb ‫ שרה‬in its adaption of the Penuel story can therefore be taken as crucial evidence that the Hosea passage adapts the Jacob story of the book of Genesis. See Wellhausen, Die kleinen Propheten (see n. 6), 128; S ellin , Hosea–Micha (see n. 2), 122; G ertner , “Masorah” (see n. 4), 275–279; W olff , Hosea (see n. 2), 274–276; R udolph , Hosea (see n. 2), 227–228; N eef , Heilstraditionen (see n. 3), 42–43; P feiffer , Heiligtum (see n. 3), 99; K öhlmoos , Bet-El (see n. 2), 147; and B lum , “Hosea 12” (see n. 3), 313, who argue that Hos 12:3–7 draws upon an (earlier) written form of the book of Genesis. Contra M arti , Dodekapropheton (see n. 3), 94–95; Good, “Hosea” (see n. 5), 150; Mays, Hosea (see n. 4), 162; Ruppert, “Herkunft” (see n. 3), 502–503; E. O tto , “Jakob in Bethel: Ein Beitrag zur Geschichte der Jakobüberlieferung,” ZAW 88 (1976), 165–190, here 176, n. 60; and Whitt, “Jacob Traditions” (see n. 4), 41–43, who suppose that the Hosea passage draws upon a different, presumably oral, version of the Jacob story. They make this argument because of some differences between Hos 12:3–7 and the Genesis account, mainly because of the allegedly different order of the Penuel story and the Bethel theophany in Hos 12:4–5 (see below). 19  S ellin , Hosea–Micha (see n. 2), 122; W olff , Hosea (see n. 2), 275–276; R udolph , Hosea (see n. 2), 229; Ruppert, “Herkunft” (see n. 3), 497–498; Andersen and Freedman, Hosea (see n. 17), 600; Jeremias, Prophet Hosea (see n. 3), 154; Pfeiffer, Heiligtum (see n. 3), 78; Köhlmoos, Bet-El (see n. 2), 144, 147; et al. 20  Scholars often used to recognize that Hos 12:4–5, with its references to Jacobʼs struggle and to a theophany in Bethel, does not fit the order of the book of Genesis with the Bethel theophany in Gen 28:10–22 and the Penuel story in Gen 32:23–33. This finding led to different

1004

Jakob Wöhrle

of the words that God speaks to Jacob at Bethel in Hos 12:7 presupposes that the theophany mentioned in 12:5b happens on Jacobʼs way back to the land. In this verse, God calls Jacob to return with the help of his god (‫)באלהיך תשוב‬. The theophany of Gen 28, however, happens on Jacobʼs flight out of the land and not on his return. Thus, reference to a theophany happening after Jacobʼs struggle at Penuel in Hos 12:4b, 5, 7, on the way back to the land, does not really fit the theophany of Gen 28. Remarkably enough, the Jacob account in Genesis reports a second theophany at Bethel, which indeed happens after the Penuel scene on Jacobʼs way back to the land in Gen 35:9–13. This text narrates that God appeared to Jacob at Bethel on his return from his eastern relatives, that God spoke to him, and that he finally went up from him. However, whereas all of the aforementioned references to the Jacob account in Hos 12 refer to non-Priestly texts, the theophany in Gen 35:9–13 is generally held to be part of the Priestly passages of the book of Genesis.21 This seems to be why no one has ever seriously considered that Hos 12:5b could refer to this text.22 But, if we understand the retrospects of the Jacob story in Hos 12:3–7*, 13–14 as later additions to this chapter, the obvious conclusion that Hos 12:5bʼs reference to a theophany at Bethel on Jacobʼs way back to the land is a reference to Gen 35:9–13 seems to be more than probable. explanations: Sellin, Hosea–Micha (see n. 2), 122; Gertner, “Masorah” (see n. 4), 274–275; R udolph , Hosea (see n. 2), 227–228; P feiffer , Heiligtum (see n. 3), 99; and K öhlmoos , Bet-El (see n. 2), 147, point to the authorʼs creative freedom. M arti , Dodekapropheton (see n. 3), 94–95; G ood , “Hosea” (see n. 5), 150; M ays , Hosea (see n. 4), 162; R uppert , “Herkunft” (see n. 3), 502–503; O tto , “Jakob” (see n. 18), 176, n. 60; and W hitt , “Jacob Traditions” (see n. 4), 41–43, think that the author of Hos 12:4–7 refers to a version of the Jacob account which differs from that in Genesis (see above, n. 18). Neef, Heilstraditionen (see n. 3), 41–42, argues that Hos 12:5b refers to Jacobʼs second stay in Bethel documented in Gen 35:1–8 (see below, n. 22). 21  T. Nöldeke, Untersuchungen zur Kritik des Alten Testaments (Kiel: Schwers, 1869), 27; M. N oth , Überlieferungsgeschichte des Pentateuch (Stuttgart: Kohlhammer, 1948), 18; D. M. Carr, Reading the Fractures of Genesis: Historical and Literary Approaches (Louisville: Westminster John Knox, 1996), 88–90; R. G. Kratz, Die Komposition der erzählenden Bücher des Alten Testaments: Grundwissen der Bibelkritik (Göttingen: Vandenhoeck & Ruprecht, 2000), 242, and many others. For a detailed analysis of Gen 35, see also Wöhrle, Fremdlinge (see n. 18), 85–95. 22  As mentioned above (n. 20), Neef, Heilstraditionen (see n. 3), 41–42, supposes that Hos 12:5b refers to the portrayal of Jacobʼs second stay in Bethel in Gen 35:1–8. However, these verses, which are part of the non-Priestly stratum, do not mention a theophany. A theophany in Bethel on Jacobʼs way back to the land is presented only in the subsequent Priestly verses 35:9–13. That Hos 12:5 could refer to this text has been supposed only by Gertner , “Masorah” (see n. 4), 278–279. But Gertner ascribes Hos 12:4–7 to the prophet Hosea and does not discuss the fact that Gen 35:9–13 is commonly seen as a Priestly and thus as a rather late (post-Hoseanic) pentateuchal text.

Jacob, Moses, Levi

1005

Some further arguments can be put forward for this reading. The phrase ‫בית‬ ‫אל ימצאנו‬, “at Bethel he found him,” in Hos 12:5b indicates a direct encounter with God. This fits poorly with the story in Gen 28:10–22, where God reveals himself to Jacob in a dream standing at the top of a ladder that reaches the heaven. Hosea 12:5b, with its presentation of a direct encounter with God, fits much better with the story of Gen 35:9–13, according to which God appears to Jacob, speaks to him, and goes up from him. The formulation ‫ושם ידבר עמנו‬, “there he spoke with him,” in Hos 12:5b is also very close to the phrase ‫מקום‬ ‫אשר דבר אתו‬, “the place where he had spoken to him,” in Gen 35:13.23 Thus, there is every indication that the notice about a theophany at Bethel in Hos 12:5b takes up the Priestly text in Gen 35:9–13. If this argumentation is correct, the result is noteworthy: Hos 12:3–7*, with its references to the Jacob story, does not only – in a rather general way – presuppose a written form of the Jacob account from the book of Genesis. Rather, it presupposes a form of the Jacob account that already combined non-Priestly and Priestly texts. The further references to the Jacob and exodus accounts in Hos 12:13–14 are remarkable against this background. Hosea 12:13 mentions Jacobʼs flight to the fields of Aram and his service there for two wives. This notice, without doubt, first and foremost alludes to the story about Jacobʼs flight to his eastern relatives in Gen 27 and to the story about his service for Rachel and Leah in Gen 29.24 Hosea 12:13 is connected to these stories by two catchwords, the verbs ‫( ברח‬Gen 27:43) and ‫( עבד‬Gen 29:15, 18, 20, 25, 27, 30). Besides these references to the stories of Gen 27 and Gen 29, both of which are part of the non-Priestly layer of the Jacob account, Hos 12:13 again contains a rather distinctive reference to the Priestly passages of the Jacob account. Hosea 12:13 documents the phrase ‫שדה ארם‬, “fields of Aram.” This term is noteworthy, since the non-Priestly passages of the Jacob account designate Jacobʼs uncle Laban as Aramean,25 but they never designate his dwelling place with the geographical term Aram; instead, they speak of Haran.26 The very specific phrase ‫ שדה ארם‬in Hos 12:13, however, strongly evokes the phrase ‫פדן ארם‬, Paddan Aram, which the Priestly passages of Genesis use to designate the dwelling

23  The formulation ‫ ושם ידבר עמנו‬in Hos 12:5b is not only close to ‫ מקום אשר דבר אתו‬in Gen 35:13, it is even closer to Gen 35:15, ‫המקום אשר דבר אתו שם אלהים‬. Since Gen 35:14–15 is most likely a late post-Priestly addition to Gen 35:9–13 (see W öhrle , Fremdlinge [see n. 18], 91), this could speak for the assumption that Hos 12:5b presupposes not only the Priestly version of the ancestorsʼ account but rather a post-Priestly version. 24  S ellin , Hosea–Micha (see n. 2), 126; W olff , Hosea (see n. 2), 279; A ndersen and F reedman , Hosea (see n. 17), 620; J eremias , Prophet Hosea (see n. 3), 157; B ons , Buch Hosea (see n. 15), 157; et al. 25  Gen 25:20; 28:5; 31:20, 24. 26  Gen 27:43; 28:10; 29:4.

1006

Jakob Wöhrle

place of the eastern relatives.27 Interestingly enough, it has often been proposed – albeit without positing a direct connection – that the phrase ‫ שדה ארם‬in Hos 12:13 is nothing other than the Hebrew equivalent for the Priestly phrase ‫פדן‬ ‫ארם‬,28 which is obscure in both origin and meaning. Thus, the phrase ‫שדה ארם‬ in Hos 12:13 can be taken as further evidence that the retrospects of the Jacob story in Hos 12:3–7*, 13–14 presuppose a form of the Jacob account that has already combined the non-Priestly and Priestly texts. The reference in Hos 12:14 to YHWH bringing Israel up from Egypt through a prophet is also noteworthy in that it designates Moses as prophet. Besides Hos 12:14, Moses is designated as prophet only within the book of Deuteronomy, namely, in Deut 18:15, 18 in the context of the law about prophets and in Deut 34:10 in the context of Mosesʼs death. Thus it has frequently been assumed that the designation of Moses as prophet in Hos 12:14 is influenced by the book of Deuteronomy, which is why this verse is sometimes considered a Deuteronomistic addition to the book of Hosea.29 Against this view, however, it has been objected that the person of Moses in Hos 12:14 has another function than he has in the Deuteronomy passages.30 Whereas in Hos 12:14 the function of the prophet Moses consists of leadership of the people, in Deuteronomy his prophetic task is the proclamation of Godʼs word. This difference between Hos 12:14 and Deuteronomy is then taken as an argument for an early, pre-Deuteronomic dating of Hos 12:14.31 It is certainly true that the prophetic task of Moses differs in Hos 12:14 and in the book of Deuteronomy, but the question remains, whether it would have been possible for Hos 12:14 to designate Moses (without mentioning his name) as a prophet without knowledge of a tradition presenting Moses as prophet. The reference to a prophet who leads Israel out of Egypt seems to presuppose that a tradition of Moses as prophet was already known. Deuteronomy is the only place in the Hebrew Bible to present Moses as a prophet. Hosea 12:14 thus seems to  Gen 25:20; 28:2, 5, 6, 7; 31:18; 33:18; 35:9, 26; 46:15; cf. 48:7.  H. Gunkel, Genesis (3rd ed.; HKAT 1/1; Göttingen: Vandenhoeck & Ruprecht, 1964), 385; R. de Vaux , “Les patriarches hébreux et les découvertes modernes,” RB 55 (1948), 321–347, here 323, n. 2; HAL 864; M. G örg , “Paddan-Aram,” Neues Bibel-Lexikon 3:58; W. Gesenius, Hebräisches und aramäisches Handwörterbuch über das Alte Testament (18th ed.; ed. H. Donner; Berlin: Springer, 2013), 1039; et al.; but see also the critical remarks by W. T. Pitard, “Paddan-Aram,” ABD 5:55. 29  See Marti, Dodekapropheton (see n. 3), 98; Vollmer, Geschichtliche Rückblicke (see n. 12), 110; Pfeiffer, Heiligtum (see n. 3), 94; Vielhauer, Werden (see n. 12), 179. 30  W hitt , “Jacob Traditions” (see n. 4), 26; S chmitt , “Kampf” (see n. 2), 172; B lum , “Hosea 12” (see n. 3), 316–317. 31  An early, pre-Deuteronomic dating of Hos 12:14 is also advocated by Wolff, Hosea (see n. 2), 280; Rudolph, Hosea (see n. 2), 231, n. 28; Mays, Hosea (see n. 4), 170; L. Perlitt, “Mose als Prophet,” in Deuteronomium-Studien (FAT 8; Tübingen: Mohr Siebeck, 1994), 1–19, here 16; Jeremias, Prophet Hosea (see n. 3), 157; Neef, Heilstraditionen (see n. 3), 55; Bons, Buch Hosea (see n. 15), 159; et al. 27 28

Jacob, Moses, Levi

1007

be influenced by the book of Deuteronomy and its presentation of Moses as a prophet. Hosea 12:3–7*, 13–14, then, with its retrospects of the Jacob and exodus accounts, not only presupposes a form of the Pentateuch that has already combined non-Priestly and Priestly texts, it also presupposes a form of the Pentateuch that combines non-Priestly texts, Priestly texts, and the book of Deuteronomy. Hosea 12:3–7*, 13–14 is a late, post-Deuteronomic, post-Priestly, even postpentateuchal addition to the book of Hosea. The adaptation of the Jacob and exodus stories in Hos 12:3–7*, 13–14 can thus not be taken as an argument for an early form of the Jacob account that already existed in the eighth century. This text can also not be taken as an argument that a literary connection between the ancestorsʼ account and the exodus story existed in this time. Hosea 12:3–7*, 13–14 rather shows how only in postexilic times, after the completion of the Pentateuch, the traditions of the Pentateuch have been taken up and have been integrated into a prophetic scripture.

2 Moses and the Exodus in Micah 6 Another adaptation of pentateuchal traditions can be found in Mic 6. The chapter presents a comprehensive judgment speech, which first blames the people generally for not adhering to YHWH and then mentions some concrete transgressions, especially economic misdeeds. In this context, Mic 6:3–5 reads as follows: Micah 6:3–5 My people, what have I done to you, wherein have I made you tired of me? Answer me! For I brought you up from the land of Egypt, I redeemed you (‫ )פדה‬from the house of bondage (‫)בית עבדים‬. And I sent before you Moses, Aaron, and Miriam. My people, remember what Balak king of Moab planned, and how Balaam the son of Beor answered him, from Shittim to Gilgal (‫)מן השטים עד הגלגל‬, that you may know the saving deeds of YHWH.

Micah 6:3–5, after the introductory question of what YHWH has done to his people, refers to several occurrences around the exodus. It first mentions generally that YHWH has brought his people up from Egypt, from the house of bondage. It then points to the leadership of Moses, Aaron, and Miriam and to the commission of the prophet Balaam by the Moabite king Balak. These references to the exodus account are remarkable not least because they refer, as will be shown in detail, to certain events documented in the book of Numbers. Micah 6:3–5 thus could be taken as evidence that a precursor of the book of Numbers already existed by the time of the prophet Micah, i.e., the

1008

Jakob Wöhrle

eighth century.32 This would speak strongly against the view put forward in some new approaches to the formation of the Pentateuch that the book of Numbers was created in the postexilic period in order to build a bridge between the books of Leviticus and Deuteronomy.33 Thus the formation of Mic 6 becomes relevant. Since the groundbreaking work of Bernhard Stade in 1881, there has been a relatively broad consensus that the literary core of the book of Micah is to be found in Mic 1–3.34 Most parts of Mic 4–7, and thus also chapter 6, are commonly seen as later additions to the book.35 With regard to Mic 6:3–5, it can be shown that these verses – or at least their primary layer – trace back to a Deuteronomistic redaction, in the broad sense.36 This is indicated by the phrase ‫בית עבדים‬, “house of bondage,” and by use of the verb ‫פדה‬, “redeem,” in the context of the exodus tradition, both of which are mainly documented in Deuteronomic or Deuteronomistic texts.37 Thus, a Dtr redaction of the book of Micah becomes apparent, which (although it cannot be shown in detail here) is to be dated to the time of the late exile.38 But Mic 6:3–5 is not a literary unity. This text can be divided into a primary layer in Mic 6:3, 4a, with its appeal to the people and its remark that God has  See B aden , “Continuity” (see n. 10), 179, who mentions Mic 6:4–5 among some extrapentateuchal references that he sees as evidence for early pentateuchal traditions. 33  T. Römer, “Das Buch Numeri und das Ende des Jahwisten: Anfragen zur ‘Quellenscheidung’ im vierten Buch des Pentateuch,” in Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (ed. J. C. Gertz et al.; BZAW 315; Berlin: de Gruyter, 2002), 215–231; R. A lbertz, “Das Buch Numeri jenseits der Quellentheorie: Eine Redaktionsgeschichte von Num 20–24,” ZAW 123 (2011), 171–183, 336–347, here 336–338. 34  B. S tade , “Bemerkungen über das Buch Micha,” ZAW 1 (1881), 161–172; cf., e.g., J. L. Mays, Micah: A Commentary (OTL; London: SCM, 1976), 23–24; H. W. Wolff, Micha (BKAT 14/4; Neukirchen-Vluyn: Neukirchener Verlag, 1982), xxxvi–xxxvii; R. K essler , Micha (HThKAT; Freiburg im Breisgau: Herder, 1999), 41–42; K. Schmid, “Hintere Pro­phe­ ten (Nebiim),” in Grundinformation Altes Testament: Eine Einführung in Literatur, Religion und Geschichte des Alten Testaments (ed. J. C. Gertz; 4th ed.; UTB 2745; Göttingen: Vandenhoeck & Ruprecht, 2010), 313–412, here 396; Carr, Formation (see n. 8), 329; E. Zenger, Einleitung in das Alte Testament (ed. C. Frevel; 8th ed.; Stuttgart: Kohlhammer, 2012), 666. 35  For a detailed analysis of the book of Micah, see Wöhrle, Die frühen Sammlungen (see n. 1), 138–197. 36  W olff , Micha (see n. 34), 143–144; E. O tto , “Techniken der Rechtssatz-Redaktion israelitischer Rechtsbücher in der Redaktion des Prophetenbuches Micha,” SJOT 2 (1991), 119–150, here 148; Nogalski, Literary Precursors (see n. 1), 141–144; Schart, Entstehung (see n. 1), 191–197; Wöhrle, Die frühen Sammlungen (see n. 1), 174–175. 37  For the phrase ‫בית עבדים‬, “house of bondage,” see Exod 13:3, 14; 20:2; Deut 5:6; 6:12; 7:8; 8:14; 13:6, 11; Josh 24:17; Judg 6:8; Jer 34:13. For the verb ‫פדה‬, “redeem,” in the context of the exodus tradition, see Deut 7:8; 9:26; 13:6; 15:15; 24:18; 2 Sam 7:23. 38  Concerning the date of the Dtr redaction of the book of Micah, which is part of a broader redaction responsible for the formation of a Book of the Four (Hosea, Amos, Micah, and Zephaniah), see Wöhrle, Die frühen Sammlungen (see n. 1), 193, 272–275. 32

Jacob, Moses, Levi

1009

brought his people out of Egypt, and a secondary layer in Mic 6:4b, 5 referring to the leadership of Moses, Miriam, and Aaron and to the commission of Balaam.39 This division is supported by the sudden change from perfect forms in 6:4a to past narrative forms in 6:4b. Additionally and mainly, this division is supported by the fact that after the peopleʼs exodus mentioned in 6:4a Godʼs sending of Moses, Aaron, and Miriam in 6:4b comes too late. All other passages within the Hebrew Bible that mention the leading role of Moses and Aaron mention their appointment before the exodus.40 Thus, Mic 6:4b, 5 is a secondary addition to the preceding Deuteronomistic verses Mic 6:3, 4a. It is therefore an even later, post-Deuteronomistic and consequently postexilic expansion of the book of Micah. Against this background, the concrete references to the Pentateuch in Mic 6:4b, 5 are important. These verses first and foremost show significant parallels to the book of Numbers. The reference to Moses, Aaron, and Miriam in Mic 6:4b is remarkable because these three persons are mentioned together only in Num 12, within the story about a conflict over the leadership of the people. It is thus highly probable that Mic 6:4b depends upon Num 12.41 The commission of Balaam in Mic 6:5 then surely alludes to the Balaam account in Num 22–24.42 Noteworthy is the subsequent note in Mic 6:5b, ‫מן השטים עד הגלגל‬, “from Shittim to Gilgal.” This phrase refers to the last stage on the peopleʼs way from Egypt to the promised land. According to Num 25:1, Shittim was the last dwelling place in Transjordan; according to Josh 4:19, Gilgal was the first dwelling place after their entry into the land. Thus, with its reference to both Shittim and Gilgal, Mic 6:5b refers to traditions not only from the book of Numbers but also from the book of Joshua.43 Micah 6:5b thus presupposes both books and, with the book of Joshua, probably a form or preform of the Deuteronomistic History. 39  Thus already M arti , Dodekapropheton (see n. 3), 291; T. L escow , “Redaktions­ge­ schichtliche Analyse von Micha 6–7,” ZAW 84 (1972), 182–212, here 185–186; cf. Wöhrle, Die frühen Sammlungen (see n. 1), 174. 40  Josh 24:5–6; 1 Sam 12:6, 8; Ps 105:26–27. 41  Kessler, Micha (see n. 34), 264; F. I. Andersen and D. N. Freedman, Micah: A New Translation with Introduction and Commentary (AB 24E; New York: Doubleday, 2000), 519; E. Noort, “Balaam the Villain: The History of Reception of the Balaam Narrative in the Pentateuch and the Former Prophets,” in The Prestige of the Pagan Prophet Balaam in Judaism, Early Christianity and Islam (ed. G. H. van Kooten and J. van Ruiten; TBN 11; Leiden: Brill, 2008), 3–23, here 11, n. 33. 42  Marti, Dodekapropheton (see n. 3), 293; Wolff, Micha (see n. 34), 150; W. Rudolph, Micha – Nahum – Habakuk – Zephanja (KAT 13/3; Gütersloh: Gütersloher Verlagshaus, 1975), 111; Kessler, Micha (see n. 34), 265; H. Utzschneider, Micha (ZBK 24/1; Zürich: Theologischer Verlag, 2005), 133; J. J eremias , Die Propheten Joel, Obadja, Jona, Micha (ATD 24/3; Göttingen: Vandenhoeck & Ruprecht, 2007), 201. 43  Sellin, Hosea–Micha (see n. 2), 342; Wolff, Micha (see n. 34), 150–151; Rudolph, Micha (see n. 42), 110, n. 9; K essler , Micha (see n. 34), 266; J eremias , Propheten (see n. 42), 202, n. 244.

1010

Jakob Wöhrle

Previous research commonly understood the reference to Balaamʼs commission in Mic 6:5a and the subsequent “from Shittim to Gilgal” in 6:5b to refer to two separate, independent occurrences on the way from Egypt to the land. Micah 6:5 is generally read as referring first, in 6:5a, to the hostilities of Balak and then, in 6:5b, to the peopleʼs entry into the land.44 The problem with this view is that the phrase “from Shittim to Gilgal” is by no means marked as referring to a new fact that has to be distinguished from the reference to Balaamʼs commission. The note “from Shittim to Gilgal” rather seems to be a geographical qualification of the aforementioned events. It reveals that the commission of Balaam took place while the people were on their way “from Shittim to Gilgal,” i.e., directly before their passage into the promised land. Interestingly enough, in Num 22:1 and thus directly before the Balaam account in Num 22:2–24:25, one finds the following note: Numbers 22:1 The Israelites set out and encamped in the steppes of Moab, at the Jordan, beyond Jericho.

According to Num 22:1, the Israelite people were located at the boundary between Transjordan and the land when the offensive acts of King Balak mentioned in the subsequent Balaam story occurred. The note “from Shittim to Gilgal” in Mic 6:5, which associates Balakʼs hostilities with the situation of imminent entry into the land, thus seems to presuppose and adapt this note in Num 22:1. However, whereas the Balaam account beginning in Num 22:2 is usually assigned to the non-Priestly stratum, Num 22:1 is commonly assigned to the Priestly stratum.45 The note “from Shittim to Gilgal” in Mic 6:5b thus depends upon a form of the book of Numbers that had already combined non-Priestly and Priestly texts.46 The references to the exodus events in Mic 6:4b, 5, then, constitute a late, post-Deuteronomistic addition to the book that presupposes not only the non Marti, Dodekapropheton (see n. 3), 293; Sellin, Hosea–Micha (see n. 2), 342; Wolff, Micha (see n. 34), 150–151; Rudolph, Micha (see n. 42), 110; Kessler, Micha (see n. 34), 266; U tzschneider , Micha (see n. 42), 133; J eremias , Propheten (see n. 42), 201–202; Noort, “Balaam” (see n. 41), 11. 45  Nöldeke, Untersuchungen (see n. 21), 86; M. Noth, Das 4. Buch Mose: Numeri (ATD 7; Göttingen: Vandenhoeck & Ruprecht, 1966), 151; H. Seebass, Numeri 10,11–22,1 (BKAT 4/2; Neukirchen-Vluyn: Neukirchener Verlag, 2003), 366; L. Schmidt, Numeri 10,11–36,13 (ATD 7/2; Göttingen: Vandenhoeck & Ruprecht, 2004), 122; Albertz, “Buch Numeri” (see n. 33), 182, 339; et al. 46  Without the Priestly verse Num 22:1, the story of the book of Numbers would locate the dwelling place of the Israelites during the events of the subsequent Balaam story in Bashan (Num 21:33) or – since Num 21:33–35 is likely a very late Dtr addition to the text; see Noth, Das 4. Buch Mose (see n. 45), 142; Schmidt, Numeri 10,11–36,13 (see n. 45), 114; Albertz, “Buch Numeri” (see n. 33), 182; et al. – “in the field of Moab, by the top of Pisgah” (Num 21:20). Thus, Mic 6:5 seems to presuppose Num 22:1 (P), as it is the only verse to locate the Israelites at the crossover from Transjordan to the land during the events of the Balaam story. 44

Jacob, Moses, Levi

1011

Priestly passages of the book of Numbers but also the book of Joshua and thus the Deuteronomistic History, as well as the Priestly passages of the book of Numbers. Like Hos 12:3–7*, 13–14, with its retrospects of the Jacob and exodus accounts, Mic 6:4b, 5 is a late, post-Deuteronomic (or post-Deuteronomistic), post-Priestly, and even postpentateuchal text. It can therefore not be taken as evidence of an early date for the pentateuchal traditions. It especially cannot be taken as an argument against the assumption of some new approaches that the book of Numbers is a late literary bridge between the books of Leviticus and Deuteronomy. Rather, Mic 6:4b, 5 again documents an adaptation of Pentateuchal traditions that happened not before the postexilic time.

3 The Covenant with Levi in Malachi 2 The last reference to pentateuchal figures in the Book of the Twelve that shall be treated in this article is to be found in the book of Malachi. In Mal 1:6–2:9 one finds a disputation speech that, in its present shape, condemns transgressions of the priesthood. Malachi 1:6–14 opposes certain wrongs with the offering of sacrifices. Malachi 2:1–9 is directed against misbehavior in communicating the Torah. In this context, Mal 2:4–6 presents the following words: Malachi 2:4–6 You will know that I sent you this commandment to maintain my covenant (‫ )ברית‬with Levi, says YHWH Sabaoth. My covenant with him was (‫ )היה‬life and peace (‫)שלום‬. I gave him fear, and he feared me. He was afraid before my name. Reliable Torah (‫ )תורה‬was in his mouth, and guilt was not found upon his lips. He walked with me in peace and uprightness, and he turned many away from iniquity.

Malachi 2:4–6 refers to a covenant with Levi, which is characterized as a covenant of life and peace. The main task of Levi under this covenant is the communication of the Torah. The consensus holds that the book of Malachi is a product of the postexilic period.47 The book probably underwent a longer history of re­dac­ M arti , Dodekapropheton (see n. 3), 457; W. R udolph , Haggai – Sacharja 1–8 – Sacharja 9–14 – Maleachi (KAT 13/4; Gütersloh: Gütersloher Verlagshaus, 1976), 248–249; A. E. Hill, “Dating the Book of Malachi: A Linguistic Reexamination,” in The Word of the Lord Shall Go Forth (ed. C. L. Meyers and M. OʼConnor; Winona Lake, IN: Eisenbrauns, 1983), 77–89, here 86; B. Glazier-McDonald, Malachi: The Divine Messenger (SBLDS 98; Atlanta: Society of Biblical Literature, 1987), 14–18; H. G. R eventlow , Die Propheten Haggai, Sacharja und Maleachi (ATD 25/2; Göttingen: Vandenhoeck & Ruprecht, 1993), 130–131; P. L. R edditt , Haggai, Zechariah, Malachi (NCB; London: Marshall Pickering, 47

1012

Jakob Wöhrle

tion.48 It likely received its essential shape in the late fifth century.49 Against this background, the pentateuchal traditions in Mal 2:4–6 are noteworthy.50 An important parallel to the covenant with Levi mentioned in this passage can be found in Num 25:12–13.51 In this text, YHWH says the following about the priest Phinehas, the grandson of Aaron: Numbers 25:12 Behold, I give him my covenant of peace (‫)בריתי שלום‬, and it shall be (‫ )היה‬for him and his descendants after him a covenant of an everlasting priesthood.

The features common to Num 25:12–13, a text commonly grouped with the Priestly passages of the Pentateuch, and Mal 2:4–6 are significant. Both passages mention a covenant (‫ )ברית‬established with a priest: in Num 25 with the Aaronide Phinehas, in Mal 2 with Levi. Both passages designate this covenant as a covenant of peace (‫)שלום‬. Both passages describe this covenant by using the verb ‫היה‬. And both passages emphasize the generation-spanning significance of this covenant. The number and specificity of the parallels make it more than probable that the author of Mal 2:4–6 drew upon Num 25 and developed his concept of a Levite covenant by transferring the covenant with Phinehas to the ancestor Levi.

1995), 149–151; A. M einhold , Maleachi (BKAT 14/8; Neukirchen-Vluyn: Neukirchener Verlag, 2006), xxii–xxiii; R. K essler , Maleachi (HThKAT; Freiburg im Breisgau: Herder, 2011), 75–77. 48  For a detailed analysis of the bookʼs formation, see W öhrle , Abschluss (see n. 1), 219–263. 49  Cf. Wöhrle, Abschluss (see n. 1), 255–263. 50  An overview of scholarly assumptions regarding the adaptation of pentateuchal traditions in Mal 2:4–6 is given by K. W. Weyde, Prophecy and Teaching: Prophetic Authority, Form Problems, and the Use of Traditions in the Book of Malachi (BZAW 288; Berlin: de Gruyter, 2000), 176–186. 51  G lazier -M c D onald , Malachi (see n. 47), 77–80; H. U tzschneider , Künder oder Schreiber? Eine These zum Problem der “Schriftprophetie” auf Grund von Maleachi 1,6–2,9 (BEATAJ 19; Frankfurt: Lang, 1989), 64–70; J. M. OʼB rien , Priest and Levite in Malachi (SBLDS 121; Atlanta: Society of Biblical Literature, 1990), 105–106; R. F uller , “The Blessing of Levi in Dtn 33, Mal 2, and Qumran,” in Konsequente Traditionsgeschichte (ed. R. Bartelmus et al.; OBO 126; Göttingen: Vandenhoeck & Ruprecht, 1993), 31–44, here 38–39; R eventlow , Propheten (see n. 47), 143; M. K rieg , Mutmassungen über Maleachi: Eine Monographie (ATANT 80; Zürich: Theologischer Verlag, 1993), 156–158; Weyde, Prophecy (see n. 50), 183; L.-S. T iemeyer , Priestly Rites and Prophetic Rage: Post-Exilic Prophetic Critique of the Priesthood (FAT 2/19; Tübingen: Mohr Siebeck, 2006), 242–243; C. Frevel, “ ‘Mein Bund mit ihm war das Leben und der Friede’: Priesterbund und Mischehenfrage,” in Für immer verbündet: Studien zur Bundestheologie der Bibel (ed. C. Dohmen and C. Frevel; SBS 211; Stuttgart: Katholisches Bibelwerk, 2007), 85–93, here 90–92; Kessler, Maleachi (see n. 47), 168.

Jacob, Moses, Levi

1013

Malachi 2:4–6 is also commonly understood to adapt a second text from the Pentateuch, namely the blessing of Levi in Deut 33:8–11.52 Verses 9–10 are particularly significant: Deuteronomy 33:9–10 For they have kept your word and held your covenant (‫)ברית‬. They teach Jacob your laws and Israel your Torah (‫)תורה‬. They put incense-offerings before you and whole-offerings on your altar.

Deuteronomy 33:9 mentions a covenant (‫)ברית‬, even though this passage, unlike Num 25:12–13, probably refers not to a specific priestly covenant but to Godʼs covenant with his people in general. Additionally, Deut 33:9–10 presents the communication of the Torah (‫ )תורה‬as the main task of the priests apart from sacrificial ministry. Within the whole Pentateuch, only this passage decidedly says that the communication of the Torah is a priestly duty.53 Thus, it seems highly probable that Mal 2:4–6ʼs presentation of a covenant with Levi that has the priests especially responsible for communicating the Torah also draws upon the blessing of Levi in Deut 33. So, this text again takes up both Priestly and Deuteronomic traditions and, consequently, presupposes a form of the Pentateuch in which both Priestly texts and the book of Deuteronomy are combined. Malachi 2:4–6 can thus be taken as further evidence of a late reception of pentateuchal traditions presupposing the completed or almost completed form of the Pentateuch.

4 Conclusion The previously discussed texts from the Book of the Twelve referring to pentateuchal figures lead to an astonishing result. All of these texts – and further texts could be mentioned54 – are late, post-Deuteronomic, post-Priestly, and even postpentateuchal. They presuppose, at the least, a nearly completed form of the Pentateuch and show how after the Pentateuchʼs completion the traditions preserved in this work were taken up and integrated into prophetic writings. Thus, unlike some previous assumptions, the references to pentateuchal traditions documented in the Book of the Twelve cannot bear the burden of proof for an early date of the pentateuchal traditions. Hosea 12, with its retrospects to the Jacob and the exodus accounts, cannot be taken as evidence that a written 52  R udolph , Haggai (see n. 47), 266; OʼB rien , Priest (see n. 51), 104–105; F uller , “Blessing” (see n. 51), 38; D. L. P etersen , Zechariah 9–14 and Malachi: A Commentary (OTL; London: SCM, 1995), 190; Weyde, Prophecy (see n. 50), 183; Meinhold, Maleachi (see n. 47), 147; Kessler, Maleachi (see n. 47), 168–169. 53  Cf. C. Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 265. 54  Cf. esp. Mic 7:20 (Abraham and Jacob); Mal 1:2–5 (Jacob and Esau); Mal 3:22 (Moses).

1014

Jakob Wöhrle

form of the Jacob story already existed in the eighth century, nor can it be taken as evidence that a literary connection between the ancestral and the exodus accounts existed at this time. And Mic 6, with its references to the exodus and especially to the Balaam story, does not speak for an early, preexilic origin of the book of Numbers. To be sure, the findings within the Book of the Twelve certainly cannot definitively exclude the possibility of an early date for the formation of pentateuchal traditions. However, it is and remains remarkable that within the Book of the Twelve, with its long-term and multilayered history of formation, the reception of pentateuchal traditions does not begin before the later postexilic period but then begins simultaneously at several places within this comprehensive collection of prophetic books. This seems to be an important indication that the crucial formative phase of the Pentateuch was essentially later than presumed in traditional models on the Pentateuchʼs formation.

Ezekiel and the Holiness Legislation A Plea for Nonlinear Models Christophe L. Nihan 1 Introduction: The State of the Discussion The relationship between the Holiness legislation (H) and Ezekiel remains a much disputed issue and one where the dividing lines between various contemporary scholarly traditions are particularly manifest.1 It has long been observed that the book of Ezekiel shares with H many distinctive expressions and motifs, arguably more so than any other prophetic tradition. Furthermore, there is also very broad agreement among scholars that this phenomenon points to some sort of interaction between these two collections. However, no agreement has been reached regarding the nature of the relationship between H and Ezekiel, and despite many recent publications on the topic, the positions have tended to drift further apart rather than to get closer. Although parallels between Lev 17–26 and Ezekiel were already noted by ancient commentators,2 the issue took on new significance after these chapters of Leviticus were identified by K. H. Graf as part of a distinct collection (H) within the Priestly traditions of the Pentateuch.3 Graf himself proposed to explain the similarities between H and Ezekiel by the notion that the prophet was the 1  In this essay, I use the expression Holiness legislation rather than Holiness Code in order to highlight the fact that this collection was arguably never transmitted separately but was composed from the beginning as a supplement to Lev 1–16. See on this the comments by B. J. Schwartz, “The Strata of the Priestly Writings and the Revised Relative Dating of P and H,” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions (ed. S. Shectman and J. S. Baden; ATANT 95; Zürich: Theologischer Verlag, 2009), 1–12. 2  For examples, see M. Lyons, From Law to Prophecy: Ezekielʼs Use of the Holiness Code (LHBOTS 507; New York: T&T Clark, 2009), 1–4. 3  K. H. Graf, Die geschichtlichen Bücher des Alten Testaments: Zwei historisch-kritische Untersuchungen (Leipzig: T. O. Weigel, 1866), 32–85. However, Graf himself assigned only Lev 18–26 to H; the inclusion of Lev 17 goes back to J. Wellhausen, Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (4th ed.; Berlin: de Gruyter, 1963), 150. For a survey of earlier scholarship on H, see C. Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 1–11.

1016

Christophe L. Nihan

author of H, yet this hypothesis found few adherents.4 Instead, A. Kuenen and J. Wellhausen argued that H postdated Ezekiel, from whom this legal collection had borrowed many distinctive expressions;5 this view, which implies an exilic or early postexilic dating of H, soon became dominant. However, it was quickly challenged by some scholars, like A. Klostermann, who maintained a preexilic dating of H and argued accordingly for the reverse relationship between the two collections.6 This model was consistently adopted by other scholars in the twentieth century who similarly endorsed a preexilic origin for H,7 and it has found further support among a number of recent scholars, in particular R. Levitt Kohn and M. Lyons, who argue – albeit from distinct perspectives – that Ezekiel has creatively reused and adapted the language of H.8 Simultaneously, several recent studies of Lev 17–26 (or portions thereof) are characterized by the assumption that H consistently borrows from Ezekiel, rather than the other way round, even though these studies differ with regard to their assessment of this legislation and its place in the formation of the Pentateuch.9 There are many significant issues involved in that discussion, not least regarding the criteria for identifying a literary relationship between two texts and

 Graf, Die geschichtlichen Bücher (see n. 3), 75–83.  A. K uenen , The Religion of Israel to the Fall of the Jewish State (trans. A. H. May; 2 vols.; London: Williams and Norgate, 1874–1875), 2:189–192. Wellhausenʼs dating of H was based on the conviction that this collection constitutes an intermediate stage in a historical sequence extending from Ezekiel to P; see J. W ellhausen , Prolegomena to the History of Ancient Israel (trans. J. Sutherland Black and Allan Enzies; New York: Meridian Books, 1957), 379. 6  A. Klostermann, Der Pentateuch: Beiträge zu seinem Verständnis und seiner Ent­steh­ ungs­geschichte (Leipzig: A. Deichert, 1893), 368–418. 7  See, e.g., L. B. Paton , “The Holiness-Code and Ezekiel,” Presbyterian and Reformed Review 26 (1896), 98–115; L. Elliott-Binns, “Some Problems of the Holiness Code,” ZAW 67 (1955), 26–40, esp. 35. For further references to earlier advocates of a similar view, see Lyons, From Law to Prophecy (see n. 2), 23, n. 27. 8  R. L evitt K ohn , A New Heart and a New Soul: Ezekiel, the Exile and the Torah (JSOTSup 358; Sheffield: Sheffield Academic Press, 2002); idem, “A Prophet Like Moses? Rethinking Ezekielʼs Relationship to the Torah,” ZAW 114 (2002), 236–254; Lyons , From Law to Prophecy (see n. 2). Compare also, e.g., K. L. Wong, The Idea of Retribution in the Book of Ezekiel (VTSup 87; Leiden: Brill, 2001), esp. 80–119. 9  See, e.g., E. O tto , “Innerbiblische Exegese im Heiligkeitsgesetz Levitikus 17–26,” in Levitikus als Buch (ed. H.-J. Fabry; BBB 119; Berlin: Philo, 1999), 125–196; K. Grünwaldt, Das Heiligkeitsgesetz Leviticus 17–26: Ursprüngliche Gestalt, Tradition und Theologie (BZAW 271; Berlin: de Gruyter, 1999); and most recently, R. M üller , “A Prophetic View of the Exile in the Holiness Code: Literary Growth and Tradition History in Leviticus 26,” in The Concept of Exile in Ancient Israel and Its Historical Contexts (ed. E. Ben Zvi and C. Levin; BZAW 404; Berlin: de Gruyter, 2010), 207–228. Compare also my earlier discussion in Nihan, From Priestly Torah (see n. 3), 401–545. 4 5

Ezekiel and the Holiness Legislation

1017

for determining the direction of that relationship.10 It is arguably no exaggeration to state that the scholarly disagreements noted above regarding the relationship between H and Ezekiel betray to a significant extent the methodological difficulties involved in construing meaningful models of literary derivation on the basis of the language shared by these two collections. Many arguments are based on the occurrence of isolated idioms occurring in H and Ezekiel, which do not necessarily provide sufficient evidence for asserting and assessing the literary relationship between these collections. In other words, “shared language” is not automatically equivalent to “textual borrowing.”11 Yet an even more basic issue, in my opinion, is the tendency to conceptualize the relationship between these two collections as a linear relationship exclusively, be it from H to Ezekiel or from Ezekiel to H. The problems with this approach had already been noted by some earlier critics, such as G. Fohrer or W. Zimmerli in his commentary on Ezekiel. Zimmerli, for instance, aptly remarked that the comparison between the two collections offers a “confusingly many-sided picture,” which should forbid “any hasty formula for the mutual relationship of H and Ezekiel.”12 At that time, however, Fohrer and Zimmerli were unable to provide an alternative model that would satisfactorily account for specifics of the scribal mechanics involved in the mutual influence of H and Ezekiel: Fohrer envisioned a “common source” used by H and Ezekiel,13  See on this the essay in this volume by M. A. Lyons, “How Have We Changed? Older and Newer Arguments about the Relationship between Ezekiel and the Holiness Code,” pp. 1055–1074. I want to thank Dr. Lyons for making accessible to me a copy of his essay. In his recent study, From Law to Prophecy (see n. 2), 47–75, Lyons also provides a helpful discussion of the main criteria for evaluating the literary dependence between two texts. 11  On these and related issues, see now especially the discussion by D. M. C arr , “The Many Uses of Intertextuality in Biblical Studies: Actual and Potential,” in Congress Volume: Helsinki, 2010 (ed. M. Nissinen; VTSup 148; Leiden: Brill, 2010), 505–535. Many of the parallels between H and Ezekiel that have been identified by scholars, especially when they concern single words or idioms, may arguably qualify as instances of what Carr aptly terms “semi-canonical fallacy” in the study of intertextual relationships between two (or more) texts in the Hebrew Bible. 12  W. Z immerli , Ezechiel, Teilband 1: Ezechiel 1–24 (BKAT 13/1; Neukirchen-Vluyn: Neukirchener Verlag, 1969), 78*: “ein verwirrend vielgestaltiges Bild, das auf jeden Fall jede pauschale Formel für die gegenseitige Beziehung von H un Ez verbietet.” See more generally his discussion of this issue on pp. 70*–79*. See also G. Fohrer , Die Hauptprobleme des Buches Ezechiel (BZAW 72; Berlin: Töpelmann, 1952), 148–154. To be clear, Zimmerli does not dispute the fact that in specific instances the direction of influence is easier to discern; compare, e.g., his discussion of Lev 17 and Ezek 14 in W. Z immerli , “Die Eigenart der prophetischen Rede des Ezechiel,” ZAW 66 (1954), 1–26. Rather, the argument is that when H and Ezekiel are compared in their entirety, it is no longer possible to identify a single, linear relationship between these two collections. As we will see, Zimmerliʼs insight in this regard is actually confirmed by the comparison between the MT and Old Greek of Ezekiel. 13  Fohrer, Hauptprobleme (see n. 12), esp. 147. 10

1018

Christophe L. Nihan

whereas Zimmerli explained this phenomenon by the assumption that both collections were transmitted among similar (priestly) groups.14 As a result, the case for the mutual influence of H and Ezekiel did not establish itself, and most recent studies have returned instead to linear models based on the alternative between Hʼs precedence over Ezekiel or Ezekielʼs precedence over H.15 In what follows, I will briefly argue why I consider this alternative to be problematic and unsatisfactory and why more complex models for assessing the relationship between these two collections need to be considered. I will do so on the basis of two sorts of evidence: the textual evidence provided by the comparison between the Masoretic Text (MT) and the Old Greek (OG) of Ezekiel and the internal evidence provided by the comparison between different passages of the book for which parallels with the Holiness legislation have been claimed. Though both types of evidence are clearly distinct, they are nevertheless complementary, in that they point to a similar conclusion. Namely, the “coordination” between H and Ezekiel is the result of a complex scribal process, which accompanies the formation and the transmission of these collections.16 A brief conclusion will derive some of the main implications of this study for the future discussion on the relationship between H and Ezekiel.

14  See, e.g., Z immerli , Ezechiel 1 (see n. 12), 78*–79*: “Die Kreise, die H seine (vorpriesterliche) Gestalt gegeben haben, müssen nicht zu ferne vom Tradentenkreis des Buches Ez gesucht werden.” 15  This approach is clear, for instance, in the recent monograph by Lyons , From Law to Prophecy (see n. 2). One may note, in this context, that an alternative approach was offered by some earlier scholars who reconstructed a complex literary history for the Holiness legislation, in which some portions of this collection would predate Ezekiel whereas others would postdate it. See B. B aentsch , Das Heiligkeits-Gesetz, Lev. XVII–XXVI: Eine historisch-kritische Untersuchung (Erfurt: Hugo Güther, 1893); further, e.g., G. Bettenzoli, Geist der Heiligkeit: Traditionsgeschichtliche Untersuchung des QDŠ-Begriffes im Buch Ezechiel (Quaderni di semitistica 8; Florence: Istituto di linguistica e di lingue orientali Università di Firenze, 1979). Yet these models have similarly found little acceptance. Instead, recent studies of H have almost unanimously emphasized the compositional unity of this collection. See, e.g., Otto, “Innerbiblische Exegese” (see n. 9); Nihan, From Priestly Torah (see n. 3), 394–545; and compare also J. Stackert, Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007). The following discussion assumes that H is largely a unified work, with some possible exceptions (like, e.g., Lev 24:10–23). 16  The concept of scribal coordination, which I am using here, is borrowed from the recent study by T. P. Mackie, Expanding Ezekiel: The Hermeneutics of Scribal Addition in the Ancient Text Witnesses of the Book of Ezekiel (FRLANT 257; Göttingen: Vandenhoeck & Ruprecht, 2015), 59–67. I thank my colleague Bill Tooman (St. Andrews) for drawing my attention to this important study. Specifically, Mackie identifies four main types of coordinating expansions in Ezekiel: harmonization; assimilation of related texts; assimilation of phraseology; and “cross-reference” or allusion (65–67).

Ezekiel and the Holiness Legislation

1019

2 Textual Evidence for Scribal Coordination between H and Ezekiel A basic issue, when discussing the relationship between H and Ezekiel, has to do with the textual form(s) on which the comparison between these collections is made. Many studies of the relationship between H and Ezekiel are predominantly – sometimes even exclusively – based on the comparison between the MT of Lev 17–26 and Ezekiel.17 However, this approach is problematic. Especially in the case of Ezekiel, we have documented evidence that the book underwent a rather complex process of transmission and that the traditions associated with the eponymous prophet remained fluid until at least the second century BCE. The earliest witness to the Greek text of Ezekiel, P.967 (from the late second or early third century CE), preserves a text that is significantly shorter than MT and that is also characterized by a distinct arrangement for chapters 36–40, where the oracle against Gog (Ezek 38–39) is placed before chapter 37 and not after as in MT.18 Furthermore, in the case of Ezek 36–40, the main features of P.967, such as, especially, the absence of 36:23c–38 or the placement of the oracle against Gog before chapter 37, are confirmed by a manuscript of the 17  Compare, e.g., J. Milgrom, “Leviticus 26 and Ezekiel,” in The Quest for Context and Meaning: Studies in Biblical Intertextuality in Honor of James A. Sanders (ed. G. A. Evans and S. Talmon; BIS 28; Leiden: Brill, 1997), 57–62; similarly Levitt Kohn, New Heart (see n. 8), 30–85. In this regard, the study by Lyons, From Law to Prophecy (see n. 2), represents a step forward, since he does acknowledge that (a) MT does not necessarily represent the earliest form of the book and (b) this finding has some significant implications for the assessment of the relationship between H and Ezekiel, since the MT of Ezekiel is frequently closer to Lev 17–26 than the OG. Note, in particular, the following statement: “With respect to the text of Ezekiel, I have found that MT Ezekiel reflects a textual form that has been brought into an even closer connection with Lev 17–26 than the shorter LXX Ezekiel” (13). However, it is questionable whether Lyons himself derives all the implications of this finding. See further below. 18  P.967 represents the earliest known copy of a Septuagintal codex and contains Ezekiel, Daniel (with the additions of Susanna and Bel and the Dragon), and Esther. In the case of Ezekiel, only the beginning of the manuscript (1:1–11:24) is missing. The papyrus is preserved in different locations: Dublin (formerly housed in London), Princeton, Cologne, and Madrid. For the edition, see F. J. Kenyon, The Chester Beatty Biblical Papyri: Ezekiel, Daniel, Esther (2 vols.; London: Emery Walker, 1937); A. C. Johnson et al., The John H. Schiede Biblical Papyri: Ezekiel (Princeton University Studies in Papyrology 3; Princeton: Princeton University Press, 1938); L. G. J ahn , Der griechische Text des Buche Ezechiel nach dem Kölner Teil des Papyrus 967 (Papyrologische Texte und Abhandlungen 15; Bonn: Habelt, 1972); M. Fernández-Galiano, “Nuevas Paginas del codice 967 del AT griego,” SPap 10 (1971), 7–76. For a recent and comprehensive analysis of P.967, see especially I. L illy, Two Books of Ezekiel: Papyrus 967 and the Masoretic Text as Variant Literary Editions (VTSup 150; Leiden: Brill, 2012).

1020

Christophe L. Nihan

Old Latin from the fifth century CE, the Codex Wurceburgensis (LaW), which appears to be independent of P.967.19 Although there have been various attempts to explain the differences between P.967 and the MT of Ezekiel as reflecting textual accidents, these explanations have been shown to be unconvincing.20 Instead, the recent and detailed investigation of I. Lilly has confirmed the thesis of earlier scholars like J. Lust, according to which P.967 is a witness to an early Hebrew text distinct from the one preserved in MT, which formed the basis for the Old Greek (OG) of Ezekiel.21 Moreover, the other main Greek versions of Ezekiel, such as especially Codex Vaticanus (GB) and Codex Alexandrinus (GA), likewise preserve several variant readings vis-à-vis MT, as well as a slightly shorter text.22 The relationship between P.967, GB, and GA as witnesses to the early Greek text of Ezekiel admittedly remains a complex issue that has not yet been entirely clarified. What seems clear, however, is that the MT of Ezekiel generally represents an expansionist textual form with regard to the shorter Hebrew base-text (Vorlage) that provided the basis for the translation of the OG of Ezekiel, a conclusion also corroborated (with some qualifications) by the recent study of Lilly.23

 See on this P.-M. Bogaert, “Le témoignage de la Vetus Latina dans lʼétude de la tradition des Septante: Ézéchiel et Daniel dans le Papyrus 967,” Bib 59 (1978), 384–395. Cf. also J. Lust, “The Sequence of Ez 36–40 and the Omission of Ez 36,23c–38 in Pap. 967 and in Codex Wurceburgensis,” BIOSCS 14 (1981), 45–46. 20  For the history of research on this issue, see the discussion by L illy , Two Books (see n. 18), 28–62, esp. 34–62, and for a discussion of the overall quality of P.967 as a witness to an early Hebrew text of Ezekiel, see pp. 80–130. Lilly concludes that P.967 preserves a literary edition of Ezekiel that she describes as “more often than not faithful to a Hebrew Vorlage, lacking in textual errors” and that she says “rarely reflects Greek innovation”; she concludes that “p967ʼs literary edition is a viable, early text of Ezekiel” (130). Compare also, e.g., the similar conclusion reached by A. S. Crane, Israelʼs Restoration: A Textual-Comparative Exploration of Ezekiel 36–39 (VTSup 122; Leiden: Brill, 2008), 263, who likewise characterizes P.967 as “a viable witness to the Old Greek, and the Hebrew Urtext.” He further concludes that the Old Greek of Ezekiel is “most likely represented” in P.967 (266). 21  See, e.g., J. L ust , “Ezekiel 36–40 in the Oldest Greek Manuscript,” CBQ 43 (1981), 517–533; idem, “The Use of Textual Witnesses for the Establishment of the Text: The Shorter and Longer Texts of Ezekiel,” in Ezekiel and His Book: Textual and Literary Criticism and Their Interrelation (ed. J. Lust; BETL 74; Louvain: Presses Universitaires, 1986), 7–20; idem, “Major Divergences between LXX and MT in Ezekiel,” in The Earliest Text of the Hebrew Bible: The Relationship between the Masoretic Text and the Hebrew Base of the Septuagint Reconsidered (ed. A. Schenker; Septuagint and Cognate Studies 52; Atlanta: Society of Biblical Literature, 2003), 83–92; compare also Crane, Israelʼs Restoration (see n. 20). 22  See on this also E. Tov, The Greek and Hebrew Bible: Collected Essays on the Septuagint (VTSup 72; Leiden: Brill, 1999), 367–410. 23  For this view, see L ust , “Use of Textual Witnesses” (see n. 21); idem , “Major Divergences” (see n. 21). For a qualified and comprehensive restatement, see Lilly, Two Books (see n. 18), esp. 301–304. She concludes, in particular, that, “p967ʼs base text is, in many cases, 19

Ezekiel and the Holiness Legislation

1021

This conclusion, in turn, has a direct bearing on the discussion of the relationship between H and Ezekiel. As some authors have already observed, in several instances the expansions evidenced by the comparison between the MT and OG of Ezekiel appear to be characterized by a concern to coordinate more closely the prophecy of Ezekiel with the phraseology of Leviticus and, especially, of the Holiness legislation (Lev 17–26).24 A comprehensive discussion of this phenomenon is beyond the scope of this short essay, but a few examples may nevertheless be briefly mentioned here in order to illustrate this point. In some cases, the scribal expansions coordinating Ezekiel with H may be restricted to a single word or phrase. One example can be found in Ezek 34:27: Ezekiel 34:27b MT

‫וידעו כי אני יהוה בשברי את מטות עלם והצלתים מיד העבדים בהם‬ They will know that I am YHWH, when I break the bars of their yoke and rescue them from the power of those who enslave them. OG καὶ γνώσονται ὅτι ἐγώ κύριος ἐν τῷ συντρῖψαί με τὸν ζυγὸν αὐτῶν καὶ ἐξελοῦμαι αὐτοὺς ἐκ χειρὸς τῶν καταδουλωσαμένων αὐτούς25 They will know that I am the Lord when I break their yoke and deliver them from the hand of those who enslave them.

This statement is part of an oracle, in Ezek 34:25–30, that presents several significant parallels with Lev 26:3–13.26 The formulation of v. 27b MT is clearly reminiscent of Lev 26:13b: the reference to YHWH “breaking” (‫ )שבר‬the “bars of the yoke” (‫ )מטות עול‬only occurs in these two passages in the Hebrew Bible, and furthermore, in both contexts it is combined with the formula of divine self-assertion, ‫אני יהוה‬. The OG provides a fairly literal translation of the Hebrew but does not contain the word “bars.” It has sometimes been argued that this

closer to a shorter Hebrew parent text. This shorter text likely resembles the basis for many of MTʼs developments” (302). 24  On this phenomenon, see now the study by M ackie , Expanding Ezekiel (see n. 16), 181–205. Mackie concludes, in particular, that “Leviticus and Jeremiah provided the most common source” for scribal expansions that seek to coordinate Ezekiel with other scriptural books (214). Compare also the brief remark in this regard by Lyons, From Law to Prophecy (see n. 2), 13. In addition, J. Stromberg, “Observations on Inner-Scriptural Scribal Expansions in MT Ezekiel,” VT 58 (2008), 68–86, observes that scribal expansions reflected in the MT of Ezekiel were based on other scriptural compositions (“inner-scriptural scribal expansions”). 25  For the OG, see J. Z iegler , Septuaginta: Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum, Vol. 16/1; Ezechiel (2nd ed., with a complement by D. Fraenkel; Göttingen: Vandenhoeck & Ruprecht, 1977), 257, 343. 26  See the discussion below.

1022

Christophe L. Nihan

word was omitted in the Greek translation.27 Though this is possible, the reasons for such an omission are not very clear, and the shorter idiom (‫ )שבר עול‬is well attested elsewhere in the Hebrew Bible.28 As recently argued by T. P. Mackie, it seems more likely, therefore, that this word was missing from the Hebrew base text used by the Greek translator and represents an expansion introduced in the textual form preserved by the MT in order to highlight the proximity of Ezek 34:27 to Lev 26:13.29 In other instances, however, scribal expansions coordinating Ezekiel with H concern more significant segments of text. One clear example can be found in Ezek 6:5. This verse is part of an oracle against the “mountains of Israel” (Ezek 6:3b–6), which presents several significant parallels with Lev 26:30–33. In the case of Ezek 6:5, the formulation of the first half of this verse (5a) has a close parallel in Lev 26:30a, as the following comparison demonstrates: Ezekiel 6:5a MT

‫ונתתי את פגרי בני ישראל לפני גלוליהם‬ I will place the carcasses of the Israelites upon the carcasses of your idols. Leviticus 26:30a

‫ונתתי את פגריכם על פגרי גלוליכם‬ I will place your (pl.) carcasses upon the carcasses of your idols.

Yet this part of v. 5 is missing from the OG of Ezekiel.30 Here also, there is no obvious textual mechanism to explain this omission, and it is likely that MTʼs plus reflects a scribal expansion, the purpose of which was to bring the oracle of Ezek 6 in closer proximity to the language of Lev 26:30–33.31 Another example of a substantial expansion betraying the influence of Lev 17–26 may arguably be preserved in Ezek 33:25–26 MT, a divine oracle of judgment against the nonexiles who have remained in the land:32 Ezekiel 33:25–26 MT ‫לכן אמר אליהם כה אמר אדני יהוה על הדם תאכלו ועינכם תשאו אל גלוליכם ודם תשפכו והארץ‬ ‫תירשו עמדתם על חרבכם עשיתן תועבה ואיש את אשת רעהו טמאתם והארץ תירשו‬ 27  E.g., W. Z immerli , Ezechiel, 2. Teilband: Ezechiel 25–48 (BKAT 13/2; Neukir­chenVluyn: Neukirchener Verlag, 1969), 831. 28  Namely, in Jeremiah: see Jer 2:20; 5:5; 30:8. 29  Mackie, Expanding Ezekiel (see n. 16), 197. 30  For the Greek witnesses to Ezek 6:5, see Ziegler, Septuaginta (see n. 25), 110. 31  For this conclusion, see, e.g., Z immerli , Ezechiel 1 (see n. 12), 139–140; further S tromberg , “Observations” (see n. 24), 71–72; Lyons , From Law to Prophecy (see n. 2), 63, 139. As observed by M. Greenberg, Ezekiel 1–20 (AB 22; Garden City, NY: Doubleday, 1983), 132, v. 5a in the MT may be regarded as an explicative comment intended to alleviate the tension created in the shorter version preserved in the OG by the fact that the oracle is initially addressed to the “mountains of Israel” (v. 3) but later involves human addressees (see especially the reference to “your slain ones” in v. 4b). 32  Unless otherwise specified, all translations of the Hebrew and the Greek are my own.

Ezekiel and the Holiness Legislation

1023

Therefore say to them: “Thus has the Lord YHWH declared: You eat (flesh) with its blood,33 you raise your eyes34 toward your idols, you shed blood – and yet you would claim the land? You have relied on the sword, you have done35 abominable things, you have defiled one anotherʼs wives – and yet you would claim the land?”

As various authors have observed, the formulation of this passage takes up various accusations against the people of Judah and Jerusalem found in the first part of the book (Ezek 4–24), which are combined with references to prohibitions stated in Lev 17–26. The reference in v. 25 to eating (flesh) together with its blood (‫)על הדם תאכלו‬, in particular, corresponds to Lev 19:26, which is taken up here in reverse order (‫)לא תאכלו על הדם‬.36 Additionally, the reference in v. 26 to the nonexiles “defiling” themselves with one anotherʼs wives, which is found in Ezek 18:6, 11, 15 (further 22:11), also has a parallel in Lev 18:20.37 In the main witnesses to the OG of Ezekiel, only the introduction to the divine speech in v. 25aα is preserved; the entire accusation against the nonexiles in vv. 25aβ–26 (together with the beginning of v. 27) is missing. The shorter Greek text has sometimes been explained by the assumption that it would reflect a case of homoioteleuton, due to the repetition of the phrase ‫ כה אמר אדני יהוה‬at the beginning of vv. 25 and 27.38 However, as other scholars have observed, this explanation is not entirely satisfactory. The presence of a new introduction to the divine speech in v. 27aα (which seems somewhat redundant after v. 25aα), combined with the fact that the direct address to the nonexiles that characterizes the divine speech in vv. 25–26 stands in some tension with the situation of communication presupposed by the following verses (v. 27–29), may more likely suggest that the Greek translator of Ezekiel had before him a Hebrew text that did not yet contain the accusation found in vv. 25aβ–26.39 If this is correct,

33  For this translation, see D. I. B lock , The Book of Ezekiel: Chapters 25–48 (NICOT; Grand Rapids, MI: Eerdmans, 1998), 257, n. 91. Emending to ‫( על ההרים‬as proposed by BHS) is unnecessary; the formulation here corresponds to Lev 19:26; see below. 34  Hebrew ‫ ועינכם‬is a defective spelling for ‫ ;ועיניכם‬the correct form is preserved in some Hebrew MSS. 35  Reading ‫ עשיתם‬instead of ‫( עשיתן‬e.g., Zimmerli, Ezechiel 2 [see n. 27], 816). 36  Z immerli , Ezechiel 2 (see n. 27), 819–820; L. C. A llen , Ezekiel 20–48 (WBC 29; Dallas: Word Books, 1990), 153. 37  The remaining accusations have parallels in other parts of Ezekiel and are arguably derived from there. For “raising the eyes toward idols,” see Ezek 18:6, 12, 15; “shedding blood,” see Ezek 22:3, 4, 6, 9, 12, 27, further 18:10 and 24:7; “to do abominable things,” see, e.g., Ezek 18:12, 13, 24; 22:11. As these parallels indicate, besides H, Ezek 18 and 22 appear to constitute the main sources for the accusations listed in 33:25–26. 38  See, e.g., Zimmerli, Ezechiel 2 (see n. 27), 815; Allen, Ezekiel 20–48 (see n. 36), 150; Block, Book of Ezekiel (see n. 33), 257, n. 90. 39  S. P etry , Die Entgrenzung JHWHs: Monolatrie, Bilderverbot und Monotheismus im Deuteronomium, in Deuterojesaja und im Ezechielbuch (FAT 2/27; Tübingen: Mohr Siebeck, 2007), 331–332; further, e.g., A. K lein , Schriftauslegung im Ezechielbuch:

1024

Christophe L. Nihan

this passage may be viewed as another instance in which Ezekiel MT preserves an expansion betraying the influence of the phraseology of Lev 17–26. Although such examples of scribal expansions aligning Ezekiel with Leviticus, and especially with Lev 17–26, are often found in the MT, it is important to observe that they are not restricted to this textual form and can also occasionally be found in the OG of Ezekiel. One clear example, recently discussed by Mackie, can be seen in Ezek 18:8: Ezekiel 18:8a MT He does not lend with interest and he does not take profit.

‫בנשך לא יתן ותרבית לא יקח‬

OG καὶ τὸ ἀργύριον αὐτοῦ ἐπὶ τόκῳ οὐ δώσει καὶ πλεονασμὸν οὐ λήμψεται He does not lend his silver with interest and he does not take profit.

The formulation of this passage is similar to Lev 25:36–37, which prevents “taking profit” (‫לקח נשך ותרבית‬, v. 36a) from an impoverished Israelite or “lending” him “silver with interest” (‫נתן כסף בנשך‬, v. 37a).40 However, the reference to “silver” (‫ )כסף‬is missing from the shorter text preserved by MT in this case. The inclusion of silver (ἄργυρος) in the longer text preserved by the OG should therefore be regarded as an attempt by a later scribe to provide a more exact parallel with the formulation of Lev 25:36–37; as aptly noted by Mackie, in the OG of Ezek 18:8a “every Hebrew element” found in the law of Lev 25:36a + 37a “has an exact quantitative representation.”41 On the whole, the various examples discussed here, to which others could be added,42 corroborate the view that one of the features characterizing the expansion of the text of Ezekiel during its transmission in the Second Temple period was a concern to coordinate the prophecy of Ezekiel with the phraseology of Leviticus and, especially, of the Holiness legislation in Lev 17–26. This scribal process of textual coordination with H through expansion and amplification Redaktionsgeschichtliche Untersuchungen zu Ez 34–39 (BZAW 391; Berlin: de Gruyter, 2008), 355. 40  Zimmerli, Ezechiel 1 (see n. 12), 406: “Mit seiner in 22:12 wiederkehrenden Paralleli­ sierung von ‫( נשׁך‬eig. ‘Ab-Biß, Abzug’) und ‫( תרבית‬eig. ‘Vermehrung, Zuschlag’) steht Ez 18:8 in seiner sprachlichen Formulierung wieder besonders nahe bei H.” Further, e.g., Lyons, From Law to Prophecy (see n. 2), 178. 41  Mackie, Expanding Ezekiel (see n. 16), 193. 42  Compare, for instance, the case of MTʼs statement, ‫ופרו ורבו‬, “they will multiply and become numerous,” in Ezek 36:11a, which is missing from the OG. The verbs ‫ רבה‬and ‫פרה‬ are characteristic of the Priestly traditions in Genesis (where they always appear, however, in the reverse order), and they also occur once in the context of H (see Lev 26:9). For further discussion of this case, see especially Stromberg, “Observations” (see n. 24), 73–74, as well as Crane, Israelʼs Restoration (see n. 20), 52–53.

Ezekiel and the Holiness Legislation

1025

of the prophetic text is particularly apparent in the textual form preserved by Ezekiel MT – an observation that corresponds to the fact that MT generally represents an expansionist form of a shorter Hebrew text – but it can also occasionally be found in the OG of Ezekiel. Additionally, and although the evidence in this case is significantly more limited, there are some indications suggesting that a similar process was at work in the transmission of Lev 17–26, whose text was occasionally brought into closer connection with Ezekiel. One example would be Lev 19:26 LXX, which reads Μὴ ἔσθετε ἐπὶ τῶν ὀρέων, “Do not eat on/ over the mountains,” instead of MT ‫לא תאכלו על הדם‬. The expression found in Lev 19:26 LXX otherwise occurs exclusively in some passages of Ezekiel43 and arguably reflects the influence of this prophetic tradition over the Greek version of Leviticus. However, such examples appear to be much more restricted in number, which corresponds to the fact that the text of Leviticus became stable (although not canonical) significantly earlier than the text of Ezekiel in the Second Temple period.44 In and of itself, the conclusion reached here regarding increasing coordination between Ezekiel and H in the course of the transmission of Ezekiel (and, to a lesser extent, of Leviticus) should not surprise us. Ancient scribes copying and transmitting the text of Ezekiel would naturally note the connections between Ezekiel and H. The perception of these connections led them in turn to highlight the relationship between the two collections by introducing further references to the Holiness legislation in the text of Ezekiel, especially at places where they felt these connections were appropriate; these “coordinating expansions” are preserved in the various pluses evidenced by the comparison between the textual forms represented in Ezekiel MT and OG. Nevertheless, the scribal phenomenon identified here has some significant implications for the discussion of the relationship between H and Ezekiel that have not always been sufficiently considered. In particular, and contrary to the approach adopted in most scholarly treatments, the previous discussion indicates that the agreements between H and Ezekiel are not limited to a specific stage in the forming of these collections. Rather, these agreements are part of a long and complex formative process, which extended over several centuries and which somehow culminated in the creation of the textual form preserved in Ezekiel MT.45 At this stage of the discussion, it becomes necessary to ask whether, and to what extent, further  See Ezek 18:11, 15; 22:9.  On Leviticus being one of the most stable texts transmitted during the Second Temple period, see, e.g., S. Metso, “Evidence from the Qumran Scrolls for the Scribal Transmission of Leviticus,” in Editing the Bible: Assessing the Task Past and Present (ed. J. S. Kloppenborg and J. H. Newman; Resources for Biblical Study 69; Atlanta: Society of Biblical Literature, 2012), 67–79. 45  As argued by Tov , Greek and Hebrew Bible (see n. 22), esp. 101, Ezekiel MT itself must be viewed as one redactional stage in the forming of Ezekiel and not merely a copy of the book. For the need to treat variant literary editions of a given book as providing evidence 43

44

1026

Christophe L. Nihan

consideration of the relationship to H in the materials shared by Ezekiel MT and OG may provide us with an additional perspective on this issue.46

3 Reception of H in the Materials Shared by Ezekiel MT and OG Though several recent studies tend to consider the book of Ezekiel to be a largely homogeneous composition, this approach is contradicted by the evidence provided by the comparison between the OG and MT of Ezekiel. The existence of at least two distinct editions (MT and OG) is a clear indication that Ezekiel, like other prophetic books such as Isaiah and Jeremiah, is the product of a complex compositional history. Moreover, redaction criticism has long identified evidence in Ezekiel for various scribal techniques that are documented not only in biblical and non- or parabiblical writings of the Second Temple period but also in other ancient Near Eastern traditions. The attempt to identify earlier stages in the composition of Ezekiel on the basis of such criteria must, by necessity, remain a speculative enterprise, especially when there is no documented evidence for such stages. But this does not mean that this approach can simply be ignored or discarded, as it can often provide us with some significant insights into the internal dynamics of the book itself. My aim, in the context of this short essay, is not to survey the various models for the composition of the book that have been advanced or to present an alternative reconstruction but only to illustrate some of the implications of this point for the issue of the relationship between Ezekiel and the Holiness legislation. I will do so on the basis of one significant example, namely, Ezek 34:25–30 and 37:24–28, two passages that are preserved (with some variations) in all the extant versions of the book. Both passages share several phraseological and topical features and, in addition, they are also often considered to present significant parallels with Lev 26. The parallels with Lev 26:3–13 are particularly clear in the case of Ezek 34:25–30.47 The reference in Ezek 34:25a to the elimination of wild, destructive for the redactional history of that book, see also the comments by E. Ulrich, The Dead Sea Scrolls and the Origins of the Bible (Grand Rapids, MI: Eerdmans, 1999). 46  The distinction posited here does not correspond to the classical (and in my view problematic) distinction between the composition and the transmission of the text of Ezekiel but, rather, to the difference between expansions to the base text of Ezekiel that are documented by the comparison between ancient versions of this book, especially the MT and the OG, and expansions that are inferred from internal observations on the text shared by these versions and that correspond to scribal techniques identified by redaction criticism. The need for a closer articulation between these two approaches has been voiced in some recent publications; compare, e.g., L illy , Two Books (see n. 18), 309, and the remarks by M ackie , Expanding Ezekiel (see n. 16), 47–49. 47  As far as I can see, the delineation between 34:23–24, 25–30 and 31 is fairly standard in the commentaries and the main studies and requires no further justification here. The main

Ezekiel and the Holiness Legislation

1027

beasts from the land, ‫והשבתי חיה רעה מן הארץ‬, has a literal equivalent in Lev 26:6b. Ezekiel 34:26b, 27a, for its part, has a close parallel in Lev 26:4. In particular, both passages preserve a similar sequence, albeit with some differences, which involves the giving of “rain” (‫ )גשם‬or “your rains” (‫ )גשמיכם‬at their “appointed time(s)” (‫ בעתו‬or ‫)בעתם‬, followed by the promise that the land and trees will deliver their produce. In Ezek 34:27a, this description is concluded with the promise that the people will reside safely in the land, ‫והיו על אדמתם‬ ‫לבטח‬, which is also reminiscent of the statement found at the end of Lev 26:5b (‫)וישבתם לבטח בארצכם‬. Although the verbs used in Ezek 34:27 and Lev 26:5 are different, the phrase ‫ ישב לבטח‬also occurs later in Ezek 34:28b, where it is combined with the expression ‫ואין מחריד‬, “no one will threaten them,” which is also found in Lev 26:6a. Finally, the formulation of Ezek 34:27b, combining the formula of divine self-assertion (‫ )כי אני יהוה‬with a reference to YHWH breaking Israelʼs “yoke,” is likewise found at the end of Lev 26:3–13 (v. 13b).48 Though none of these parallels is necessarily significant in itself, their combined presence within comparably short textual segments, frequently in the same sequence, may be considered sufficient evidence for substantiating the assumption of a literary relationship between Lev 26 and Ezek 34.49 Determining which text borrowed from the other is, however, more difficult to decide, especially because close examination of Lev 26 and Ezek 34 demonstrates that both texts evince distinct expansions of the materials they share.50 As such, and contrary debate concerns whether vv. 23–24, 25–30 and 31 are subsections within a single section comprising vv. 23–31 (thus, e.g., A llen , Ezekiel 20–48 [see n. 36], 159; B lock , Book of Ezekiel [see n. 33], 296) or whether vv. 23–24 and 25–30 (31) represent two distinct sections in chapter 34 (e.g., F. S edlmeier , Das Buch Ezechiel: Kapitel 25–48 [NSKAT; Stuttgart: Verlag Katholisches Bibelwerk, 2013], 162–163). My own preference is to interpret vv. 23–31 as constituting a single section; yet this point is not decisive for the following analysis and can be left open here. 48  As argued above, the parallel between the two passages has been further highlighted in the MT of Ezek 34:27, which adds the word “bars” (‫ )מטות‬before “yoke,” as in Lev 26:13b. 49  This view appears to be accepted by most commentators on Lev 26 and Ezek 34, although the relationship between these texts remains disputed. For a comprehensive discussion of this issue, see in particular M ilgrom , “Leviticus 26” (see n. 17), 57–59; G rünwaldt , Heiligkeitsgesetz (see n. 9), 349–350; as well as Klein, Schriftauslegung (see n. 39), 184–190. 50  Though the text of Ezek 34 contains several motifs that are unparalleled in Lev 26 (compare, e.g., the reference in 34:29 to the establishment of Israel as a “plantation of peace,” which will no longer suffer from hunger or bear the disgrace of the nations), there are, simultaneously, some motifs in Lev 26 that have likewise no equivalent in Ezek 34. For instance, the sequence found in Ezek 34:26b–27a comprising the giving of rain, abundant harvests, and secure dwelling in the land is also found in Lev 26:4–5 but is supplemented there in v. 5a with a detailed description of these abundant harvests (‫)והשיג לכם דיש את בציר ובציר ישיג את זרע‬ that has no equivalent in Ezek 34. Likewise, the reference to the elimination of “evil beasts” (‫ )חיה רעה‬from the land, which occurs in both passages, is followed in Lev 26:6b with a further promise (“the sword will not go through your land”) that is unparalleled in the corresponding passage of Ezek 34:25a.

1028

Christophe L. Nihan

to the approach demonstrated in several recent treatments, the identification in either Lev 26 or Ezek 34 of such expansions is not a sufficient criterion, in and of itself, for assessing which of these two texts is chronologically prior.51 Nevertheless, some observations appear to support the general assumption that Ezek 34 derives from Lev 26 rather than the other way around.52 First, as noted by J. Milgrom and M. Lyons, although both Ezek 34:26 and Lev 26:4 contain similar references to the sending of rain by YHWH, Ezek 34 adds an explicative comment, ‫גשמי ברכה יהיו‬, (rendered as ὑετὸν εὐλογίας in G), “they will be rains of blessing.” The presence of this explicative comment is easier to explain if one assumes that the reference to the sending of rain did not originate in Ezek 34 but was taken from Lev 26, so that the author of Ezek 34 felt the need to explain the meaning of this statement.53 Second, both Ezek 34 and Lev 26 make a connection between YHWHʼs breaking of Israelʼs yoke and the deliverance from slavery, with the root ‫( עבד‬see Lev 26:13a; Ezek 34:27b). However, the comment ‫“ והצלתים מיד העבדים בהם‬and I will deliver them from the hand of those who enslave them,” is somewhat surprising. In the context of Ezek 34:25–30 it is unclear who are the ‫העבדים‬, a term that refers here to the “enslavers” of Israel. The broader context of the chapter suggests that it is a reference to the foreign nations from which the Israelites will be brought back by YHWH in the future (Ezek 34:10, 11–16). Yet the use of this motif in 34:27b is unique: both the reference to the Israelites as the slaves of the nations and, correspondingly, the comparison of the return from exile as a deliverance from slavery are never found otherwise in Ezekiel, contrary to other prophetic books.54 51  This observation explains why recent studies that predominantly use this criterion come to opposed results: compare, e.g., M ilgrom , “Leviticus 26” (see n. 17), 57–59, and Grünwaldt, Heiligkeitsgesetz (see n. 9), 349–350. 52  Here, I must correct my earlier opinion, stated in C. N ihan , “Heiligkeitsgesetz und Pentateuch: Traditions- und kompositionsgeschichtliche Aspekte von Levitikus 26,” in Abschied von der Priesterschrift? Zum Stand der Pentateuchdebatte (ed. F. Hartenstein and K. Schmid; Veröffentlichungen der Wissenschaftlichen Gesellschaft für Theologie 40; Leipzig: Evangelische Verlagsanstalt, 2015), 161–185, where I was still arguing for the possible priority of Ezek 34:25–30. 53  See Milgrom, “Leviticus 26” (see n. 17), 58; Lyons, From Law to Prophecy (see n. 2), 125. The fact that this explicative comment in Ezek 34 uses the plural form, ‫גשמי‬, instead of the singular ‫ גשם‬as in the previous clause, could also be a reference to Lev 26:4, which uses the plural ‫גשמי‬. The Greek text, which uses the singular (ὑετὸν) in both instances, presumably harmonizes the stylistic tension caused by the alternation between singular and plural in the MT of Ezek 34:26. 54  In Ezekiel, the singular form of the substantive ‫ עבד‬is used to describe specific Israelite figures as the “servant of YHWH,” such as Jacob (28:25; 37:25) and David (34:23, 24; 37:24, 25). The plural is also used once to describe the prophets (“my servants the prophets,” Ezek 38:17) but never the Israelites as a whole. The closest analogy to this notion would be in Ezek 20:40, which refers to the Israelites serving YHWH on his mountain after being brought back into the land. But the description of the Israelites as the “slaves” of the nations

Ezekiel and the Holiness Legislation

1029

It is apparent that this concept is much more at home in Lev 26:13, where the reference is explicitly to YHWH delivering Israel from Egypt (cf. ‫אשר הוצאתי‬ ‫אתכם מארץ מצרים‬, v. 13a). Here also, therefore, it seems more logical to infer that the reference to Israelʼs deliverance from slavery in Ezek 34 is not original but has been adapted from Lev 26.55 Additionally, the notion that Ezek 34 is more likely to borrow from Lev 26 rather than the other way around is consistent with the general observation that, even though both texts present significant pluses as noted above, the materials shared by the two texts are often used in a more complex and elaborated way in Ezek 34 than in Lev 26. One clear example, in particular, can be found in the reference to the peopleʼs secure dwelling in the land, which is phrased in a significantly more sophisticated way in Ezek 34 than in Lev 26. Whereas Lev 26:5 merely refers once to the people “dwelling” (‫ )ישב‬securely (‫ )לבטח‬on their land (‫)ארץ‬, the same motif occurs no less than three times in Ezek 34 MT. The first occurrence, in 34:25, relates the motif of the Israelitesʼ secure dwelling to the elimination of “wild beasts.” The second occurrence, in v. 27, connects this motif with deliverance from foreign oppression. Finally, the third and last occurrence, in v. 28, combines the two related themes of deliverance from foreign oppression and from wild beasts (see v. 28a) and adds the concluding explicative comment ‫ואין מחריד‬, “and no one will threaten them.” As such, this example represents a fitting illustration of the way in which a fairly simple motif in Lev 26 has been taken over and significantly developed in Ezek 34.56 When we turn to Ezek 37:24–28, however, the situation is somewhat distinct and the relationship with Lev 26:3–13 is significantly less clear.57 The main is absent outside of Ezek 34:27b. The identification of the exile with the enslavement of the Israelites – and, correspondingly, of the return from exile with deliverance from slavery – is especially attested in Jeremiah: see, e.g., Jer 5:19; 17:4; 30:8, and compare also Isa 14:2–3. 55  In addition, the formulation of Ezek 34:27b has a further parallel in Jer 30:8 MT. It is possible that both passages represent adaptations of Lev 26:13 (I thank Dalit Rom-Shiloni for this suggestion). Note, however, that the Greek version of Jeremiah preserves a shorter, and presumably earlier, text at Jer 37:8. It is likely, therefore, that Jer 30:8 MT represents a case of secondary alignment with Lev 26, comparable to the instances surveyed in the previous section of this essay. 56  One should note, however, that the first occurrence of ‫ לבטח‬is missing from the OG of Ezek 34:25. Commentators usually assume an omission by the Greek translator (e.g., Zimmerli, Ezechiel 2 [see n. 27], 831; Allen, Ezekiel 20–48 [see n. 36], 158). Alternatively, it could also be argued that ‫ לבטח‬was introduced later in the textual form preserved by the MT. In this case, MTʼs plus in v. 25 would represent a further example of a scribal expansion serving to coordinate the prophecy of Ezekiel with H (Lev 26:5). Even so, however, in light of the use of the motif of Israelʼs secure dwelling in the land in vv. 27 and 28, the point would remain that this motif has been further developed in Ezek 34 vis-à-vis Lev 26. 57  The division of the concluding section of Ezek 37 is not entirely clear and remains a disputed issue among scholars. Some authors regard v. 25 as the beginning of a new unit; compare, e.g., K lein , Schriftauslegung (see n. 39), 75; K.-F. P ohlmann , Das Buch des

1030

Christophe L. Nihan

parallel arguably concerns the reference to the conclusion of a covenant between YHWH and Israel in Ezek 37:26–27 and Lev 26:9–12, which involves similar motifs (albeit in a somewhat different sequence):58 Shared motifs

Ezek 37

Lev 26

conclusion of a covenant

26a

9b

multiplication of the people

26bα

9a

establishment of YHWHʼs sanctuary

26bβ, 27aα

11a

covenantal formula

27aβ, b

12b

In this case, however, it is unlikely that such structural parallelism points to the dependence of Ezek 37 on Lev 26, contrary to what various authors have argued.59 To begin with, the lexical parallels between Ezek 37 and Lev 26 are significantly more limited than in the case of Ezek 34:25–30. The closest parallel is the covenantal formula, occurring in Ezek 37:27aβ, b and Lev 26:12b, respectively, “I will be their/your god and they/you will be my people.” Yet a similar formula occurs several times in Ezekiel and in Jeremiah, so there is no reason to assume that it should derive from Lev 26 specifically.60 The connection between Propheten Hesekiel (Ezechiel): Kapitel 20–48 (ATD; Göttingen: Vandenhoeck & Ruprecht, 2001), 503–505. However, the division between vv. 24 and 25 looks artificial. There is no indication of the beginning of a new unit in v. 25; instead, the reference to the people as the subject of the action described in v. 25a continues the formulation of v. 24b, and the two half-verses appear to be complementary (namely, the people will obey the divine laws and consequently settle permanently in the land of their ancestors). For this reason, a number of scholars propose to take v. 24b with vv. 25–28; compare, e.g., Zimmerli, Ezechiel 2 (see n. 27), 907, 913; Block, Book of Ezekiel (see n. 33), 408–409; Sedlmeier, Das Buch Ezechiel (see n. 47), 231–233. This solution is admittedly possible but does not account for the presence of the covenant formula at the end of v. 23 (“They will be my people and I will be their god”), which is best viewed as signaling the end of this unit. On balance, the division between vv. 24a and 24b seems questionable, and it is arguably better to regard all of vv. 24–28 as forming a distinct subsection within Ezek 37. 58  Namely, the reference to the peopleʼs multiplication is placed somewhat differently in the two sequences, since it occurs before the mention of the covenant in Lev 26:9 and after in Ezek 37:26. 59  See, e.g., Milgrom, “Leviticus 26” (see n. 17), 60; Klein, Schriftauslegung (see n. 39), 184–190. The opposite view has already been argued by several authors; compare, e.g., C. Levin, Die Verheißung des neuen Bundes in ihrem theologieschichtlichen Zusammenhang ausgelegt (FRLANT 137; Göttingen: Vandenhoeck & Ruprecht, 1985), 222–228; Grünwaldt, Heiligkeitsgesetz (see n. 9), 348–354. From a methodological perspective, the following discussion assumes that structural or topical parallels are important when assessing possible influence between two passages, as well as the direction of that influence, but such features are not sufficient per se. In particular, they need to be compared with the presence or absence of shared language, which remains a basic criterion for postulating textual borrowing between two passages. 60  With the same word order, see Jer 7:23; 31:33; see further Jer 31:1, where the formula

Ezekiel and the Holiness Legislation

1031

YHWHʼs covenant and Israelʼs multiplication, which is found in Ezek 37:26 and Lev 26:9, is unusual in Ezekiel. In Lev 26:9 the language used (with ‫ פרה‬hiphil and ‫ רבה‬hiphil) clearly refers to the Priestly covenants with Noah (Gen 9) and with Abraham (Gen 17).61 Although the description of Israelʼs multiplication in Ezek 37:26 is somewhat distinct, it is possible that it likewise involves a reference to the Priestly covenant. However, the phrase ‫ ונתתים והרביתי אותם‬in v. 26aβ is missing from the OG of Ezek 3762 and may therefore represent yet another instance of the tendency to coordinate the text of Ezekiel more closely with Leviticus and other “Priestly” texts in the Pentateuch.63 The twofold characterization in Ezek 37:26 of the covenant as a “covenant of peace” (‫)ברית שלום‬ and an “everlasting covenant” (‫ )ברית עולם‬has nothing to do with Lev 26 or with H in general. The concept of a “covenant of peace” established between YHWH and Israel is not attested in P or H.64 The designation of this covenant as an “everlasting covenant” might involve a reference to Pʼs covenant with Abrahamʼs offspring (Gen 17:7, 13, 19) but certainly does not refer to the covenant of Lev 26, which is never described as a ‫ברית עולם‬.65 In fact, it should be clear that the joint designation of the covenant as ‫ ברית שלום‬and ‫ ברית עולם‬in Ezek 37:26

has been expanded. In other passages the same covenantal formula occurs in the reverse order: see Jer 11:4; 24:7; 30:22; 32:28; Ezek 11:20; 14:11; 36:28 (MT); 37:23. 61  See my discussion in Nihan, From Priestly Torah (see n. 3), 537–538. 62  This phrase is missing from all the main witnesses to the OG of Ezek 37, including the Alexandrinus (which tends to follow MTʼs pluses). In the Hexapla, the phrase και δωσω αυτους και πλεθυνω αυτους, corresponding to MΤ ‫ונתתים והרביתי אותם‬, is placed under the asterisk, which provides further evidence that it was not part of the Greek translation of Ezek 37 originally. See Ziegler, Septuaginta (see n. 25), 271. 63  For this conclusion, see the recent discussion by Crane, Israelʼs Restoration (see n. 20), 133–134. 64  The only mention of a ‫ ברית שלום‬in the Priestly literature is in Num 25:12, where it refers to YHWHʼs covenant with Phinehas, not with the Israelites. There are some grounds for thinking that the conception laid out in Num 25 is based on Ezek 37, as argued by several scholars; see the discussion in C. Nihan, “The Priestly Covenant, Its Reinterpretations, and the Composition of ‘P,’ ” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions (ed. S. Shectman and J. S. Baden; ATANT 95; Zürich: Theologischer Verlag, 2009), 86–134, here 125–126, with further references. 65  Contra, e.g., Block, Book of Ezekiel (see n. 33), 419, who argues that the notion of a ‫ברית‬ ‫ עולם‬in Ezek 37:26 “stems from Lev 26:4.” Since Lev 26:4 contains neither the term ‫ ברית‬nor the term ‫עולם‬, this suggestion seems particularly unlikely. The fact that Lev 26, contrary to P, never uses the expression ‫ ברית עולם‬is not a coincidence but corresponds to a basic purpose of this text, which is precisely to revise and reformulate Pʼs concept of an “everlasting covenant” with Israel. See further my comments in Nihan, “Priestly Covenant” (see n. 64), 104–115; and compare also the helpful and nuanced discussion of this issue by J. S tackert, “Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction: Leviticus 26 as a Test Case,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 369–386.

1032

Christophe L. Nihan

represents the logical sequel to the utopian description in vv. 24–25 and does not require a reference to P or H in order to be understood.66 Finally, the description in Ezek 37:26bβ, 27aα of YHWHʼs sanctuary is also unlikely to be based on Lev 26. However, the formulation of this passage has often been misunderstood and therefore deserves brief discussion. In general, scholars who identify in Ezek 37:26bβ, 27aα a reference to P, or P and H, base their argument on two observations. First, the wording of v. 26bβ, with its reference to the sanctuary (‫ )מקדש‬being established “in the midst” (‫ )בתוך‬of Israel, would be reminiscent of the conception stated in Exod 25:8 and related passages of P. Second, the use in Ezek 37:26bβ and 27aα of the synonyms ‫ מקדש‬and ‫משכן‬ to refer to the sanctuary would be distinctive of P (and H).67 The first argument is possible although not necessary: the wording of Ezek 37:26bβ has no exact parallel in P and, more important, this statement provides an apt summary for the general concept developed in chapters 40–48, which describe YHWHʼs sanctuary precisely as being located “in the midst” (again with ‫ )בתוך‬of Israelʼs land (Ezek 48:8–22).68 The second argument is more problematic, because it ignores the specifics of the formulation of v. 27aα, which asserts that YHWHʼs ‫ משכן‬will be “over” the Israelites and not “amidst” them: ‫והיה משכני עליהם‬.69 The notion that YHWHʼs ‫ משכן‬will reside “over” the Israelites is unparalleled in P. The difference between vv. 26bβ and 27aα suggests that v. 27aα does not merely repeat v. 26bβ but provides additional information with regard to the mode of divine presence associated with Israelʼs sanctuary.70 Accordingly, ‫משכן‬  The term ‫ עולם‬occurs twice in the previous description of v. 25. In addition, the maintenance of ‫ שלום‬represents a traditional prerogative of the king in the ancient Near East as well as in the Hebrew Bible (see especially 1 Kgs 5:4–5; Isa 9:5–6; Mic 5:4–5; Zech 9:9–10; Ps 72:7), so that the reference to the “covenant of peace” in v. 26a also follows quite well after the reference to the reestablishment of a Davidic ruler that frames vv. 24–25 (vv. 24a, 25b). This connection is further highlighted in the Old Greek of Ezek 37, which adds a reference to David in v. 26a: “I will conclude a covenant of peace with David.” 67  See, e.g., Levin, Verheißung (see n. 59), 217; similarly T. A. Rudnig, Heilig und Profan: Redaktionskritische Studien zu Ez 40–48 (BZAW 287; Berlin: de Gruyter, 2000), 69. But contrast the comments on this point by Klein, Schriftauslegung (see n. 39), 175. 68  Cf. Ezek 48:8, further 48:10, 21. See on this, e.g., Zimmerli, Ezechiel 2 (see n. 27), 914, who aptly comments with regard to the formulation of Ezek 37:26b: “das ‫ בתוכם‬wird in der Landverteilung 48 1–29 geradezu geographisch anschaulich gemacht.” Compare similarly the comments by Pohlmann, Hesekiel (see n. 57), 505. 69  Some commentators have proposed rendering ‫ עליהם‬as “with them,” or even “among them,” thereby harmonizing the formulation of v. 27a with that of v. 26b; compare, e.g., Zimmerli, Ezechiel 2 (see n. 27), 915 (“bei ihnen”). Though this rendering is grammatically possible, it does not do justice to the parallel between v. 27a and v. 24a, where ‫ עליהם‬clearly means “over them.” 70  For this reason, it is unnecessary to regard v. 27aα as a doublet of v. 26bβ (pace Levin, Verheißung [see n. 59], 217). For this interpretation of ‫ משכן‬in Ezek 37:27, see the observations by A. H urvitz, A Linguistic Study of the Relationship between the Priestly Source and the 66

Ezekiel and the Holiness Legislation

1033

in Ezek 37:27 should be rendered as “dwelling-presence,” and the statement in v. 27aα should be viewed as completing (rather than repeating) the previous statement in v. 26b: YHWH will establish his sanctuary “in the midst” of Israel (Ezek 37:26b), so that his “dwelling-presence” may reside “over them” (Ezek 37:27aα). The vertical dimension introduced by v. 27a is, again, fitting as a summary for the general concept developed in Ezek 40–48, since the sanctuary in which YHWH will dwell (‫ )שכן‬is established not only in the center of the land (Ezek 48:8–22) but also on its “topmost mountain” (Ezek 40:2). Significantly enough, a similar usage of ‫ משכן‬appears to be presupposed by Lev 26:11. In P, the term ‫ משכן‬consistently refers to the totality of the tent structure constituting the wilderness sanctuary.71 In Lev 26, however, this meaning hardly fits the context. In 26:11, YHWH promises that, should the Israelites observe his laws, he will establish (‫ )נתן‬his ‫“ משכן‬in their midst,” yet according to P the ‫ משכן‬was already built by Moses (Exod 40:18–33) and consecrated by YHWHʼs ‫( כבוד‬Exod 40:34–35).72 The problem created by the use of ‫ משכן‬in Lev 26 was already manifest for the Greek translator of Leviticus, since the OG of Leviticus (represented here by GA and GB) reads at this place τὴν διαθήκην μου (= ‫)בריתי‬, “my covenant,” presumably as an attempt to alleviate the apparent tension between the statement found in Lev 26:11 and the rest of Pʼs account. The tension disappears, however, when we understand that ‫ משכן‬in Lev 26 refers not merely to the physical sanctuary but, rather, to the mode of divine presence associated with the latter.73 This conclusion, in turn, implies that ‫ משכן‬in Lev 26:11 has the same meaning as in Ezek 37:27 and should also be interpreted accordingly as denoting YHWHʼs “dwelling-presence.” It is unclear whether this distinct usage of ‫משכן‬, which characterizes both Lev 26 and Ezek 37, points to a textual dependence between these two passages. In principle, it is also possible that such usage was developed separately in H and Ezekiel on the basis of P. Nevertheless, the parallel noted above between the structure of Lev 26:9–12 and Ezek 37:26–27, combined with the observation that in both cases the reference to the ‫ משכן‬is followed by the same covenantal Book of Ezekiel: A New Approach to an Old Problem (CahRB 20; Paris: Gabalda, 1982), 109–113, and further J. M ilgrom , Leviticus 23–27: A New Translation with Introduction and Commentary (AB 3B; New York: Doubleday, 2001), 2300, both of whom establish a parallel with the rabbinic notion of šĕkînâ. However, Milgrom is incorrect in my opinion when he defines the ‫ משכן‬as a “spatially unbounded presence.” Rather, the formulation of Ezek 37:26b–27a implies that the ‫משכן‬, as a manifestation of the divine presence, is closely related to the sanctuary (‫ )מקדש‬mentioned immediately before. This conception is also consistent with the more fully developed description of Ezek 40–48. See further below. 71  As can be seen, e.g., in the description provided by Exod 35:11: “the ‫משכן‬, its tent and its covering, its clasps and its frames, its bars, its pillars, and its bases.” 72  For this observation, see my comments in N ihan , From Priestly Torah (see n. 3), 539–540; further, e.g., Müller, “Prophetic View,” (see n. 9), 211–212. 73  For this conclusion, see Milgrom, Leviticus 23–27 (see n. 70), 2229–2300.

1034

Christophe L. Nihan

formula, may arguably favor the view that one text has borrowed from the other. If this is the case, then, the previous discussion suggests that it is more likely that Lev 26 borrowed from Ezek 37 rather than the other way round. For the reasons noted above, the use of the term ‫ משכן‬to denote YHWHʼs “dwelling-presence” is consistent with the context of Ezek 37:24–28, as an introduction to Ezek 40–48. By contrast, this same usage of ‫ משכן‬is introduced more abruptly in Lev 26, as the evidence provided by the OG indicates: it stands in tension not only with the broader context of the Priestly narrative, as noted above, but also with the previous usage of this term in H itself (Lev 17:4), where ‫ משכן‬obviously refers to the physical sanctuary and not to the divine presence. Additionally, the secondary character of Lev 26 vis-à-vis Ezek 37 is corroborated by a further observation. Although in both texts the reference to the ‫ משכן‬is followed by the covenantal formula, in Lev 26 two additional statements have been introduced in between: ‫ולא תגעל נפשי אתכם‬, “my throat will not loathe you” (Lev 26:11b), and ‫והתהלכתי‬ ‫בתוככם‬, “I will walk in your midst” (Lev 26:12a). There is no good reason why the author of Ezek 37:26–27 would have omitted these statements had he used Lev 26 as his source, and it is easier therefore to regard Lev 26:11–12 as an expansion and an amplification of Ezek 37:26–27.74 In short, the previous discussion indicates that the relationship to Lev 26 is distinct in the case of Ezek 34 and 37. In Ezek 34:25–30, there are good reasons to assume that the author of this passage already knew Lev 26 and reused the phraseology of the first part of this chapter, vv. 3–13. In Ezek 37:24–28, it is possible to find occasional references to P (although even this point is open to debate),75 but there is no evidence supporting the view that this passage would be dependent on Lev 26. If a literary relationship between Ezek 37:24–28 and Lev 26:3–13 is to be assumed – which is possible, although not certain – it is more likely to go the other way around. This finding, in turn, may be accounted for when we consider the literary relationship between Ezek 34:25–30 and 74  It has sometimes been suggested that the reference to YHWH walking “in the midst” of Israel would have been omitted in Ezek 37 because of its anthropomorphic connotations (e.g., M ilgrom , Leviticus 23–27 [see n. 70], 2351), but this explanation is not very satisfactory since other passages in Ezekiel do not refrain from using anthropomorphic features to refer to the deity; compare, e.g., the reference to “the soles of my [i.e., YHWHʼs] feet” in Ezek 43:7a. The omission of the first statement in Ezek 37 is even more difficult to account for, since the reference to YHWHʼs “loathing” the Israelites is found earlier in Ezekiel (see Ezek 16:5), and scholars who assume the priority of Lev 26 over Ezek 37 have usually been at pains to find a convincing explanation for this omission (cf. Milgrom, “Leviticus 26” [see n. 17], 60; similarly idem, Leviticus 23–27, 2351). 75  The main parallels with P concern the designation of the covenant as a ‫ ברית עולם‬in v. 26a, as well as the reference to the sanctuary (‫ )מקדש‬being “in the midst” (‫ )בתוך‬of the Israelites in v. 26b. However, as remarked above, both expressions can be satisfactorily explained against the background of Ezekiel and do not require the assumption that the author of Ezek 37 would be referring to P here.

Ezekiel and the Holiness Legislation

1035

37:24–28. There are good reasons, in effect, to assume that these two texts belong to distinct compositional layers within the book. This is particularly manifest when we consider the references to David in both passages, which I will briefly discuss now to illustrate this point.76 In Ezek 34, the reference to the conclusion of a “covenant of peace” (‫ברית‬ ‫ )שלום‬in vv. 25–30 is preceded by the announcement of the reestablishment of a Davidic leader in vv. 23–24: Ezekiel 34:23–24 ‫והקמתי עליהם רעה אחד ורעה אתהן את עבדי דויד הוא ירעה אתם והוא יהיה להן לרעה‬ ‫ואני יהוה אהיה להם לאלהים ועבדי דוד נשיא בתוכם אני יהוה דברתי‬ I will establish one shepherd77 over them, who will tend them, my servant David: (it is he who will tend them)78 and it is he who will be their shepherd. Whereas I, YHWH, will be their god, (my servant)79 David will be a prince among them – I, YHWH, have spoken.

This description has a close counterpart in Ezek 37, where the reference to the “covenant of peace” in v. 26 is similarly preceded by the mention of the reestablishment of “David” in v. 24a: Ezekiel 37:24

‫ועבדי דוד מלך עליהם ורועה אחד יהיה לכלם‬ My servant David will be king over them,80 and there will be one shepherd for them all.  In a recent and insightful essay, W. Tooman, “Covenant and Presence in the Composition and Theology of Ezekiel,” in Divine Presence and Absence in Persian Period Judaism (ed. N. MacDonald and I. J. de Hulster; FAT 2/61; Studies of the Sofja Kovalevskaja Research Group on Early Jewish Monotheism 2; Tübingen: Mohr Siebeck, 2013), 151–182, rightly observes that the language of Ezek 37:24–28 may be satisfactorily accounted for in light of other passages in the book and that the assumption of the dependence of this passage on Lev 26 is unnecessary. However, he proposes to explain this by the notion that Ezek 37:24–28 already knew 34:25–30 and therefore did not need to repeat the references to Lev 26. Yet this solution does not account for the evidence suggesting that the two passages do not belong to the same compositional layer in Ezekiel and that Ezek 34:25–30 is arguably later (and not earlier) than 37:24–28. See below. 77  Several Greek witnesses, including P.967, LaW, and GA, read ἕτερον = ‫ אחר‬instead of MT ‫ אחד‬here (but see GB and Sinaiticus); cf. Ziegler, Septuaginta (see n. 25), 256. Whether this reading of G is deliberate or accidental is difficult to decide, but it seems likely in any case that the MT preserves the older reading, as is generally acknowledged (see also Klein, Schriftauslegung [see n. 39], 29, n. 23). 78  This statement is missing from the main witnesses to the Greek text and is placed under the asterisk in the Hexapla (cf. Ziegler, Septuaginta [see n. 25], 256–257). It was therefore most certainly absent from the Greek tradition initially and may well represent a later gloss in the MT. 79  The phrase ‫עבדי‬, “my servant,” in MT is missing from the Greek witnesses. It is difficult to decide whether this phrase is a gloss in MT or whether its absence in G is due to the fact that it was omitted in the course of the transmission. 80  The main witnesses to the Old Greek (P.967, GA, and GB) read ἄρχων ἐν μέσῳ αὐτῶν, “a 76

1036

Christophe L. Nihan

This announcement is continued in v. 25b, where, however, David is now described as “prince” (‫ )נשיא‬rather than “king” (‫)מלך‬: ‫ודוד עבדי נשיא להם לעולם‬, “and David, my servant, will be their prince forever.” The two statements, in vv. 24a and 25b, frame a description of the future of the Israelites in the land of their ancestors (vv. 24b–25a). The parallels between the descriptions of the Davidic leader in Ezek 34 and 37 are obvious: both passages refer to David as “my servant” (‫)עבדי‬, “the one shepherd” (‫)רעה אחד‬, and a “prince” (‫ )נשיא‬in Israel; additionally, in both cases these references to David precede the mention of the ‫ברית שלום‬, the “covenant of peace,” that will characterize the relationship between YHWH and Israel in the future. Yet there are also some significant differences. In particular, Ezek 37 (in the MT) characterizes David both as ‫( מלך‬in v. 24a) and as ‫( נשיא‬in v. 25b), whereas Ezek 34 exclusively designates him as ‫ נשיא‬and never uses ‫מלך‬. Moreover, Ezek 37:24a refers to David as “king over them” (‫)מלך עליהם‬, whereas 34:24 mentions instead that David will be “a prince in their midst” (‫נשיא‬ ‫)בתוכם‬. Taken together, these observations support the assumption that Ezek 37 presents a more elevated – and more vertical – view of Davidic leadership than the one stated in Ezek 34.81 Moreover, the tension between these two views of Davidic rule seems to have been apparent to the ancient audience(s) of the book, since the main witnesses to the OG (P.967, GA, and GB) offer a revised reading of Ezek 37:24a: ἄρχων ἐν μέσῳ αὐτῶν, “a prince in their midst,” which corresponds, in effect, to the conception previously stated in Ezek 34:24.82 In this way, the tension between the descriptions of Davidic leadership that are preserved in Ezek 34 and 37 are significantly attenuated in the Greek tradition, and the conception found in Ezek 37 is brought more closely in line with the less exalted view of Davidic rule already established in Ezek 34.83 prince in their midst,” here. See below. 81  As noted, e.g., by Sedlmeier, Das Buch Ezechiel (see n. 47), 171–172. 82  Cf. Ezek 34:24a: ‫ = ועבדי דוד נשיא בתוכם‬καὶ Δαυιδ ἐν μέσω αὐτῶν ἄρχων in the Greek. 83  Because the term ἄρχων normally renders the Hebrew ‫ נשיא‬in the Greek text of Ezek 1–39, it is often assumed that the Greek translation was based on a Hebrew Vorlage that read ‫ נשיא‬instead of MTʼs ‫מלך‬. Furthermore, a number of authors have also assumed that the reading ‫ נשיא‬would be original here, whereas ‫ מלך‬would reflect a change introduced by a later edition. For a detailed justification of this model, see J. Lust, “Messianism in LXX-Ezekiel: Toward a Synthesis,” in The Septuagint and Messianism (ed. M. A. Knibb; BETL 195; Leuven: Leuven University Press, 2006), 417–430; Crane, Israelʼs Restoration (see n. 20), 119–126, 250–263. However, as I have argued in detail elsewhere, this reconstruction is not necessarily compelling and there are reasons to retain MTʼs ‫ מלך‬as original in 37:24; see C. N ihan , “The nāśîʼ and the Future of Royalty in Ezekiel,” in History, Memory, Hebrew Scriptures: A Festschrift for Ehud Ben Zvi (ed. I. D. Wilson and D. Edelman; Winona Lake, IN: Eisenbrauns, 2015), 229–246, here 232–234. First, it is not clear that ἄρχων in the Greek text of Ezekiel must necessarily render the Hebrew ‫נשיא‬, as already pointed by some critics: see, e.g., A. van der K ooij , “The Septuagint of Ezekiel and Hasmonaean Leadership,” in Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust (BETL 192;

Ezekiel and the Holiness Legislation

1037

The tensions noted here between the descriptions of Davidic power in Ezek 34 and 37, as well as the evidence provided by the Greek tradition for attempts to harmonize these descriptions, suggest that the two passages are unlikely to come from the same hand. Although the chronological relationship of Ezek 34:23–24 to 37:24–25 is disputed, the cumulative evidence suggests that Ezek 34 is a rewriting of Ezek 37.84 First, the fact that Ezek 34 only refers to David as ‫נשיא‬ and avoids the term ‫ מלך‬suggests a concern to harmonize the titles ascribed to David, a phenomenon that, as argued above, seems likewise reflected in the Greek text of Ezek 37:24–25.85 Second, the emphasis on David as the “one shepherd” (‫ )רעה אחד‬makes better sense against the background of Ezek 37, since the theme of Israelʼs unity is the leitmotif of the second half of this chapter (see 37:15–19) and follows well after the reference to “one nation” (‫)גוי אחד‬ and “one king” (‫ )מלך אחד‬in 37:22. This observation appears to support the view that the presentation of David as the “one shepherd” originated in Ezek 37 rather than in Ezek 34.86 Third, the material shared by Ezek 34 and 37 tends to be preserved in a more expansionist version in the case of Ezek 34. This is clear, for instance, in the case of Davidʼs description as the “one shepherd.” In 34:23a, the title ‫ רעה אחד‬is followed by an explicative comment, absent from Ezek 37: ‫ורעה אתהן‬, “and he will shepherd them”; and in 34:23b, the reference to David as the “one shepherd” is further explained by two more comments (in MT) highlighting the uniqueness of Davidʼs role that likewise have no Leuven: Leuven University Press/Peeters, 2005), 437–446, here 441. Second, the argument of Lust and Crane rests upon some problematic assumptions, such as the fact that the term ‫מלך‬ would have more military overtones than ‫( נשיא‬Crane, Israelʼs Restoration [see n. 20], 126, 255), or that the order of chapters 36–40 would point to the greater implication of David in the upcoming battle against Gog. Third, and last, the use of the term ‫ מלך‬in the MT is consistent with a dominant theme in the oracle of 37:15–28, which refers to the recreation of a Davidic kingdom including Judah and Samaria (Ephraim). For these reasons, the wording of the OG in Ezek 37:24a is unlikely to be original and presumably reflects an alignment with the concept of Davidic leadership stated in Ezek 34:23–24. 84  For this view, compare also, e.g., L evin , Verheißung (see n. 59), 218–222, as well as Pohlmann, Das Buch (see n. 57), 467–468. 85  For a similar view, see Sedlmeier, Das Buch Ezechiel (see n. 47), 171–172. 86  As noted, e.g., by Z immerli , Ezechiel 2 (see n. 27), 841, who remarks in this regard: “Das betonte ‫ אחד‬ist von 37:15ff. her zu verstehen.” Pace Rudnig, Heilig (see n. 67), 65–68 (cf. similarly P ohlmann, Das Buch [see n. 56], 504–505), it is unlikely that vv. 24–28 followed immediately after v. 14 originally and that vv. 15–23 as a whole should be viewed as a later interpolation. This reconstruction has little evidence to support it, and it raises some issues, especially since the transition from v. 14, with its direct address in the second-person plural, to v. 25, where the Israelites are indirectly addressed (“they”), is awkward. Additionally, the separation between vv. 24 and 25 is unwarranted in my opinion. In short, the final statements constituting 37:24–28 presuppose at least some of the previous discourse on Israelʼs unification in vv. 15–19, 20–23. For a similar conclusion, see also Klein, Schriftauslegung (see n. 39), 172–175, with n. 234.

1038

Christophe L. Nihan

counterpart in Ezek 37.87 Fourth, and last, it is difficult to understand why Ezek 34 fails to mention the reestablishment of the sanctuary and the restitution of the divine presence in Israel, even though these figure prominently in the end of Lev 26:3–13 (see vv. 9–12), which, as argued above, was the source for Ezek 34. This observation can be satisfactorily accounted for, however, if the scribe who composed 34:23–30 already knew Ezek 37:24–28, where the restitution of the divine presence is expressly mentioned (see vv. 26–28). This assessment of the relationship between Ezek 34 and 37, if it is correct, provides an explanation for the difference between these two texts with regard to H. Though Ezek 37:24–28 does not (yet) appear to reflect the influence of H and may even have served as a source for the latter, the later composition of Ezek 34:23–31, which rewrites 37:24–28, is clearly concerned with coordinating the prophecy of Ezekiel with the phraseology of H (specifically, of Lev 26:3–13). This conclusion, in turn, has some significant implications for the broader issue of the relationship between Ezekiel and the Holiness legislation. On the one hand, it suggests that the scribal trend toward coordinating the prophecies of Ezekiel with the phraseology of H not only characterizes the expansions documented by the comparison between OG and MT but may also be identified in the materials shared by these textual forms – in this case Ezek 34:23–31. On the other hand, the difference between Ezek 34 and 37 with regard to H also suggests, simultaneously, that this scribal trend is not equally reflected in the various stages of the composition of the book, since it appears to be absent from the earlier text of Ezek 37:24–28. This conclusion must, by necessity, remain provisional: further analysis of the main stages that can be identified in the composition of the book and their relationship to H is required. Nevertheless, the previous analysis points to the need to consider the evidence for literary and compositional growth within the materials shared by Ezekiel MT and OG when assessing the relationship between these materials and the Holiness legislation. In particular, the finding that the reception of H is apparent in the later text of Ezek 34:23–31 but not in the earlier text of Ezek 37:24–28 contradicts any simplistic model that would construe the relationship between H and Ezekiel as straightforward and one-sided and suggests instead that the reception of H may be more characteristic of the later stages in the composition of Ezekiel than of the earlier forms of this book.

 See 34:23b: ‫הוא ירעה אתם והוא יהיה להן לרעה‬, “it is (only) he who will tend them, and it is only he who will be their shepherd.” As noted above, the first comment is missing from G and may represent a later gloss. 87

Ezekiel and the Holiness Legislation

1039

4 Conclusion: Toward Nonlinear Models Some general conclusions derive from this analysis with regard to the relationship between the Holiness legislation and Ezekiel, which can be briefly summarized by way of a conclusion. First, in keeping with some recent studies, the present essay has argued that the parallels between the two collections is part of a complex formative process, which impacted both the composition and the transmission of Ezekiel in the Second Temple period and which is documented by the comparison between the textual forms of this book preserved in the OG and the MT.88 Second, although the presence of scribal expansions characterized by the concern to coordinate the prophecy of Ezekiel more closely with the Holiness legislation is documented in both the OG and the MT, the textual evidence clearly suggests that the textual form preserved in the MT should be situated toward the end of this scribal process of coordinating Ezekiel with H. This conclusion, in turn, implies that any approach to the relationship between these two collections exclusively (or even predominantly) based on the MT of Ezekiel is inherently flawed. Third, the evidence provided by the comparison between Ezek 34:23–31 and 37:24–28 with regard to H (Lev 26) indicates that the relationship to H may differ according to the compositional stage reflected by these shared materials: though the later text of Ezek 34:23–31 arguably reflects the influence of H, this does not appear to be the case for the earlier text of 37:24–28. This conclusion, as noted above, suggests that the reception of H may, in fact, be more characteristic of the later stages in the composition of Ezekiel than of the earlier forms of this book. Overall, the findings presented here point to the need to elaborate complex, nonlinear models in order to adequately describe the relationship between the Holiness legislation and Ezekiel.

 Although it should go without saying, this point implies that models ascribing the book of Ezekiel to the eponymous prophet have a limited value for understanding the relationship between H and Ezekiel. 88

What Kind of Priestly Writings Did Ezekiel Know? Ariel Kopilovitz The literary connections between Ezekiel and the Priestly writings in the Pentateuch are well-known. The direction of influence is, however, a matter of debate. Some argue that Ezekiel preceded and influenced the Priestly writings,1 whereas others maintain that the Priestly writings as a whole – both P and H (hereafter referred to together as P) – preceded Ezekiel and had a major influence on his perceptions, language, and style.2

I. The Strata of Leviticus 26 Within P, Lev 26 seems to have the most literary connections to Ezekiel. But at what stage of Lev 26ʼs development did these literary connections come about? Every discussion of the literary connections between one biblical text and any other cannot ignore the question of composition history and the different strata that might exist within it. Leviticus 26 consists of four major parts: Verses 1–2 contain few directives and are actually the conclusion of the former literary unit. Verses 3–13 describe I am grateful to Baruch J. Schwartz, Yair Zakovitch, and Dalit Rom-Shiloni for their helpful comments on early drafts of this paper and to Martin (BG) Beifield for his help in preparing this paper for print. 1  See, e.g., J. Wellhausen, Prolegomena zur Geschichte Israels (Berlin: Georg Reimer, 1905), 377–378; J. L ust , “Exodus 6,2–8 and Ezekiel,” in Studies in the Book of Exodus: Redaction–Reception–Interpretation (ed. M. Vervenne; Leuven: Leuven University Press, 1996), 209–224; C. Nihan, From Priestly Torah to Pentateuch (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 543–545. 2  See, e.g., M. Fishbane, Text and Texture: Close Readings of Selected Biblical Texts (New York: Schocken Books, 1979), 131–132; J. Milgrom, Leviticus 23–27 (AB 3C; New York: Doubleday, 2001), 2352–2363; R. L evitt K ohn , A New Heart and a New Soul: Ezekiel, the Exile, and the Torah (JSOTSup 358; London: Sheffield Academic, 2002); M. A. Lyons, From Law to Prophecy: Ezekielʼs Use of the Holiness Code (New York: T&T Clark, 2009). For a different approach, which argues that the literary relations between Ezekiel and P are a result of a common priestly heritage and not of direct dependence of one upon the other, see A. Hurvitz, A Linguistic Study of the Relationship between the Priestly Source and the Book of Ezekiel: A New Approach to an Old Problem (Paris: J. Gabalda, 1982); M. Haran, “Ezekiel, P, and the Priestly School,” VT 58 (2008), 211–218.

1042

Ariel Kopilovitz

the promises if Israel maintains the covenant. Verses 14–39 describe the threats if Israel fails to maintain the covenant and are divided into five rounds by use of the phrase “if you do not obey Me” (vv. 14, 18, 21, 23, 27). Finally, vv. 40–45 depict Israelʼs restoration after exile. Many Lev 26 exegetes have pointed to its composite character and claimed that at least vv. 40–45 are a secondary addition.3 I accept this conclusion and wish to add a few more arguments to sustain it: Verse 38 says: “And you shall perish among the nations and the land of your enemies shall consume you.” The prime meaning of ‫ אבד‬is “to be extinct or die”;4 the basic meaning of ‫אכל‬, “eat,” is well-known, but frequently it also describes a total destruction that leaves nothing afterward.5 Verse 39 states that the remains will “rot in their iniquity,” and the use of ‫ נמק‬also denotes the gradual extinction of the few who will survive.6 It seems that vv. 38–39, which come after five rounds of threats, serve as a dim conclusion for the threats section. They depict the destruction and exile as the absolute end of the relationship between God and Israel, after which there will be no restoration, and they certainly do not anticipate the verses the now follow.7 Further indication to the secondary character of vv. 40–45 can be found in Lev 26ʼs structure. A similar literary structure,8 in which a few rounds of calamities are followed by an absolute and irrecoverable destruction, can be found in Amos 4:4–5:2. This prophetic unit contains five rounds of calamities that have already occurred (v. 6: hunger; vv. 7–8: withholding of rain; v. 9: pests to agriculture; v. 10: pestilence and sword; v. 11: destruction), with the repeating formula “Yet you did not turn back to me” in between. Here, as in Lev 26, the number of opportunities is limited and, after Israel has failed five times to turn back to God, it describes the final blow about to come: “Prepare to meet your God, O Israel!” (4:12). The consequences of this meeting will be tragic to Israel: 3  See, e.g., K. E lliger , Leviticus (HAT 4; Tübingen: Mohr Siebeck, 1966), 370; R. E. Friedman, The Exile and Biblical Narrative (Chico, CA: Scholars Press, 1981), 60–65, 102–104; H. L. Ginsberg, The Israelian Heritage of Judaism (New York: Jewish Theological Seminary of America, 1982), 79–81; B. A. L evine , Leviticus (JPS Torah Commentary 3; Philadelphia: Jewish Publication Society of America, 1989), 182–192; M ilgrom , Leviticus 23–27 (see n. 2), 2329–2330. 4  See the threat in Lev 23:30, which parallels the threat to be cut off (‫ )כרת‬in the preceding verse, and Num 33:50–56, where ‫ אבד‬is parallel to ‫שמד‬, “destroy.” See also Ezek 22:27; 26:17–20. 5  See Gen 31:15; Num 13:32; Deut 7:16; Jer 10:25; Ezek 36:13. 6  See Isa 34:4; Ezek 4:17; 24:23; 33:10; Zech 14:12; Ps 38:6. See also BDB 596–597, s.v. ‫ ;מקק‬HALOT 2:628–629, s.v. ‫מקק‬. 7  Note also the threat ‫וגעלה נפשי אתכם‬, “my soul shall abhor you” (v. 30), that certainly does not anticipate the promise ‫ולא געלתים‬, “neither will I abhor them” (v. 44). 8  See H. W. W olff , Joel and Amos (Hermeneia; Philadelphia: Fortress Press, 1977), 211–214; J. A. Soggin, The Prophet Amos: A Translation and Commentary (trans. J. Bowden; OTL; London: SCM, 1987), 78–79.

What Kind of Priestly Writings Did Ezekiel Know?

1043

“Hear this word which I intone as a dirge over you, O House of Israel: Fallen, not to rise again, is Maiden Israel; Abandoned on her soil With none to lift her up” (5:1–2).9 An examination of the opposite parallelism between the promises and the threats in Lev 26 will reinforce our conclusion that the original stratum of Lev 26 ended with the final calamity described in vv. 38–39. The threats are a broader and more intimidating mirror image of the promises. Usually they use the same phrases in order to invert the promises, and almost every promise has its opposite in the threats.10 Thus the promisesʼ opening formula, “If you follow my laws” (v. 3), parallels the threatsʼ opening formula, “if you reject my laws” (v. 15). The promise to “grant your rains in their season” (v. 4) is negated by the threat to “make your skies like iron and your earth like copper” (v. 19). The promise that the “earth shall yield its produce and the trees of the field their fruit” (v. 4) is negated by the threat that “your land shall not yield its produce, nor shall the trees of the land yield their fruit” (v. 20). The promise “you shall eat your fill of bread” (v. 5) is negated by the threat, “though you eat, you shall not be satisfied” (v. 26). The promise to “dwell securely in your land” (v. 5) is negated by the threats to desolate the land and scatter its inhabitants in vv. 31–33. The promise that “no sword shall cross your land” (v. 6) is negated by the threat to “bring a sword against you” (v. 25). The promise that “you shall lie down untroubled by anyone” (v. 6) is negated by the threats to bring “terror” (v. 16) and “faintness into their hearts” (v. 36). The promise to “remove dangerous beasts from the land” (v. 6) is negated by the threat to “loose wild beasts against you” (v. 22). The promise to “give chase to your enemies” (v. 7) is negated by the threat to be “routed by your enemies” (v. 17). The promise that God “will look with favor upon you” (v. 9) is negated by the threat, “I will set My face against you” (v. 17). The promise of Godʼs presence within Israel (vv. 11–12) is negated by the threats to desolate the sanctuaries and not to accept Israelʼs cultic offerings (v. 31). The promise that God will not “abhor” Israel (v. 11) is negated by the threat, “I will abhor you” (v. 30). Finally, the description of Israelʼs walk as

9  See S. M. Paul , Amos (Hermeneia; Minneapolis: Fortress, 1991), 141–160; M. W eiss , The Book of Amos (2 vols.; Jerusalem: Magnes, 1992), 1:112–133 (Hebrew). For the perception that, in one specific moment, Israelʼs verdict will become final and irreversible with no possibility of turning back, see Amos 7:8. For a comprehensive discussion of the relationship between Lev 26 and Amos 4, see John Kesslerʼs article in this volume, pp. 943–984. A similar repeating formula is found in Isa 9:7–10:4. This unit will not be discussed here, however, since the final, irreversible calamity is not mentioned explicitly and since its complicated literary connection (if there is one) to Isa 5:25–30 is a matter of debate. See G. H. A. Ewald, Jesaja mit den übrigen älteren Propheten (Göttingen: Vandenhoeck & Ruprecht, 1867), 286; O. K aiser , Isaiah 1–12 (OTL; Philadelphia: Westminster, 1983), 220; W. P. B rown , “The So-Called Refrain in Isaiah 5:25–30 and 9:7–10:4,” CBQ 52 (1990), 423–443. 10  See Levine, Leviticus (see n. 3), 185.

1044

Ariel Kopilovitz

“erect” is negated by the threat that “you shall have no power to stand against your enemies” (v. 37). At first sight, it seems that some promises and threats remained without an opposite parallel. Godʼs promises, “I will maintain My covenant with you” (v. 9) and “I will be your God, and you shall be My people” (v. 12), have no opposite threat, and the threats to “perish among the nations,” to be “consumed by the land of your enemies,” and to “rot in iniquity” (vv. 38–39) have no opposite promise. I would like to suggest that, despite the different terminology, exactly this was the opposite parallelism intended by the author of Lev 26:3–39. If Israel maintains the covenant, God will do so as well: he shall be their God and they shall be his people. But if Israel fails to maintain the covenant, the opposite will occur: they will perish in the lands of their enemies, and their covenant with God will reach its ultimate end; they shall no longer be his people and he shall no longer be their God.11 By assuming that the option to turn back will be exhausted in one particular moment, and that Israelʼs calamity will be final and irreversible, Lev 26:3–39 resembles many other treaties from the ancient Near East.12 The structure and phrasing of all these covenants indicate that like them, the original stratum of Lev 26 ended at v. 39. All these treaties ended with horrible threats of total and absolute extinction for the rebellious vassal, without any promise that he would be restored in the future. The images used in these threats and curses emphasized in the most explicit way the nature of the calamity that the rebellious vassal would suffer. Thus, for example, Shattiwaza of Mitanni declared in his covenant with Suppiluliuma I of Hatti, as a fir tree when it is felled has no more shoots, like this fir tree let me Shattiwaza [. . .] have no progeny. As the water of a drainpipe never returns to its place, let us, like the water of a drainpipe, not return to our place. [. . .] As salt has no seed, let me Shattiwaza [. . .] like salt indeed have no progeny.13  For a different opinion that argues that the Priestly covenant cannot be broken from Godʼs side, see J. S tackert , “The Holiness Legislation and Its Pentateuchal Sources: Revision, Supplementation, and Replacement,” in The Strata of the Priestly Writings (ed. S. Shectman and J. S. Baden; ATANT 95; Zürich: Theologischer Verlag, 2009), 187–204; Nihan, Priestly Torah (see n. 1), 535–541. However, both of them rely mainly on Lev 26:40–45 and not on Lev 26:3–39. 12  See, e.g., the treaty between Shattiwaza of Mitanni and Suppiluliuma I of Hatti in G. Beckman, Hittite Diplomatic Texts (WAW 7; Atlanta: Scholars Press, 1999), 44–50; the treaty between Bir-Ga‍ʾyah and Matîʾel in J. A. Fitzmyer, The Aramaic Inscriptions of Sefire (BibOr 19A; Rome: Pontifical Biblical Institute, 1995), 42–55; the treaty between Ashur-Nerari V and Matiʾ-ilu in S. Parpola and K. Watanabe , Neo-Assyrian Treaties and Loyalty Oaths (SAA 2; Helsinki: Helsinki University Press, 1988), 8–13; and the vassal treaties of Esarhaddon in D. J. Wiseman, “The Vassal Treaties of Esarhaddon,” Iraq 20 (1958), 1–99. 13  Beckman, Hittite Diplomatic Texts (see n. 12), 49. 11

What Kind of Priestly Writings Did Ezekiel Know?

1045

Esarhaddon threatened his vassals: as a caterpillar does not see and does not return to its cocoon, so may you not return to your women, your sons, your daughters, and to your houses.14

The felled tree that will never grow new shoots, the water drop that will never beat gravitation and return to its source, the salt that will never function as a growth substrate for anything, and the caterpillar that will never reverse metamorphosis all denote the terminal and irreversible nature of the calamity that the rebellious vassal will suffer. The symbolic acts conducted during these covenants were also meant to make clear that a rebellious vassal will never be restored. The “seizing of the throat” (napištam lapātum),15 which meant touching the throat of the covenant parties with either a knife or a finger, symbolized the death of those who would breach the covenant. Burning wax images symbolized the burning of the king and his land, breaking bows and arrows symbolized the loss of his military strength, and cutting an animal in two symbolized the destiny the rebellious king could expect.16 The acts, probably conducted in front of Matiʾ-ilu, illustrated in the most figurative way what would happen to him if he dared breach his treaty with Ashur-Nerari: This spring lamb has not been brought out of its fold for sacrifice [. . .] it has been brought to conclude the treaty of Aššur-nerari, king of Assyria with Matiʾ-ilu. If Matiʾilu [sins] against th[is] sworn treaty, then, just as this spring lamb has been brought from its fold and will not return to its fold [. . .] may, alas, Matiʾ-ilu [. . .] not return to his country and not [behold] his country again. This head is not the head of a spring lamb, it is the head of Matiʾ-ilu [. . .] if Matiʾ-ilu [should sin] against this treaty, so may, just as the head of this lamb is c[ut] off, and his knuckle placed in his mouth, [. . .] the head of Matiʾ-ilu be cut off and his sons [and magnates] be th[rown] into [. . .].17

All the images used in the threats and curses and the acts conducted during the conclusion of these covenants were meant to prevent the vassal from even thinking that rebellion would be worthwhile. The goal of the covenant was thus achieved, and the uncompromising loyalty of the vassal was guaranteed. It is not surprising, therefore, that none of these covenants included a promise of restoration to the rebellious vassal, since such a promise in advance would not achieve the desired deterrence. The curses in Deut 28:15–68 are no exception to this pattern. They stress several times that, if Israel does not maintain the covenant, they will become extinct, using the terms ‫שמד‬, ‫אבד‬, and ‫ כלה‬to em Parpola and Watanabe, Neo-Assyrian Treaties (see n. 12), 53–54.  See CAD 9:84–85, s.v. lapātu; H. Tadmor, “Treaty and Oath in the Ancient Near East: A Historianʼs Approach,” Shnaton 5–6 (1981–1982), 149–173, here 157 (Hebrew). 16  Fitzmyer, Aramaic Inscriptions (see n. 12), 46–47. 17  Parpola and Watanabe, Neo-Assyrian Treaties (see n. 12), 9. 14 15

1046

Ariel Kopilovitz

phasize this.18 It seems that, like Lev 26, Deut 28 holds that the consequence of breaching the covenant will be a terminal and irreversible calamity, with no option of restoration afterward.19 Leviticus 26:3–39 is thus another reflection of an ancient perception found in both treaties from the ancient Near East and biblical texts such as Amos 4–5 and Deut 28. Leviticus 26:38–39 does not anticipate the remembrance of the covenant and the renewal of the relationship between God and Israel mentioned just after,20 and they serve as a dim and hopeless end to the first stratum of the chapter. Accordingly, every discussion of literary connections to Lev 26 must consider the question whether these connections exist in regard to all its strata or only one of them.

II. Ezekiel in Its Contacts with Leviticus 26:3–39 What is the relation between Lev 26ʼs different strata and the book of Ezekiel? Our discussion will begin in Ezek 33,21 which has many literary connections to P: the phrase ‫דמו בו‬, “his bloodguilt shall be upon himself” (33:5);22 the phrase ‫דרש דם‬, “demand blood” (33:6), depicting God as demanding a reckoning for the blood of those who died unjustly;23 the description of sins as a heavy burden on the sinner (33:10);24 describing “following the laws” as guaranteeing life  See Deut 28:20, 21, 22, 24, 45, 48, 51, 61, 63.  The promises of restoration in Deut 30:1–10 were recognized by many as secondary additions. See, e.g., J. H. T igay , Deuteronomy (JPS Torah Commentary 5; Philadelphia: Jewish Publication Society of America, 1996), 432; R. D. N elson , Deuteronomy (OTL; Louisville: Westminster John Knox, 2002), 346. For a different opinion, see T. R ömer and M. Z. Brettler, “Deuteronomy 34 and the Case for a Persian Hexateuch,” JBL 119 (2000), 401–419, here 412. 20  See J. Kesslerʼs article in this volume, pp. 943–984, who suggests that Lev 26:40–45 are a secondary clarifying rereading of Lev 26:3–39 aiming to present the notion of the eternal validity of the covenant between God and Israel, which is lacking in 26:3–39. I am grateful to Professor Kessler for allowing me to see a former version of his paper. 21  This chapter was unjustly regarded by many as a secondary collection of paragraphs from other places in Ezekiel. See, e.g., G. B rin , “The Date and Meaning of the Prophecy against ‘Those Who Live in These Ruins in the Land of Israel’ (Ezekiel 33:23–29),” in Texts, Temples, and Traditions: A Tribute to Menahem Haran (ed. M. V. Fox et al.; Winona Lake, IN: Eisenbrauns, 1996), 29*–36* (Hebrew); L. Boadt, “The Function of the Salvation Oracles in Ezekiel 33–37,” HAR 12 (1990), 1–21; D. I. Block, The Book of Ezekiel, Chapters 25–48 (NICOT; Grand Rapids, MI: Eerdmans, 1998), 241; M. Greenberg, Ezekiel 21–37 (AB 22B; New York: Doubleday, 2004), 671–692. 22  See Lev 20:9, 11–13, 16, 27. 23  See Gen 9:5. 24  See Lev 5:1, 17; 10:17. See also B. J. S chwartz , “The Bearing of Sin in the Priestly Literature,” in Pomegranates and Golden Bells: Studies in Biblical, Jewish, and Near Eastern 18

19

What Kind of Priestly Writings Did Ezekiel Know?

1047

(33:15);25 and eating with or upon the blood as a forbidden abomination (33:25).26 Ezekielʼs retribution, ‫אתם‬ ֶ ‫וְ ִאיׁש ֶאת ֵא ֶׁשת ֵר ֵעהּו ִט ֵּמ‬, “you have defiled other menʼs wives” (33:26), shows that P influenced Ezekiel and not the other way around, ְ ‫וְ ֶאל ֵא ֶׁשת ֲע ִמ‬ since it is a conflation of two priestly rules: ‫יתָך לֹא ִת ֵּתן ְׁש ָכ ְב ְּתָך ְלזָ ַרע‬ ‫ל ָט ְמ ָאה ָבּה‬,ְ “Do not have carnal relations with your neighborʼs wife and defile yourself with her” (Lev 18:20), and ‫וְ ִאיׁש ֲא ֶׁשר יִ נְ ַאף ֶאת ֵא ֶׁשת ִאיׁש ֲא ֶׁשר יִ נְ ַאף ֶאת‬ ‫א ֶׁשת ֵר ֵעהּו‬, ֵ “If a man commits adultery with a married woman, committing adultery with another manʼs wife” (Lev 20:10).27 Besides these general connections to P, there are also specific resemblances between Ezek 33 and Lev 26. This is understandable given that Lev 26 is the only place in P that deals with what happens to Israel after they breach the covenant and are exiled, and Ezek 33 is the opening of a new unit of prophecies dealing with this same issue: Israelʼs restoration after the destruction. Indeed, both chapters mention God “bringing the sword” against Israel (Lev 26:25; Ezek 33:2), a phrase common in Ezekiel but that appears only once in the pentateuchal traditions;28 both use the unique word ‫גילולים‬, “fetishes” (Lev 26:30; Ezek 33:25);29 and both describe God as making the land desolate (Lev 26:31–33; Ezek 33:28–29). The last two examples are significant for discussion about the direction of influence. First, both chapters mention the unique idiom ‫גאון עוז‬, “proud glory,” which will be broken or ceased (Lev 26:19; Ezek 33:28). This idiom occurs in the pentateuchal traditions only once, and four times in Ezekiel, each time with a different meaning. In Ezek 24:21, it refers to the temple; in 30:6, 18 to the strength of Egypt; and in 33:28 to the strength of the land. It seems more plausible that Lev 26 influenced the several cases in the book of Ezekiel than that the author of Lev 26 gathered phrases and idioms scattered with diverse meanings all over Ezekiel and integrated them into his composition.30 Ritual, Law, and Literature in Honor of Jacob Milgrom (ed. D. P. Wright et al.; Winona Lake, IN: Eisenbrauns, 1995), 3–21. 25  See Lev 18:4–5. 26  See Lev 19:26. For the meaning of this phrase, see B. J. S chwartz , The Holiness Legislation (Jerusalem: Magnes, 1999), 344–346 (Hebrew). 27  See L evitt K ohn , New Heart (see n. 2), 40, 78; Lyons , From Law to Prophecy (see n. 2), 66: “the presence of conflation [. . .] is generally considered to be a quality of the later, dependent text.” See also Schwartz, Holiness Legislation (see n. 26), 135–144, who argues that Lev 18 and 20 are alternative literary traditions that were combined by the editors of H. This might reinforce our claim that Ezekiel was influenced by H, since such a literary connection is more likely to happen after these two traditions were compiled into the canonical H. 28  See Ezek 5:17; 6:3; 11:8; 14:17; 29:8 (against Egypt). It seems that the “sword” mentioned in Deut 28:22 is a scribal error. It is not represented in LXXB and it seems that the readings ‫חֹרב‬, “heat and dryness” (see Vulg. aestu) or ‫חרס‬, “scabies” (see Deut 28:27 and HALOT 1:355, s.v. ‫)חרס‬, are preferable and better fit the context of different diseases. 29  For further discussion of this term, see below, p. 1053. 30  See also Lyons, From Law to Prophecy (see n. 2), 84.

1048

Ariel Kopilovitz

Second, both chapters describe the exiles with the rare verb ‫ נמק‬together with ‫ עון‬or ‫( חטא‬Lev 26:39; Ezek 33:10).31 As M. Lyons has argued, this last example proves that Ezekiel relied on P and not the other way around and that, when doing so, he omitted the mention of vertical retribution he found in Lev 26:39 (“And those who are left over among you will rot in their iniquity [. . .] and furthermore, in the iniquities of their fathers with them they will rot”) since it did not tally with his perception that “the person who sins, only he shall die” (18:4; see also 33:6).32 The fact that all these literary connections also include unique phrases and idioms shows that Ezekielʼs reliance on P was more than a mere ideological reliance and that Ezekiel was familiar with some written version of Lev 26.

III. Ezekiel and Leviticus 26:40–45 There are also many differences between these two chapters. In 33:10, Ezekiel ascribes to the exiles the saying, “Our transgressions and our sins weigh heavily upon us; we are rotting because of them. How can we survive?”33 Many commentators on Ezekiel have justly noted that this saying reflects the exilesʼ recognition that the threats of Lev 26:39 were fulfilled in them and that this caused them total despair.34 But, if Ezekiel and his audience indeed knew a written version of Lev 26 that included the unique threat to “rot in iniquity,” why were they so full of despair? All they had to do was read on a few sentences and discover that it all depended on them: if only they would confess their iniquity and humble their obdurate hearts, God would remember his covenant and not reject them. It seems that the total despair ascribed to Ezekielʼs audience does not reflect awareness of Lev 31  ‫ נמק‬occurs only ten times in the Hebrew Bible. The combination of ‫ נמק‬and ‫עון‬/‫ חטא‬is unique to Lev 26 and Ezek 4:17; 24:23; 33:10. 32  Lyons, From Law to Prophecy (see n. 2), 61–62, 85. It should be mentioned, however, that the question whether Ezekiel had a systematic retribution conception is a matter of debate. 33  Some doubt the authenticity of this quote. See, e.g., H. G. May, The Book of Ezekiel (IB 6; Nashville: Abingdon, 1956), 246; J. W. Wevers, Ezekiel (CB; London: Oliphants, 1976), 177. According to M. Greenberg, “The Citations in the Book of Ezekiel as a Background for the Prophecies,” Beit Mikra 17 (1972), 273–278 (Hebrew), most of the citations in the book of Ezekiel were not actually said by the people but rather created by the prophet. Nevertheless, Greenberg admits that these citations do reflect trends within the prophetʼs audience. Regarding quotes of the nations (Ezek 25:3, 8; 26:2; 28:2; 29:3; 35:12), there is no doubt that Greenberg is right. However, regarding quotes of the Judean exiles, I agree with A. Graffy, A Prophet Confronts His People (Rome: Biblical Institute Press, 1984), 50–53, and think that there is no reason to doubt their authenticity, especially when one of them is also mentioned outside the book of Ezekiel (see Jer 31:28; Ezek 18:2). 34  See, e.g., G raffy , A Prophet Confronts His People (see n. 33), 73; B lock , Ezekiel 25–48 (see n. 21), 246.

What Kind of Priestly Writings Did Ezekiel Know?

1049

26:40–45; according to their perception, the exile was the end of the relationship between God and Israel (see also Ezek 37:11). We should notice that Ezekiel himself does not disagree with the peopleʼs basic understanding of their situation and does not declare that it has no reasonable ground. When he cites what they are saying, he says: ‫ֵּכן ֲא ַמ ְר ֶּתם ֵלאמֹר ִּכי‬ ‫אתינּו‬ ֵ ֹ ‫פ ָׁש ֵעינּו וְ ַחּט‬. ְ The role of the word ‫ כי‬in this verse is to mark the beginning of the direct quote. In this regard, it functions like the modern colon.35 Thus, the word ‫ כן‬cannot also mark the direct quote, in the sense of “thus, in this exact manner you have said,”36 but rather emphasizes Ezekielʼs evaluation of the peopleʼs speech – that is, “you are right in saying that you are rotting in inequity and cannot survive.” Ezekiel admits that the exilesʼ understanding of their situation is correct and perfectly matches Pʼs written edition, which was known to them – indeed, there is no other way to understand Lev 26ʼs threats.37 However, here lies one of Ezekielʼs great innovations: although God has promised in P that breaking the covenant will lead to death, now he desires something else – life. They were indeed sentenced to death, but this sentence is reversible and it all depends on them. Therefore, Ezekiel calls the exiles to turn back from their evil ways and promises that, if they do so, they will have hope. The exilesʼ reaction to this, however, is ‫( לֹא יִ ָּת ֵכן ֶּד ֶרְך ֲאד ֹנָ י‬33:17, 20), which, as A. Graffy has shown, emphasizes not defiance but amazement and lack of ability to measure and understand divine grace in giving the option to turn back from evil even after rotting in iniquities: “The way of God cannot be fathomed.”38 But Lev 26 depicts the “way of God” exactly and in advance: the people will rot in iniquity and afterward confess, and God will remember his covenant with them. Why, then, are the exiles so amazed  See, e.g., Gen 21:30; 29:33; Exod 3:12; Judg 6:16; 1 Sam 10:19; See also BDB 471–472, s.v. ‫כי‬. 36  This was the opinion of most commentaries on Ezekiel. See, e.g., Block, Ezekiel 25–48 (see n. 21), 244; R. Kasher, Ezekiel: Introduction and Commentary (2 vols.; Jerusalem: Magnes, 2004), 2:636 (Hebrew). However, this suggestion creates a major difficulty in the verseʼs sequence, since it claims that three of its parts – ‫כן‬, ‫לאמר‬, ‫ – כי‬have the same syntactic role. 37  Further support for this (correct) understanding can be found in the Talmud, b. Makkot 24a: “Said R. Yossi son of Hanina: Our master Moses pronounced four sentences on Israel but four prophets came and revoked them. [. . .] Moses said ‘the Lord visits the iniquity of the fathers upon the children,’ Ezekiel came and declared ‘the soul that sins, it shall die.’ Moses had said: ‘And you shall perish among the nations,’ Isaiah came and said ‘and it shall come to pass in that day that a great horn shall be blown [and the lost will come from the land of Assyria].’ Rab said: I am afraid of this verse, ‘And you shall perish among the nations.’ R. Papa demurred, saying – could it not perhaps rather be taken in the sense of something lost and searched for, as is written, ‘I have gone astray like a lost sheep, seek Thy servant’? But it was the latter part of that verse [that frightened Rab], ‘[. . .] and the land of your enemies shall consume you’ ” (The Babylonian Talmud [trans. H. M. Lazarus; London: Soncino, 1935], 173–174). 38  Graffy, A Prophet Confronts His People, 76–78 (see n. 33). 35

1050

Ariel Kopilovitz

by Godʼs way and think that it cannot be fathomed? Here as well, it seems that the reaction attributed to the exiles does not reflect any acquaintance with the last verses of Lev 26; Ezekielʼs words in chapter 33 are (rightly) a complete innovation to them. Not only does the exilesʼ distress reflect lack of acquaintance with Lev 26:40– 45, but so does the solution Ezekiel offers them. According to Lev 26:40–41, the peopleʼs restoration will be a result of a few elements: it is expected that they will confess – meaning, specify their sins in words;39 afterward their obdurate hearts must be humbled, a phrase depicting a change in Israelʼs rebellious attitude; and, finally, their iniquity must be paid or atoned for without specifying how exactly this will be executed.40 Only afterward will God remember his covenant and refrain from destroying them. Ezekiel, however, never mentions the term ‫התוודה‬, “confess,” either in chapter 33 or in any other place in his book. This fact is striking since the peopleʼs declaration, “Our transgressions and our sins weigh heavily upon us” (33:10), is in fact a confession!41 The humbling of the heart and the payment of iniquity are also not mentioned in Ezekiel. Instead, Ezekiel adopts the rhetoric of repentance and return from evil ways,42 a term mentioned neither in Lev 26 nor in Pʼs atonement ideology. That Ezekiel refrains from referring to Lev 26:40–41 in such a place shows that the restoration procedure mentioned there was not familiar to him. Leviticus 26 and Ezekiel even have two different approaches to the reason why God will accept the change in Israelʼs behavior. In Leviticus, remembering the covenant with the patriarchs and the ancients will prevent God from destroying Israel. In Ezekiel, however, two alternative explanations can be found. According to the first, in chapter 33, the divine motive is dependent on human actions. God does not desire the death of the wicked, and if the wicked change their ways, God will accept this change (33:11–20). The second explanation, found in chapter 36, is theocentric in character. Israelʼs existence in exile causes Godʼs name to be profaned and, in order to stop it, Israel must be taken out  The confession appears few times in P as a part of the atonement process, especially in regard to deliberate sins. See Lev 5:5; 16:21; Num 5:7. See also J. Milgrom, Leviticus 1–16 (AB 3A; New York: Doubleday, 1991), 301–303. 40  It seems that the meaning of ‫ רצה עון‬is a complete and satisfying payment of the debt created as a result of the sin. See Isa 40:2; BDB 953, s.v. ‫ ;רצה‬HALOT 3:1282, s.v. ‫;רצה‬ G. Gerleman, “rṣh,” TLOT 3:1259–1261. According to M. J. Boda, A Severe Mercy: Sin and Its Remedy in the Old Testament (Siphrut 1; Winona Lake, IN: Eisenbrauns, 2009), 82–85, the payment will be executed through the suffering caused by the curses (threats) and exile. 41  See D. Rom-Shiloni, God in Times of Destruction and Exile (Jerusalem: Magnes, 2009), 280–281 (Hebrew). 42  According to Levitt Kohn, New Heart (see n. 2), 92, 103, this reflects the use of Deuteronomic concepts and expressions. In chapter 36, Ezekiel presents a different option, according to which Israelʼs restoration will occur not as a result of their actions but in order to avoid defiling Godʼs name. This option also does not tally with Lev 26. 39

What Kind of Priestly Writings Did Ezekiel Know?

1051

of there, whether they change their ways or not (36:16–36). The relationship between these two explanations cannot be discussed here, but neither of them is related to the covenants with the patriarchs and the ancients mentioned in Leviticus. Moreover, Lev 26:44 is an essential part of the restoration program in Lev 26: “Yet, even then, when they are in the land of their enemies, I will not reject them or abhor them so as to destroy them, annulling My covenant with them: for I the LORD am their God.” According to this, Israelʼs restoration is inevitable since, no matter what happens, God will not annul his covenant with them. If Ezekiel was familiar with this notion, why didnʼt he use it as a response to the exilesʼ despair? It seems that this is a further indication that Ezekiel was not familiar with Lev 26ʼs restoration program. An alternative explanation for the differences between these chapters is that Ezekiel did know Lev 26:40–45 but deliberately decided to replace its restoration program with another.43 This explanation seems less plausible, for a number of reasons. First, all parts of Lev 26 have parallels in Ezekiel except verses 40–45.44 Indeed, there are some idioms that appear both in Lev 26:40–45 and in Ezekiel. However, methodologically, when discussing the question of literary connections between texts, one must point out the unique resemblances that show the connections between these specific texts. In this case, idioms such as ‫ מעל מעל‬and ‫משפטי מאסו‬, which do appear in Lev 26:40–45, also appear in other parts of P.45 Therefore, their occurrences in Ezekiel are not necessarily a result of literary dependence on Lev 26:40–45, and it could equally be a result of dependence on other occurrences of these idioms in P. On the other hand, idioms that are unique to Lev 26:40–45, such as confession, humbling of the heart, and payment of iniquity, do not occur in Ezekiel at all. This fact serves as further reinforcement of the conclusion that these verses were not part of the written edition of Leviticus that influenced Ezekiel but were added to it only afterward. Second, Ezekiel does not show any awareness of Lev 26:40–45. Matters such as Israelʼs actual restoration, which are presented in Leviticus as inevitable or at least possible, are understood by Ezekielʼs audience as impossible. Finally, the watchman prophecy in Ezek 33:1–9 reflects Godʼs desire for Israelʼs repentance, the possibility of this process, and the crucial influence it might have on their fate. Its positioning as the opening chord of Ezekielʼs restoration oracles emphasizes its importance and centrality in Ezekielʼs restoration program46 and proves that it was not merely a rhetorical device.47 In this regard,  Thus, e.g., Lyons, From Law to Prophecy (see n. 2), 85–87, 123.  See J. Milgrom, “Leviticus 26 and Ezekiel,” in The Quest for Context and Meaning (ed. C. A. Evans and S. Talmon; Leiden: Brill, 1997), 57–62. 45  See, e.g., ‫ מעל מעל‬in Lev 5:15, 21; Num 5:6, 12, 27 and ‫ משפטי מאסו‬in Lev 26:15. 46  See Ezek 3:19, 20; 13:22; 14:6; 18:21, 23, 24, 26, 27, 28, 30, 32; 33:9, 11, 12, 14, 18, 19. 47  As suggested by P. M. Joyce, Divine Initiative and Human Response in Ezekiel (JSOTSup 51; Sheffield: JSOT Press, 1989). 43 44

1052

Ariel Kopilovitz

Ezek 33:12–20 invents the principle not only that a person does not suffer for his fatherʼs deeds but also that his repentance might save him from suffering for his own past deeds. In these circumstances, it seems less probable that Ezekiel would use Lev 26:39 to describe Israel rotting in their iniquity and deliberately choose not to continue using vv. 40–41. These verses are highly relevant. They match Ezekielʼs restoration program perfectly, and Ezekielʼs audience behaves exactly as Lev 26:39–41 predicted it would after rotting in iniquity: they confess their guilt and humble their hearts. Although a dependent text does not necessarily have to use all parts of the text on which it is dependent, it seems that, in this case, we would expect it to do so. The fact that this did not occur reinforces the conclusion that Ezekiel did not know these verses. It seems, then, that Ezekiel knew a written edition of P that did not include Lev 26:40–45. The independent discussion of the exegetical problems within Lev 26 that led to the conclusion that the chapter is composed of more than one stratum is thus reinforced through discussion of its relationship with Ezekiel. The book of Ezekiel, and perhaps other prophetic books as well, may function as a camera that captures the Pentateuchʼs text in one specific moment, enabling us to better understand the different strata within it and the process and date of their composition. It is well accepted today that P was composed not in a single stroke but rather through a long process of composing, compiling, and editing that probably spanned generations. It is just as implausible to claim that Priestly literary activity came to its absolute end after the destruction of the first temple as it is to claim that it began suddenly in the Persian period. Our discussion shows that, even after the composition of Ezekiel, Priestly circles continued to create Scripture and their work was combined into P. Did Ezekielʼs prophecies in turn influence the composition of later strata in P?48 I would like to mention two instances where the answer might be affirmative and perhaps lead to further discussion. Ezekiel 6 and Lev 26 share many similarities.49 Methodologically, when trying to determine the direction of influence between two texts with similar terms, one must consider where these terms fit in the general context and where they create incongruity that might indicate a secondary use. In this case, it seems that Lev 26 influenced 48  This has been offered in some changes by W. Zimmerli, A Commentary on the Book of the Prophet Ezekiel Chapters 1–24 (trans. R. E. Clements and J. D. Martin; Hermeneia; Philadelphia: Fortress, 1979), 52. See also Christophe Nihanʼs paper, in this volume, pp. 1015–1039. This phenomenon, in which text A precedes and influences text B, and afterward, text B in turn influences some later strata in text A, was defined by Yair Zakovitch as the “boomerang phenomenon.” See Y. Zakovitch, “The Book of the Covenant Interprets the Book of the Covenant: The Boomerang Phenomenon,” in Texts Temples and Traditions: A Tribute to Menahem Haran (ed. M. V. Fox et al.; Winona Lake, IN: Eisenbrauns, 1996), 59*–65* (Hebrew). 49  See, e.g., “bring a sword” (Lev 26:25; Ezek 4:2); “destroy/cut down/desolate cult places/ altars” (Lev 26:30; Ezek 6:3, 4, 5); “incense stands” (Lev 26:30; Ezek 6:4).

What Kind of Priestly Writings Did Ezekiel Know?

1053

Ezek 6. In Lev 26, the threats are addressed to the people. In Ezek 6, however, the prophecy is addressed to the “mountains of Israel” (v. 2), and the sudden reference to corpses and bones in verses 4–5 does not fit the context surrounding them, indicating their secondary nature. Nevertheless, regarding the term ‫גילולים‬, the overall picture is more complex. First, ‫ גילולים‬refers to an inanimate object functioning as a fetish or idol.50 This term occurs only once in P but is typical in Ezekiel, which contains thirty-nine out of its forty-eight occurrences in the Hebrew Bible.51 Second, though Ezekiel prophesies that the carcasses of Israel will be cast “in front of their fetishes,”52 Lev 26:30 maintains that they will be cast “upon your fetishesʼ carcasses.” The description of the death of Israel in front of their idols emphasizes not only the defilement of the sanctuaries where these idols stood (see Ezek 9:7) but also their inability to save their believers in times of need. All occurrences of the word ‫ פגר‬in the Hebrew Bible refer to a carcass of a human being or an animal; the word never describes inanimate objects such as idols. It seems that mentioning the fetishes and Israelʼs carcasses falling in front of them was original in Ezekiel and, when this term typical to Ezekiel in turn influenced Lev 26, the word ‫פגר‬, “carcass,” that originally referred to Israel alone, was expanded to describe the fetishes as well, causing incongruity in its new context. Perhaps the same process occurred regarding the idiom ‫לעיני הגוים‬, “in the sight of the nations.” The notion of divine action “in the sight of” different witnesses is frequent in Ezekiel.53 It emphasizes the impression that this action will leave and the knowledge it will create.54 Several times Ezekiel stresses that the witnesses to the divine act will be the nations. Israelʼs calamity will occur “in the sight of the nations” (5:8, 14; 22:16), their existence in exile causes Godʼs holy name to be profaned “in the sight of the nations” (36:20–23), and this by itself requires that their restoration be “in the sight of the nations” as well (20:41; 28:25; 39:27). It is well-known that Ezekielʼs descriptions of Israelʼs future restoration were considerably influenced by the manner in which Israelʼs exodus

50  See BDB 165, s.v. ‫ ;גלל‬see also its parallels to ‫( אלילים‬Ezek 30:13) and ‫( תרפים‬2 Kgs 23:24). 51  The other eight occurrences appear in Deuteronomy (one), Kings (six), and Jeremiah (one). 52  The words “I will cast the carcasses of the people of Israel in front of their fetishes” are missing in the LXX, perhaps a result of homoioteleuton, where the copyist mistakenly omitted the sentence between the two occurrences of ‫גילולים‬. 53  See, e.g., 4:12; 5:8, 14; 12:3–7, 9; 16:41; 21:11; 22:16; 28:18, 25; 36:34; 37:20; 38:16, 23; 39:27; 43:11. 54  See, e.g., the many occurrences of the phrase ‫וידעתם כי אני יהוה‬/‫וידעו‬, “And they/you shall know that I am the LORD,” in Ezekiel: 5:13; 6:10, 14; 7:27; 12:15–16; 17:24; 21:10; 24:27; 25:11, 17; 26:6; 28:22–24, 26; 29:6, 9, 16, 21; 30:8, 19, 25, 26; 32:15; 33:29; 34:27, 30; 35:15; 36:23, 36, 38; 37:28; 38:23; 39:6, 7, 22, 28.

1054

Ariel Kopilovitz

out of Egypt was described.55 However, the opposite process occurred as well, and in several cases the manner in which Ezekiel described Israelʼs present and future restoration in turn influenced the manner in which he chose to describe the past exodus events. One typical example is the metahistorical description of the exodus and the wandering in the desert in Ezek 20. The description of Israelʼs idolatry in Egypt (vv. 7–8; see also 22:3, 8, 21) has no basis in the pentateuchal traditions, and it seems that the way Ezekielʼs contemporaries behaved before and during exile influenced the manner in which he chose to depict Israelʼs behavior in Egypt. The description of the exodus as occurring “in the sight of the nations” (20:9, 14, 22) also has no basis in the pentateuchal traditions, which never state that the exodus occurred that way, and there is no report of nations other than Israel who dwelled in Egypt and might have witnessed the exodus. On the contrary, all the pentateuchal traditions (whether P, D, or non-P and non-D) stress over and over again that the exodus events occurred “in the sight of” Israel and Egypt alone, and only they were expected to draw the appropriate conclusions regarding Godʼs might.56 In this case as well, the importance of Israelʼs future restoration “in the sight of the nations” in Ezekielʼs restoration program influenced the manner in which he chose to depict the exodus. This notion fits very well in the broad context of Ezek 20, which describes Israel as sinners from the very beginning and in which Godʼs action in order to prevent profaning his holy name is a major factor in Israelʼs history. However, the description of the exodus “in the sight of the nations” in Lev 26:45 does not fit its context. Besides the fact that it has no basis in the narratives that preceded it, the manner in which God is perceived by the nations has no role within it and no effect on Israelʼs future restoration. It seems that here, as well, reference to the exodus taking place “in the sight of the nations” was original in Ezekiel and, when this notion typical to Ezekiel in turn influenced Lev 26, it caused incongruity in its new context. We may conclude, then, that when dealing with the literary relations between the pentateuchal traditions and prophetic literature, we must reexamine the methodological questions at stake: It is not sufficient to ask whether the pentateuchal traditions influenced prophetic literature. We must rather ask which strata within the pentateuchal traditions provided the influence. Regarding research on the book of Ezekiel and its relationship to P, it seems that we should ask not only whether P influenced Ezekiel or Ezekiel influenced P but also when P influenced Ezekiel and when Ezekiel in turn influenced P.  In addition to the phrase “And they/you shall know that I am the LORD” in Exod 6:7; 7:5; and 10:2, see also ‫ביד חזקה ובזרוע נטויה‬, “with a strong hand, and with an outstretched arm” (Exod 6:6; Deut 4:34; 5:15; Ezek 20:33–34); ‫עשה שפטים‬, “execute judgments” (Exod 12:2; Ezek 5:10); ‫נטה יד‬, “stretch out [oneʼs] hand against” (Exod 7:5; Ezek 6:14; 14:9, 13); and the exodus out of Egypt/exile into the desert (Ezek 20:35). 56  See Exod 7:5, 20; 8:18; 9:8, 29–30; 10:2; 11:3; 14:3, 18; Num 33:3; Deut 1:30; 4:34; 6:22; 7:19; 11:7; 29:1. Awareness of the exodus by nations other than Israel and Egypt is mentioned only in Exod 15:14–15; 18:1, but even there hearing is depicted, not directly witnessed. 55

How Have We Changed? Older and Newer Arguments about the Relationship between Ezekiel and the Holiness Code1 Michael A. Lyons 1 Introduction It is widely recognized that there are a remarkable number of locutions common to Lev 17–26 (the Holiness Code) and the book of Ezekiel. The quality, frequency, and distribution of these locutions are such that most agree they can be explained only by a model of literary dependence – either by one text borrowing from the other or by mutual dependence during a process of textual formation. There is, however, no consensus on the direction of literary dependence. This does not (to me, at least) constitute a crisis; readers will naturally construe these texts in different ways due to the complex nature of cognition and the complexities of the texts themselves. Yet the lack of consensus does suggest that we look closely at, and think critically about, the criteria we have traditionally used to determine textual relationships. In this essay, I will review early arguments about the direction of literary dependence between the Holiness Code (in particular, Lev 26) and Ezekiel. I will then examine the extent to which we have (or have not) moved beyond the criteria used to support these arguments.2 Finally, I will conclude with reflections about how we have changed. It is my hope that this will inspire greater methodological awareness when analyzing relationships between texts and that it will encourage greater dialogue between specialists in pentateuchal literature and those in prophetic literature.

1  I wish to thank Christophe Nihan for his helpful comments and suggestions. The opinions expressed in this paper (and any errors present) are my own. 2  This study is necessarily limited to representative examples; I make no attempt here to offer a comprehensive survey of scholarship on H or Ezekiel.

1056

Michael A. Lyons

2 The Initial Period: The Positions Are Developed Up to – and even after the rise of – the era of critical approaches to biblical literature, readers naturally supposed (based on the narrative frame, e.g., Lev 26:46) that the contents of Lev 17–26 were preexilic. The presence of shared locutions in Leviticus and Ezekiel were explained as instances in which the prophet borrowed from Leviticus.3 Thus commentators did not find it necessary to provide explicit arguments to support their statements about the direction of literary dependence. But the traditional approach to the relationship between Lev 17–26 and Ezekiel would undergo a radical challenge in the late 1800s. Within a span of twenty years (1876–1896), three models would be developed to explain the direction of the relationship between the Holiness Code and Ezekiel: either H was an exilic/early postexilic composition borrowing from Ezekiel (Wellhausen),4 or Ezekiel was borrowing from a preexilic H (Klostermann, Driver, Paton),5 or there was a complex process of mutual literary dependence as the Holiness Code underwent composition and growth (Baentsch).6 All these models 3  See, e.g., Calvinʼs remark in his commentary on Ezekiel (published posthumously in 1565) regarding the locution “break the staff of bread” (Ezek 4:16): “And this clearly appears from Lev. xxvi. 26, whence our Prophet has adopted this expression”; J. C alvin , Commentaries on the First Twenty Chapters of the Book of the Prophet Ezekiel (trans. T. Myers; Edinburgh: Calvin Translation Society, 1848), 187. See also W. M. L. de W ette , Lehrbuch der historisch-kritischen Einleitung in die kanonischen und apokryphischen Bücher des Alten Testaments (5th ed.; Berlin: Reimer, 1840), 317; F. Hitzig, Der Prophet Ezechiel Erklärt (Kurzgefasstes exegetisches Handbuch; Leipzig: Weidmann, 1847), 35 (on Ezek 4:16–17); C. F. Keil, Biblischer Commentar über den Propheten Ezechiel (Leipzig: Dörffling und Franke, 1868), 59 (on Ezek 6:4–6). 4  J. Wellhausen, “Die Composition des Hexateuchs,” Jahrbücher für deutsche Theologie 21 (1876), 392–450, 531–602; 22 (1877), 407–479; republished as Die Composition des Hexateuchs und der historischen Bücher des Alten Testaments (3rd ed.; Berlin: Reimer, 1899), esp. 150–68; idem , Prolegomena zur Geschichte Israels (2nd ed.; Berlin: Reimer, 1883), translated as Prolegomena to the History of Ancient Israel (trans. J. S. Black and A. Menzies; Edinburgh: A&C Black, 1885), esp. 376–385. 5  A. Klostermann, “Beiträge zur Entstehungsgeschichte des Pentateuchs,” Zeitschrift für die gesammte lutherische Theologie und Kirche 38 (1877), 401–445, republished as “Ezechiel und das Heiligkeitsgesetz,” in Der Pentateuch: Beiträge zu seinem Verständnis und seiner Entstehungsgeschichte (Leipzig: G. Böhme, 1893), 368–418; S. R. Driver, An Introduction to the Literature of the Old Testament (New York: Scribnerʼs Sons, 1891), 43–55, 138–144; L. B. Paton, “The Holiness-Code and Ezekiel,” Presbyterian and Reformed Review 26 (1896), 98–115. 6  B. B aentsch , Das Heiligkeits-Gesetz Lev. XVII–XXVI: Eine historisch-kritische Untersuchung (Erfurt: Hugo Güther, 1893). According to Baentsch, Hʼs earliest layer (H1) consisted of Lev 18–20, 23–25, and this material was used in Ezekiel. A subsequent layer (H2), consisting of Lev 21–22, contained priestly and cultic matters; this layer was composed after Ezekiel. A final redactor combined the first two layers, prefaced them with a third (H3, consisting of Lev 17), and appended a sermon (Lev 26), which he expanded. Baentsch concluded

How Have We Changed?

1057

were in some part responses to K. H. Grafʼs theory that Ezekiel himself had written the Holiness Code.7 But they should be viewed as substantive positions in their own right, attempts to explain the distinctiveness of the material in Lev 17–26 and account for its relationship with Ezekiel. It is appropriate to begin this survey in 1876 with the arguments of Julius Wellhausen, who deserves recognition not only for his literary and historical acumen but also because his model would shape the course of subsequent scholarship. One might rightly claim that Wellhausenʼs contribution lay just as much in the fact that he made arguments as in the content of his arguments. What is so significant about his work is the recognition that one cannot simply assume the temporal priority of H over Ezekiel – which is what generations of earlier commentators had done.8 From this point on, commentators would have to provide arguments for their assessment of the direction of literary dependence. This is not to say that Wellhausen was first to espouse the position that the shared material in Lev 17–26 and Ezekiel could be explained as the literary dependence of the former on the latter. Such a position had already been advanced by both M. M. Kalisch and A. Kuenen.9 However, at this time, neither Kalisch nor Kuenen treated Lev 17–26 as an originally independent unit in the same way that others would just a short while later. Furthermore, while Kalisch anticipated many of Wellhausenʼs arguments regarding the dating of P and H, he was not to enjoy the same influence.10 As for Kuenen, he both influenced Wellhausen and was in turn influenced by him.11 that the material in Lev 26 borrowed extensively from Ezekiel, and here he drew heavily on Wellhausenʼs arguments (69–71, 86, 114, 120–128). 7  K. H. Graf, Die geschichtlichen Bücher des Alten Testaments: Zwei historisch-kritische Untersuchungen (Leipzig: T. O. Weigel, 1866), 75–83. 8  Wellhausen, Prolegomena (see n. 4), 379: “To assume that Ezekiel, having the Pentateuch in all other respects as we have it, had a great liking for this piece of it, and made it his model in the foundation of his style of thought and expression – such an assumption does not free us from the necessity of seeking the historical order, and of assigning his natural place in that order to Ezekiel; we cannot argue on such a mere chance.” 9  M. M. Kalisch, A Historical and Critical Commentary on the Old Testament, with a New Translation: Leviticus, Part II (London: Longmans, Green, 1872), 291–292, 313, 360–363; A. K uenen , De Godsdienst van Israel tot den ondergang van den joodschen staat (2 vols.; Haarlem: A. C. Kruseman, 1869–1870); English translation, idem , The Religion of Israel to the Fall of the Jewish (3 vols.; trans. A. H. May; London: Williams & Norgate, 1874–1875), 2:150–153, 189–191. 10  See J. W. Rogerson, Old Testament Criticism in the Nineteenth Century (Philadelphia: Fortress, 1985), 242–244; B. J. S chwartz , “The Pentateuch as Scripture and the Challenge of Biblical Criticism: Responses of Modern Thinkers and Scholars,” in Jewish Concepts of Scripture: A Comparative Introduction (ed. B. D. Sommer; New York: NYU Press, 2012), 203–229, here 213–214. 11  See R. Smend, “Kuenen and Wellhausen,” in Abraham Kuenen (1828–1891): His Major Contributions to the Study of the Old Testament (ed. P. B. Dirksen and A. van der Kooij;

1058

Michael A. Lyons

3 How Earlier Arguments Have Been Received 3.1 “Ezekiel reports the words of the exiles, so it is the source for H.” Wellhausenʼs interaction with the shared locutions themselves was limited to a single example. He argued that the locution “rot in iniquity” (‫ )מקק בעון‬in Lev 26:39 was dependent on Ezek 33:10 because the latter depicts a similar locution as a quote spoken by the exiles.12 This argument was quickly taken up by Smend and Kuenen.13 Wellhausenʼs claim that he is simply taking Ezekielʼs statement at face value14 is somewhat ironic, since this is precisely what he is not doing with respect to Lev 26. The difficulty here is that it seems equally plausible to regard material in Ezekiel and in Lev 26 as a creative literary product. The attribution of speech in Ezek 33:10 cannot naively be taken as verbatim reporting any more than the attributions of speech in, for example, Ezek 8:12; 9:9 can. It is not impossible, then, that Ezekiel is borrowing a locution from H, modifying it (from “rot in iniquity” to “rot in transgressions and sins”), and putting it into the mouth of his contemporaries as part of his strategy to combat despair.15 What is particularly problematic about Wellhausenʼs treatment of this passage, however, is what he does not mention: namely, the fact that the locution “rot in iniquity” occurs earlier in Ezek 4:17, where it is not a quote “in the mouth of the people” but a statement of judgment employed by Ezekiel against the people.16 This occurrence could easily be taken as evidence against his assertion that Ezekiel is simply reporting what he hears and must therefore be the source of the locution.

Leiden: Brill, 1993), 113–127. Compare A. Kuenen, Historisch-Kritisch Onderzoek naar het onstaan en de verzameling van de Boeken des Ouden Verbonds (Leiden: P. Engels, 1861), 104, 132, 134, 136, 148, with the significantly different second edition of 1885, translated as idem , An Historico-Critical Inquiry into the Origin and Composition of the Hexateuch (trans. P. H. Wicksteed; London: Macmillan, 1886), 275–287, and note how K uenen , Religion of Israel (see n. 9), represents a transition between the two. 12  Wellhausen, Die Composition (see n. 4), 169, 172; cf. idem, Prolegomena (see n. 4), 384: “In the same way the phrase pine away in their iniquity is repeated by Ezekiel as he heard it in the mouth of the people. He is its originator in literature; in Lev. xxvi. it is borrowed.” 13  See R. Smend, Der Prophet Ezechiel (Kurzgefasstes exegetisches Handbuch; 2nd ed.; Leipzig: S. Hirzel, 1880), 268; Kuenen, Historico-Critical Inquiry (see n. 11), 287. 14  W ellhausen , Prolegomena (see n. 4), 384, n. 1: “Delitzsch [. . .] thinks it a piece of impertinence in me to read out of Ezek. xxxiii. what that passage says.” 15  See W. Zimmerli, Ezekiel 2: A Commentary on the Book of the Prophet Ezekiel Chapters 25–48 (trans. J. D. Martin; Hermeneia; Philadelphia: Fortress, 1983), 187: “Thus Israelʼs despair finds expression in words of lamentation (stylized by the prophet himself).” 16  The locution also appears as a negative statement in Ezek 24:23. This occurrence was noted by Wellhausen, Die Composition (see n. 4), 169; idem, Prolegomena (see n. 4), 381, but he did not use it to argue for the direction of dependence.

How Have We Changed?

1059

It is telling that modern critics, regardless of their position on the direction of literary dependence, no longer use Wellhausenʼs argument. For example, Moshe Greenberg (who argues for the priority of H) takes Ezek 4:17 as the prophetʼs use of earlier stereotypical language and Ezek 33:10 as the peopleʼs acknowledgement of the prophetʼs earlier statements.17 And Klaus Grünwaldt (who argues for the priority of Ezekiel) takes Lev 26:39 as a postexilic interpretation of the exile along Deuteronomistic lines, one that borrows an Ezekielian locution to make its point.18 3.2 “A prophet would not depend on earlier texts, so H is the borrowing text.” In the 1800s, there was a widespread belief that prophecy was characterized by great originality and that the prophets were radically independent figures who spoke solely out of their own personal genius.19 This perspective led to a good deal of anxiety and disagreement about how to account for influence and dependence in prophetic literature. Thus we find Klostermann arguing against Graf that Ezekiel could borrow extensively from earlier texts and yet do so “freely.”20 Similarly, we find Paton arguing against Cornillʼs and Baentschʼs claim that Ezekielʼs “originality” rules out the possibility that he borrowed from H.21 In an amusing twist, Paton simply denied Ezekielʼs originality (at least with respect to the issue of literary dependence): “In reality Ezekiel is perhaps the least original of all the Old Testament writers.”22 Thankfully, these debates are a 17  M. Greenberg, Ezekiel 1–20 (AB 22; New York: Doubleday, 1983), 108; idem, Ezekiel 21–37 (AB 22A; New York: Doubleday, 1997), 673. 18  K. Grünwaldt, Das Heiligkeitsgesetz Leviticus 17–26: Ursprüngliche Gestalt, Tradition und Theologie (BZAW 271; Berlin: de Gruyter, 1999), 361–362. 19  See, e.g., G. H. A. Ewald, Die Propheten des Alten Bundes (2nd ed.; 2 vols.; Göttingen: Vandenhoeck & Ruprecht, 1867), 1:1–47, 67–69; also Wellhausen, Prolegomena (see n. 4), 398–399. 20  K lostermann , “Beiträge” (see n. 5), 417, 431, esp. 444: “Man wird also vielmehr schließen müssen, das auch in diesem Punkte das Heiligkeitsgesetz nicht ein Produkt des Propheten, sondern ein Muster gewesen ist, durch dessen Sprache er sich hat bestimmen lassen, ohne seine eigene dabei aufzugeben, d.i. eben in der Weise freier Reminiscenz.” Cf. Graf, Die geschichtlichen Bücher (see n. 7), 82. 21  C. H. C ornill , Einleitung in das Alte Testament (Freiburg: J. C. B. Mohr, 1891), 78: “Dann kann aber nur eines vom andern abhängig sein; entweder H von Ez oder Ez von H. Letztere Alternative ist schon a priori ausgeschlossen. Es ist so unwahrscheinlich wie möglich, dass ein Schriftsteller von der Originalität und dem Umfange Ezs seinen Styl gerade an Einem Capital des Pentateuchs gebildet haben sollte”; Baentsch, Das Heiligkeits-Gesetz (see n. 6), 124: “Wäre Lev. XXVI älter als Ezechiel, so hätte Ezechiel seine ganze Phraseologie und sehr wesentliche Vorstellungen aus diesem Stücke entlehnt. Es bliebe für ihn nicht allzuviel Originales mehr übrig. Es ist aber undenkbar, dass ein Prophet sich in solch sklavische Abhängigkeit versetzt und seinen Sprach- und Ideenkreis aus einem einzigen Capitel, einer einzigen Rede, geschöpft haben sollte.” 22  Paton, “Holiness-Code” (see n. 5), 109.

1060

Michael A. Lyons

thing of the past.23 Older assumptions about “prophetic originality” have given way to the appreciation of prophetic literature as the product of creative literary interaction with earlier texts.24 3.3 “If the author of X possessed Y, why wouldnʼt he use it at all points?” As originally used, this argument had to do with elements of Israelite religion reflected in H that were absent in Ezekiel. For example, a number of scholars took the absence of legislation about the high priest and jubilee year in Ezekiel as evidence that these notions were unknown in his time.25 The assumption that if Ezekiel had possessed H he would have borrowed from it at all points fails to take into account the complexity of religion and the creative ways in which authors use earlier texts. This was the charge of L. B. Paton, who felt that the possibility of authorial innovation was being ruled out a priori and claimed that nothing about dating could be deduced from Ezekielʼs silences on certain laws.26 Moreover, whereas Wellhausen presented the development of Israelite religion recoverable from Ezekiel, H, and P as a smooth, linear sequence,27 others felt that the relationships between the laws contained in H and Ezek 40–48 were a good deal more complex.28

23  Though there are still debates about the degree to which Ezekiel the prophet was a legal innovator; see, e.g., M. Haran, “The Law-Code of Ezekiel XL–XLVIII and Its Relation to the Priestly School,” HUCA 50 (1979), 45–71; idem, “Ezekiel, P, and the Priestly School,” VT 58 (2008), 211–218; R. Levitt Kohn, A New Heart and a New Soul: Ezekiel, the Exile and the Torah (JSOTSup 358; Sheffield: Sheffield Academic, 2002), 110–118. 24  See, e.g., C. Levin, Die Verheißung des neuen Bundes: In ihrem theologiegeschichtlichem Zusammenhang ausgelegt (FRLANT 137; Göttingen: Vandenhoeck & Ruprecht, 1985), 67: “Pointiert kann man sagen: Die prophetische Literatur, ja mit Einschränkung das ganze Alte Testament, ist in erster Linie weder Autoren- noch Redaktorenliteratur, sondern Auslegungs­ literatur, ein großer, in Jahrhunderten gewachsener, schriftlicher Midrasch: ‘Sacra scriptura sui ipsius interpres.’ ” 25  E.g., Smend, Der Prophet Ezechiel (see n. 13), 315, 382–383; Kuenen, Historico-Critical Inquiry (see n. 11), 286; B aentsch , Das Heiligkeits-Gesetz (see n. 6), 113–114; C ornill , Einleitung (see n. 21), 78. Note that the same kind of argumentation had already been used by Kuenen as early as 1869–1870; see Kuenen, Religion of Israel (see n. 9), 2:114–116; 3:55–56; see also Kalisch, Leviticus (see n. 9), 166, 291. 26  Paton, “Holiness-Code” (see n. 5), 106–108. 27  W ellhausen , Prolegomena (see n. 4), 379: “Jehovist, Deuteronomy, Ezekiel, are a historical series; Ezekiel, Law of Holiness, Priestly Code, must also be taken as historical steps, and this in such a way as to explain at the same time the dependence of the Law of Holiness on the Jehovist and on Deuteronomy.” 28  See Kuenen, Historico-Critical Inquiry (see n. 11), 286: “the priority seems to belong now to Lev. xxvii. sqq., and now to Ezek. xl. sqq.”; Driver, Introduction (see n. 5), 141, n. 6.

How Have We Changed?

1061

The assumption that an author would use his source at all points is less common today.29 The question “Why wouldnʼt text X use all available material in text Y?” is useful for investigating differences in ideology only once the direction of literary dependence is already known – not for determining the direction of dependence. Of course, this is not to say that modern scholars do not make arguments about the relationship of H and Ezekiel with reference to developments in Israelite religion. For example, both Christoph Levin and Christophe Nihan have (in different ways) argued that the representation of covenant in Lev 26 reflects a conceptual development that is later than the representations of covenant in Ezekiel.30 And Jacob Milgrom – who held a different model of the relationship between H and Ezekiel – argued that Hʼs reference to a multiplicity of sanctuaries (Lev 26:31) reflects a preexilic date of composition.31 That we can come to different conclusions based on our reconstructions of the evolution of religious concepts and practices in ancient Israel highlights the methodological difficulties inherent in such reconstructions. But the alternative would be a kind of historical agnosticism, which is obviously problematic as well. 3.4 “If the author of X possessed Y, why would he change it?” This argument also arose from Wellhausenʼs model of the development of Israelite religion and was expressed in the context of his claim that cultic praxis was not “codified” before the exile.32 It seemed most likely to him that while the cult was functioning there was no need to codify its practice. Rather, he claimed, it was only after the loss of the cult that cultic praxis was codified, initially by Ezekiel and then by later Priestly authors. The exile itself was the motivation for its formal preservation. Given this reconstruction, Wellhausen claimed that it was impossible to explain the differences between pentateuchal legal material and Ezek 40–48 with the theory that Ezekiel had modified earlier codified legal material.33 If he possessed it, why would he change it? And conversely, if codi-

29  Regarding the issue of the high priest, see R. Kasher, “Anthropomorphism, Holiness and Cult: A New Look at Ezek 40–48,” ZAW 110 (1998), 192–208, here 197. 30  L evin , Die Verheißung (see n. 24), 222–227; C. N ihan , “The Priestly Covenant, Its Reinterpretations, and the Composition of ‘P,’ ” in The Strata of the Priestly Writings: Contemporary Debate and Future Directions (ed. S. Shectman and J. S. Baden; Zürich: Theo­ logischer Verlag, 2009), 87–134. 31  J. M ilgrom , Leviticus 17–22: A New Translation with Introduction and Commentary (AB 3B; New York: Doubleday, 2001), 1363. 32  Wellhausen, Prolegomena (see n. 4), 52–60. See already Kuenen, Religion of Israel (see n. 9), 2:153. 33  Wellhausen , Prolegomena (see n. 4), 60: “Ezekielʼs departure from the ritual of the Pentateuch cannot be explained as intentional alterations of the original; they are too casual and insignificant.”

1062

Michael A. Lyons

fied legal material already existed, why would Ezekiel have needed to compose his own legislation?34 These arguments were quickly taken up by other critics.35 The argument that the differences between H and Ezekiel cannot be explained as deliberate modifications of one by the other continues to be made. It has been voiced most recently by, for example, Georg Fohrer, Menahem Haran, and Steven Tuell.36 Unlike earlier critics, these authors all argue that neither Ezekiel nor H depend directly on each other, nor is there mutual literary dependence; rather, the differences between the two can be accounted for only by a model in which both depend on an earlier shared source. Though such a model is not in itself unlikely, some of the arguments used to support it seem dubious. For example, the inability to explain why an author would modify an earlier text probably reveals more about modern critics than ancient authors. The argument that “the author had no reason to change his source” is both overly optimistic (in its “all or nothing” approach to reconstructing motives behind differences) and overly pessimistic (about the possibility that other readers might discover something new). And the so-called problem of why Ezekiel would modify H (or vice versa) is not alleviated by positing a common source, but compounded: now we have two authors who have modified an earlier shared source, and we are no closer to explaining why these changes were made than when we were dealing with only one borrowing author. Moreover, the differences between H and Ezekiel (e.g., Lev 21:13–15 vs. Ezek 44:22) are not the kind that can be explained any better by a model in which both H and Ezekiel are modifying an earlier source. Finally, a number of recent works have plausibly explained at least some of the differences as deliberate modification.37 34  Wellhausen, Prolegomena (see n. 4), 60: “What need would there have been for it, if the realised picture, corresponding completely to his views, had actually existed, and, being already written in a book, wholly obviated any danger lest the cultus should become extinct through the mere fact of its temporary cessation?” 35  Baentsch, Das Heiligkeits-Gesetz (see n. 6), 113–114; Cornill, Einleitung (see n. 21), 78: “Wenn Ezechiel H kannte; wesshalb stellte er dann seine eigene Zukunftsthora auf?” See already Kuenen, Religion of Israel (see n. 9), 2:114–116. 36  G. F ohrer , Die Hauptprobleme des Buches Ezechiel (BZAW 72; Berlin: Töpelmann, 1952), 147: “Beide haben eine verschiedene Reihenfolge der Gesetze (vgl. S. 145f.); die Ähnlichkeit beschränkt sich auf diese Gesetze, schließt aber die ermahnenden Zusätze nicht ein. Eine Reihe bedeutsamer Unterschiede sind schon früh erkannt worden; sie sind stilistischer und sachlicher Art und schließen die Möglichkeit literarischer Abhängigkeit Ezechiels von H aus. Auch diese Unterschiede erklären sich zwanglos aus der Annahme einer von Ezechiel und H benutzten älteren Vorlage”; idem, Ezechiel (HAT; Tübingen: Mohr Siebeck, 1955), xxiii; M. Haran, “The Law-Code of Ezekiel XL–XLVIII and Its Relation to the Priestly School,” HUCA 50 (1979), 45–71, here 62; idem, “Ezekiel, P, and the Priestly School,” VT 58 (2008), 211–218, here 212; S. Tuell, Ezekiel (NIBCOT; Peabody, MA: Hendrickson, 2009), 3. 37  See, e.g., K. R. Stevenson , The Vision of Transformation: The Territorial Rhetoric of Ezekiel 40–48 (Atlanta: Scholars Press, 1996); Kasher, “Anthropomorphism, Holiness and Cult” (see n. 29), 192–208.

How Have We Changed?

1063

3.5 “X is a quoting text.” It has been suggested that the direction of literary dependence between two related texts can be determined on the basis of which author quotes other texts (though today we would speak more precisely of allusion rather than quotation). This claim was articulated by Paton in favor of Ezekiel as the borrowing author and was later repeated by Burrows.38 But this claim has also been made more recently by Eckart Otto in favor of H as the borrowing text.39 The ease with which this argument can be used by proponents of diametrically opposed models should make us cautious. In fact, when both texts are known to depend on multiple earlier sources, this argument seems to be of little value. 3.6 “X scatters/gathers and therefore it is borrowing from Y.” Early critics made arguments about the direction of literary dependence using the distribution of locutions as a criterion. For example, Paton thought it more likely that Ezek 18, 20, and 22 had gathered what was scattered in H than that H had scattered what appeared together in Ezekiel.40 Modern critics such as Baruch Levine have made the opposite argument in support of the priority of Ezekiel, arguing that Lev 26 is “replete with Ezekielʼs distinctive language” and that it “condenses or encapsulates diction that is virtually pervasive in Ezekiel.”41 Similarly, Christophe Nihan thinks it unlikely that Hʼs locutions would be scattered throughout Ezekiel.42 But both clustering and scattering are compositional techniques that can be utilized by borrowing authors and are therefore not rigorous criteria for determining the direction of dependence.43 38  Paton, “Holiness-Code” (see n. 5), 109; M. Burrows, The Literary Relations of Ezekiel (New York: Jewish Publication Society of America, 1925), xi. 39  E. Otto, “Innerbiblische Exegese im Heiligkeitsgesetz Levitikus 17–26,” in Levitikus als Buch (ed. H.-J. Fabry and H.-W. Jüngling; BBB 119; Berlin: Philo, 1999), 125–196, esp. 180–182; see also C. Nihan, From Priestly Torah to Pentateuch: A Study in the Composition of the Book of Leviticus (FAT 2/25; Tübingen: Mohr Siebeck, 2007), 543. 40  Paton , “Holiness-Code” (see n. 5), 110, 113–114; cf. 113: “Ez. xx is full also of the phrases of Lev. xxvi, and in all of these cases it is easier to suppose that Ez. in his exhortation has brought together things which were disconnected in H., than that H. has divided up and scattered in different portions of his code sets of phrases which are regularly combined in Ez.” 41  B. A. Levine, “The Epilogue to the Holiness Code: A Priestly Statement on the Destiny of Israel,” in Judaic Perspectives on Ancient Israel (ed. J. Neusner et al.; Philadelphia: Fortress, 1987), 9–34, here 29, 30. 42  N ihan , Priestly Torah (see n. 39), 543: “the literary relationship between Lev 26 and Ezekiel makes more sense if the author of H has the various oracles in Ezekiel before his eyes; the contrary assumption that the prophet (or his disciples) would have scattered their source throughout the book, as Milgrom, for instance, would have it, is difficult to admit.” 43  We see the clustering of locutions used as a technique in 4Q176 (which borrows and pulls together material, largely from Isaiah, on the basis of a shared keyword). In earlier texts, we have the examples of Ezek 38–39 (which is practically a pastiche of borrowed locutions), for

1064

Michael A. Lyons

3.7 “Ezekielʼs use of shared locutions presumes an existing body of legislation.” Both Driver and Paton argued for the priority of H on the grounds that Ezekielʼs accusations and exhortations presumed existing legislation.44 Baentsch claimed that this was true for what he reconstructed as the earliest layer of the Holiness Code (H1),45 and even Kuenen admitted the force of this argument – though he claimed that Ezekiel was using material behind H rather than H itself.46 Related to this argument is Patonʼs claim that it is more likely for a prophet to borrow from legal material than for a law code to borrow from prophecy.47 These arguments were subsequently repeated by Burrows, though they are rarely stated in this form today.48 It seems more common for modern authors to first argue for the priority of H on other grounds, then describe how Ezekiel is basing his arguments on H once it is presumed that he is using H.49 The claim that it is more likely for prophets to use legal material than the converse is plausible, but it is not conclusive. To be sure, is difficult, if not impossible, to explain the origin of individual legal regulations as dependence on prophetic statements of accusation and judgment. However, the formation of collections of legal material that function as theological literature may very well have been indebted to prophetic pronouncements that were themselves based on older legal traditions.50 With respect to Lev 26, one can imagine a setting in which the prophecies of Ezekiel have already been textualized and in which the resulting book is being read as Scripture. One can further imagine a redactor in this period who inserts a highly allusive editorial layer into the Pentateuch, referencing the book of Ezekiel (as well as Exodus, Deuteronomy, and Amos), which see W. A. Tooman, Gog of Magog: Reuse of Scripture and Compositional Technique in Ezekiel 38–39 (FAT 2/52; Tübingen: Mohr Siebeck, 2011), or Num 18 (generally regarded as a late collection of laws about priestly prebends), for which see C. Nihan, “The Priestly Laws of Numbers, the Holiness Legislation, and the Pentateuch,” in Torah and the Book of Numbers (ed. C. Frevel et al.; FAT 2/62; Tübingen: Mohr Siebeck, 2013), 109–137; R. A chenbach , Die Vollendung der Tora: Studien zur Redaktionsgeschichte des Numeribuches im Kontext von Hexateuch und Pentateuch (BZABR 3; Wiesbaden: Harrassowitz, 2003), 141–172. We see the scattering of locutions used as a technique in the book of Job, where material from a psalm collection that includes Pss 8; 39; 72–73; 107; 139 has been placed throughout the book. (Note how locutions from Ps 8:5 alone are employed in Job 7:17–18; 15:14; 25:6.) Similarly, the Apocalypse of John contains allusions to Ezekiel that have been scattered throughout the book. 44  Driver, Introduction (see n. 5), 138–139, 141; Paton, “Holiness-Code” (see n. 5), 114. 45  Baentsch, Das Heiligkeits-Gesetz (see n. 6), 69, 86. 46  Kuenen, Historico-Critical Inquiry (see n. 11), 287. 47  Paton, “Holiness-Code” (see n. 5), 114. 48  Burrows, Literary Relations (see n. 38), 35. 49  See M. A. Lyons, From Law to Prophecy: Ezekielʼs Use of the Holiness Code (LHBOTS 507; New York: T&T Clark, 2009), 59–67, 76–145. 50  See R. G. Kratz, The Composition of the Narrative Books of the Old Testament (trans. J. Bowden; New York: T&T Clark, 2005), 140–143; K. Schmid, The Old Testament: A Literary History (trans. L. Maloney; Minneapolis: Fortress, 2012), 96–99.

How Have We Changed?

1065

and all the while using the genre and language of earlier legal literature. Such an editorial layer would be not a functional law code but a kind of bold literary experiment in theological reflection and textual coordination. Something similar to this has been proposed by Eckart Otto.51 3.8 “Part (or all) of Lev 26 is exilic or later, and therefore it is the borrowing text.” Wellhausenʼs conclusion that H had borrowed from Ezekiel was based in part on his arguments that Lev 26 should be dated to the exilic or early postexilic period52 and that the author of Lev 26 was the compiler of Lev 17–25.53 For example, Wellhausen felt that the language of Lev 26:14–39 was so remarkably gloomy that no preexilic historical precedent could be found, and he concluded that this kind of language presumed the destruction and exile of 587.54 He also argued that the language of vv. 40–45 would be incongruous in a preexilic setting: Why issue threats to prevent disobedience (vv. 14–39), then undercut them with a reassurance that restoration was possible (vv. 40–45)?55 Was it not more likely that this restoration language presumed the imminent or recent end of the exile?56 Although a number of early critics shared Wellhausenʼs presupposition that an exilic date for Lev 26 was incompatible with a model in which Ezekiel borrowed from H,57 others accepted Wellhausenʼs arguments about dating only for a handful of verses in Lev 26 that they felt to be exilic interpolations.  See Otto, “Innerbiblische Exegese” (see n. 39).  W ellhausen , Prolegomena (see n. 4), 382: “the exilic or post-exilic origin of this hortatory and denunciatory oration is too plain to be mistaken”; cf. idem , Die Composition (see n. 4), 170–171. 53  Wellhausen, Prolegomena (see n. 4), 380: “If Lev. xxvi. is incontestably intended to form the conclusion of chaps. xvii.–xxv., it would be natural to suppose that the author of that collection was also the author of the oration.” 54  Wellhausen, Prolegomena (see n. 4), 383–384: “These words cannot have been written before the Babylonian exile. [. . .] In Ezekielʼs day such thoughts, feelings, and expressions as we have here can be shown to have prevailed: but it would be difficult to show that the fall of Samaria gave rise to such depression at Jerusalem”; cf. Kalisch, Leviticus (see n. 9), 361–362. 55  Wellhausen, Prolegomena (see n. 4), 384: “In such prophets as Jeremiah and Ezekiel there is a meaning in such forecasting of the joyful future, but here it contradicts both the historical position and the object of the threats, and appears to be explained most naturally as the result of an accident, i.e., of actuality.” This argument was repeated by Baentsch, Das Heiligkeits-Gesetz (see n. 6), 126–128. 56  Wellhausen, Prolegomena (see n. 4), 384: “I even think it certain that the writer lived either towards the end of the Babylonian exile or after it, since at the close of the oration he turns his eyes to the restoration.” 57  Wellhausen, Prolegomena (see n. 4), 383–384; cf. Smend, Der Prophet Ezechiel (see n. 13), xvii: “Es kann für gewiss gelten, dass das letztere Corpus jünger ist als Ez, zumal Lev 26 offenbar im babylonischen Exil geschrieben ist”; Kuenen, Historico-Critical Inquiry (see n. 11), 287: “And again, the date inferred from Lev. xxvi. 3–45 confirms the priority 51

52

1066

Michael A. Lyons

Thus, by stratifying Lev 26, August Dillmann was able to maintain that Ezekiel was borrowing from a largely preexilic Holiness Code.58 Other commentators interpreted the data differently: whereas Wellhausen felt that the depressing language of Lev 26 was unprecedented, Dillmann simply asserted the opposite, and Driver attributed it to the “certainty of approaching exile.”59 Even Wellhausenʼs argument that the language of restoration in Lev 26:40–45 would undercut the rhetorical force of vv. 14–39 in a preexilic setting was not conclusive for all. It is at least conceivable that an author could genuinely wish to discourage apostasy or respond to a perceived problem of apostasy by describing horrific consequences (note that they are depicted in Lev 26 as remedial, designed to drive the offender back to YHWH) and yet for theological reasons be unwilling to imagine that YHWH would abandon his people. Such a view seems to lie behind the response of Driver.60 There are, then, two distinct issues: the question of which parts of H presume the exile and the question of the direction of literary dependence. These issues can be related in several different ways, and practically every possible position is attested in modern scholarship. For example, the argument that Lev 26 in its current form is exilic lies behind Fohrerʼs model of a common earlier source shared by H and Ezekiel.61 And the argument that parts of Lev 26 are later interpolations lies both behind models that view H as postexilic and behind models that view the bulk of H as preexilic (and the source for Ezekiel). Some of these models reconstruct a unidirectional relationship between H and Ezekiel,62 whereas others specify a relationship of mutual literary dependence in which the interpolations are believed to have been motivated by the book of Ezekiel. Thus some take Lev 26:(39)40–45 to be an exilic insertion because they feel that the hopeful outlook of these verses would be incongruous in the preexilic period of Ezekiel; for the author of this discourse has a longer exile behind him than Ezekiel has”; Baentsch, Das Heiligkeits-Gesetz (see n. 6), 126–127. 58  Though Dillmann agreed with Wellhausen that Lev 26:40–45 would be incongruous in a preexilic setting after the warnings in vv. 14–39, he took only Lev 26:34–35, 39, 40–45 to be exilic; see A. Dillmann, Die Bücher Exodus und Leviticus (Kurzgefasstes exegetisches Handbuch; Leipzig: S. Hirzel, 1880), 620; idem, Die Bücher Numeri, Deuteronomium und Josua (Kurzgefasstes exegetisches Handbuch; 2nd ed.; Leipzig: S. Hirzel, 1886), 645–646. Wellhausen dismissed Dillmannʼs model as a “mechanical style of criticism”; see Wellhausen, Prolegomena (see n. 4), 380 n. 1. 59  Dillmann, Numeri (see n. 58), 645; Driver, Introduction (see n. 5), 143. 60  Driver, Introduction (see n. 5), 143. 61  See F ohrer , Die Hauptprobleme (see n. 36), 146–148; idem , Introduction to the Old Testament (trans. D. E. Green; Nashville: Abingdon, 1968), 141–142. A similar dating for Lev 26 can be seen in the slightly different model (mutual dependence) proposed by R. Kilian, Literarkritische und formgeschichtliche Untersuchung des Heiligkeitsgesetzes (Bonn: Peter Hanstein, 1963). 62  It should be noted that the verses that are thought to be interpolations in Lev 26 share far fewer locutions with Ezekiel than does the surrounding material.

How Have We Changed?

1067

after the threats of vv. 14–(38)39.63 Others take Lev 26:40–45 to be an exilic insertion because their description of hope for a remnant seems to be at odds with vv. 13–39, which speak of a complete and total destruction for those who rebel.64 Still others take Lev 26:33b–35, 43–44 as additions, arguing that v. 33b is a repetition of vv. 31–32, that vv. 34–35 interrupt the flow by introducing a new topic, and that the sudden change in verb form in v. 44 is evidence of a retrospective point of view from a postexilic standpoint.65 Walther Zimmerliʼs commentary on Ezekiel reflects an even more complicated model of mutual literary dependence.66 It is clear that there is considerable disagreement over how Lev 26 is to be stratified.67 63  So H. L. Ginsberg, The Israelian Heritage of Judaism (New York: Jewish Theological Seminary of America, 1982), 80–81, 105. Ginsberg sees two layers of interpolated material, consisting of vv. 39–40a, 44–45 and vv. 33b–37, 40b–43; similarly Levine, “Epilogue to the Holiness Code” (see n. 41) 19. (Levine sees both the exilic base layer and the first additions to it as dependent on Ezekiel; see pp. 24–30). The argument about the incongruity of vv. 40–45 has also been made by R. Müller , “A Prophetic View of the Exile in the Holiness Code: Literary Growth and Tradition History in Leviticus 26,” in The Concept of Exile in Ancient Israel and Its Historical Contexts (ed. E. Ben Zvi and C. Levin; BZAW 404; Berlin: de Gruyter, 2010), 207–228, here 208, 221; Baentsch, Das Heiligkeits-Gesetz (see n. 6), 126–127; Dillmann, Numeri (see n. 58), 646; Wellhausen, Prolegomena (see n. 4), 384 (though the models of these authors are hardly identical). 64  See the chapter by Kopilovitz in this volume, pp. 1041–1054, and D. R om -S hiloni , Exclusive Inclusivity: Identity Conflicts between the Exiles and the People Who Remained (6th–5th Centuries BCE) (LHBOTS 543; New York: Bloomsbury T&T Clark, 2013), 2, 173– 178, esp. 178, n. 125. However, if the references to “scattering” and “perishing” in vv. 14–39 are derived from traditional treaty language, perhaps they should be taken as stereotypical rather than as absolute claims. 65  J. M ilgrom , Leviticus 23–27: A New Translation with Introduction and Commentary (AB 3B; New York: Doubleday, 2001), 2322–2323, 2337, 2364. Note the similarities to the arguments of Dillmann, Exodus und Leviticus (see n. 58), 620. 66  W. Z immerli , Ezekiel 1: A Commentary on the Book of the Prophet Ezekiel Chapters 1–24 (trans. R. E. Clements; Hermeneia; Philadelphia: Fortress, 1979), 52, argues that “it cannot be denied that Ezekiel has been influenced by detailed material built into H, or which already underlies it.” For example, he sees material in Lev 26 as the source for Ezek 34:25–30; cf. idem, Ezekiel 2 (see n. 15), 220, 222. Yet he also claims that “it can be seen most clearly in parts of Leviticus 26 that the prophecy of Ezekiel has exercised a reciprocal influence on the development of H” (Ezekiel 1, 52); he suggests as an example the influence of prophetic preaching against the high places. He therefore wishes to “abandon the sharp alternatives” of the unidirectional dependence of Ezekiel on H or H on Ezekiel (as well as Fohrerʼs model of a common source behind both!), and he believes that Elligerʼs reconstruction of an autumn festival liturgy behind both H and Ezekiel “may be appropriate to explain a series of connections.” But it is far from clear why the similarities and differences Zimmerli lists (Ezekiel 1, 50–51) necessitate such a model and equally unclear what criteria he is using to determine the direction of dependence in any given place. 67  See, e.g., Nihan, Priestly Torah (see n. 39), 537 nn. 561–562; compare Müller, “Prophetic View” (see n. 63), 207–228.

1068

Michael A. Lyons

An exilic dating for part or even all of Lev 26 is not necessarily determinative for the question of the literary relationship between H and Ezekiel. After all, since the book of Ezekiel is a late exilic composition,68 it could be either the source text or the borrowing text for exilic portions of Lev 26. Furthermore, since it is generally recognized that the book of Ezekiel has received editorial extensions (Fortschreibungen) – particularly in Ezek 34, where the shared locutions with H are especially numerous – it cannot be assumed that an exilic date for material in Lev 26 rules out borrowing by the redactors of the book of Ezekiel.69 Even attempts to locate the entirety of H as a composition in the pre- or postexilic period by reconstructing its setting based on social factors – something at which we are perhaps more sophisticated than the critics of the 1800s – do not yield objective criteria by which we may determine the direction of dependence in every instance.70 Such reconstructions are of course necessary and demonstrate our commitment to the historical dimensions of the text under consideration. But the ease with which we can create plausible reconstructed settings in a variety of temporal periods should inspire caution.71 Ultimately, the direction of literary dependence must be ascertained from an explanation of the shared locutions in their respective contexts. 3.9 “X expands/modifies Y.” Klostermann, Driver, and Paton all agreed that expansion was one way in which the direction of literary dependence could be detected. For example, Klostermann claimed that Ezekiel had a technique of borrowing material, then adding to it or modifying it: the prophet borrowed the locution “break the staff of bread” (Lev 26:26), then added to it a statement about “eating and drinking in dismay” (Ezek 4:16); he borrowed “rot in iniquity” (Lev 26:39), then added to it a statement about “being appalled” (Ezek 4:17; cf. Ezek 33:10); he borrowed “scatter among the nations” (Lev 26:33), then changed it to “scatter to every wind” (Ezek  On the dating of the earliest form of the book of Ezekiel, see R. Albertz, Israel in Exile: The History and Literature of the Sixth Century B.C.E. (trans. D. Green; Atlanta: Society of Biblical Literature, 2003), 351–352. 69  See A. K lein , Schriftauslegung im Ezechielbuch: Redaktionsgeschichtliche Unter­ su­chungen zu Ez 34–39 (BZAW 391; Berlin: de Gruyter, 2008), 185–189; M. A. Lyons , “Extension and Allusion: The Composition of Ezekiel 34,” in Ezekiel: Current Debates and Future Directions (ed. W. A. Tooman; Tübingen: Mohr Siebeck, forthcoming). 70  For a preexilic setting, see J. J oosten , People and Land in the Holiness Code: An Exegetical Study of the Ideational Framework of the Law in Leviticus 17–26 (VTSup 67; Leiden: Brill, 1996); for a postexilic setting, see G rünwaldt , Das Heiligkeitsgesetz (see n. 18), esp. 379–381. 71  See the comments of B. D. Sommer, “Dating Pentateuchal Texts and the Perils of Pseudo-Historicism,” in The Pentateuch: International Perspectives on Current Research (ed. T. B. Dozeman et al.; FAT 78; Tübingen: Mohr Siebeck, 2011), 85–108. 68

How Have We Changed?

1069

5:10, 12; 12:14; 17:21). That the change in the latter example was introduced by Ezekiel was clear from the fact that it was motivated by the performance of his sign act, in which he is described as being told to scatter a third of his shaved hair to the wind (Ezek 5:2).72 Likewise, Ezekiel borrowed “rule with harshness” (Lev 25:43), then glossed the rare word “harshness” (‫ )פרך‬with the word “strength” (‫ )חזקה‬in Ezek 34:4.73 But this raises a question: though in some cases expansion can be plausibly explained as exegetical transformation, can we equate longer with expanded (and therefore with borrowed) in every case? Driver agreed that the direction of dependence could be detected when Ezekiel expanded Hʼs material,74 but he went further than Klostermann in two ways. First, Driver distinguished between shared locutions that were evidence of literary dependence and ones that were not, such as “technical expressions, borrowed (as seems clear) from priestly terminology, but not sufficient to prove Ez.ʼs acquaintance with the codified laws in the form in which we now have them,” and purely coincidental similarities (“expressions [. . . that] appear to arise out of the narrative in which they occur, and are not necessarily reminiscences”).75 Second, Driver argued that the earlier of two related texts could be identified by greater originality and coherence: “Lev. 26,3 ff. is in style terse and forcible; Ez. is diffuse: Lev. also appears to have the advantage in originality of expression (contrast, e.g., ‘the pride of your power’ in Lev. 26,19 and in Ez. 7,24 (LXX). 24,21. 30,6. 18. 33,28), and in the connexion of thought (contrast Lev. 26,4–6. 13 with Ez. 34,25–29).”76 Recent studies continue to treat expansion as evidence for determining literary dependence and include other kinds of textual modification (e.g., conflation, reversal) as well. For example, Klaus Grünwaldt states that the borrowing text is the one in which an intent to revise can be detected.77 In his study of the Holiness Code, Grünwaldt compares Lev 26 with Ezek 34 and finds significant differences that point to the likelihood of modification: the two texts contain the same elements, yet in Lev 26 they are conditional, whereas in Ezek 34 they are unconditional; the two texts display identical locutions in different order (e.g.,  Klostermann, “Beiträge” (see n. 5), 417–418.  K lostermann , “Beiträge” (see n. 5), 430: “aber dass Ezechiel hier alle Ausdrücke entlehnt, zeigt er deutlich durch Vorsetzung des erklärenden, nur ihm eigenen Ausdruckes ‫בחזקה‬.” 74  D river , Introduction (see n. 5), 141, 142, n. 2; so also Paton , “Holiness-Code” (see n. 5), 111–113. 75  Driver, Introduction (see n. 5), 139, 139, n. 2. Note also his skepticism regarding Klostermannʼs arguments about expansion of formulaic language; cf. Klostermann, “Beiträge” (see n. 5), 436–43; Driver, Introduction (see n. 5), 142, n. 2. 76  Driver, Introduction (see n. 5), 142, n. 2. 77  G rünwaldt , Das Heiligkeitsgesetz (see n. 18), 50: “Um einer Lösung näher zu kommen, müßte man nachweisen können, daß entweder Lev 26,4–6.13 oder Ez 34,25–30 sicher redaktionelle Tendenzen enthält, von woher sich eine Bearbeitungsintention aufzeigen ließe.” 72

73

1070

Michael A. Lyons

Lev 26:4 vs. Ezek 34:27); the freedom from the yoke of slavery is presented as having occurred in the past in Lev 26:13 but as a promise for the future in Ezek 34:27. Grünwaldt then claims that Lev 26 has “combined” what is separate in Ezek 34, has “reversed” the order of elements in Ezek 34, and has “replaced” Ezekielʼs uncommon expression with a more familiar one; he finally states that Lev 26 is most likely borrowing from Ezek 34.78 Reinhard Müller has made a similar claim: “The first section of promises (vv. 4–6) is modelled on a promise of Israelʼs salvation after the exile in Ezek 34:25–28. What Yahweh promises there without any conditions is introduced in Lev 26 only as the outcome of Israelʼs obedience to Yahwehʼs statutes and commandments. This difference shows that Lev 26 quotes Ezek 34, not vice versa.”79 The problem is that even when textual differences suggest that modification has occurred, the mere fact of difference does not in itself indicate the direction of literary dependence. This problem is particularly acute in the case of Ezek 34, for here we find that other scholars, working under the very same principle articulated by Grünwaldt, have interpreted the textual data as an instance of Ezekiel borrowing from Lev 26. In their analyses, it is Ezekiel and his redactors who have expanded and reversed locutions and turned conditional blessings into unconditional promises.80 The definitions of modification and expansion, then, need further clarification for these to be considered useful criteria. A number of scholars have undertaken this task – ultimately, an attempt to define more precisely Driverʼs terms originality of expression and connexion of thought. Millar Burrows was one of the first to methodically consider criteria for the direction of dependence, though his position on the dating of Ezekiel clearly influenced his results.81 In her study of the relation of Ezekiel to Priestly and Deuteronomic texts, Risa Levitt Kohn aimed at isolating and describing the literary profile of Ezekielʼs text-handling  Grünwaldt, Das Heiligkeitsgesetz (see n. 18), 350, 365–366; cf. 351: “Der knappe Vergleich zwischen Lev 26,4–6.13 mit der Parallele in Ez 34,25–30 hat ergeben, daß der Anhang zur Hirtenallegorie wahrscheinlich die Vorlage eines Teils der bedingten Segensverheißung ist.” For Grünwaldtʼs analysis of the relationship between Lev 26 and Ezek 5–6, see Das Heiligkeitsgesetz, 359–361. 79  M üller , “Prophetic View” (see n. 63), 209. See also C. N ihan , “The Holiness Code between D and P: Some Comments on the Function and Significance of Leviticus 17–26 in the Composition of the Torah,” in Das Deuteronomium zwischen Pentateuch und deutero­ nomistischem Geschichtswerk (ed. E. Otto and R. Achenbach; FRLANT 206; Göttingen: Vandenhoeck & Ruprecht, 2004), 81–122, here 108–110; idem , Priestly Torah (see n. 39), 544–545. 80  Zimmerli, Ezekiel 2 (see n. 15), 220; Milgrom, Leviticus 23–27 (see n. 65), 2348–2352; Klein, Schriftauslegung (see n. 69), 185–189, 346–347. 81  Burrows, Literary Relations (see n. 38), ix–15. Burrowsʼs maximalist position on the bookʼs literary sources was probably due to the influence of his Doktorvater C. C. Torreyʼs dating of the book of Ezekiel to the Hellenistic period. 78

How Have We Changed?

1071

techniques.82 And, in a 2009 monograph, I proposed criteria for identifying the direction of dependence that included ideologically based modification (where there is an identifiable motivation in one text but not the other), incongruity, conceptual dependence, and interpretive expansion.83 But even these criteria do not always yield unambiguous results. For example, Christophe Nihan has argued that the abstract use of the word ‫משכן‬ in Lev 26:11 creates a “tension” in context (since, according to P and to Hʼs narrative frame, the tabernacle has already been constructed).84 On the one hand, it might be possible to explain this apparent incongruity by postulating that H is borrowing from and reacting to the depiction of covenant and presence in Ezek 37:26–27. On the other hand, Anja Klein has offered a different analysis, arguing that the use of the word ‫ משכן‬for YHWHʼs habitation in Ezek 37:26–28 is incongruous since it is used elsewhere in Ezekiel only for human habitation. Furthermore, it is employed in immediate proximity to the word ‫מקדש‬, Ezekielʼs usual term (used thirty-one times) for YHWHʼs habitation. Klein concludes that Ezek 37 is borrowing the locution “I will put my dwelling in your midst” (‫ונתתי‬ ‫ )משכני בתוככם‬from Lev 26:11 and reworking it into two statements: “I will put my sanctuary in their midst forever” (‫ )ונתתי את מקדשי בתוכם לעולם‬in Ezek 37:26 and “my dwelling will be with them” (‫ )והיה משכני עליהם‬in Ezek 37:27. She also notes that this material is juxtaposed in v. 27 with a borrowing of the covenant formula from the same context (Lev 26:12).85 So even the criterion of incongruity does not easily translate into evidence for the direction of literary dependence. Another example: Reinhard Müller concludes from the “unparalleled use” of the term ‫ פגרי גלוליכם‬in Lev 26:30 that H is borrowing this locution from Ezek 6:5 with the intention of “depicting the idols themselves as dead persons.”86 But, again, the reverse can also be argued. The oracle in Ezek 6:2–7, addressed to the inanimate “mountains of Israel,” contains many locutions also present in Lev 26:25a, 30–31. Yet Ezekiel uses the expressions “your bones” (v. 5b) and “your  Levitt Kohn, New Heart (see n. 23), 75–85.  Lyons, From Law to Prophecy (see n. 49), 59–67; for modification with polemical intent, see Lev 26:39 → Ezek 24:23 (cf. Ezek 18:2–3); incongruity, Lev 26:4 → Ezek 34:26 (note the shift in Ezekiel from singular to plural under the influence of Lev 26); conceptual dependence, Lev 25:25–28 → Ezek 7:12b–13a (what is being bought and sold?); interpretive expansion, Lev 25:43 → Ezek 34:4. 84  See Nihan, Priestly Torah (see n. 39), 538, n. 563; Nihan explains this as “part of a deliberate literary and rhetorical strategy emphasizing the conditional character of the covenant and of the divine presence.” It seems possible that this tension could also be accounted for as a lapse betraying the gap between the composerʼs setting and the ostensible narrative setting (cf. the anachronistic slip in Lev 18:25, 27–28). 85  Klein, Schriftauslegung (see n. 69), 186, 188. 86  Müller, “Prophetic View” (see n. 63), 217; see also the chapter by Kopilovitz in this volume, pp. 1041–1054. 82 83

1072

Michael A. Lyons

dwelling places” (v. 6), expressions that presuppose the human addressees of Lev 26. The incongruity between the inanimate addressee of the oracle and the locutions borrowed from H was noted and addressed by the scribal insertion of an explicit human referent from Lev 26:30 into Ezek 6:5a (in MT), “and I will place the corpses of the sons of Israel before their idols.”87 The problem is not that our criteria are flawed. Rather, it is that the data we are working with are ambiguous: most of the shared locutions in H and Ezekiel simply do not provide sufficient evidence to conclusively determine the direction of dependence. As Nihan has noted, “the dependence of Lev 26 on Jeremiah and Ezekiel cannot be consistently demonstrated for each of these parallels taken individually.”88 So although there are models (such as my own earlier work)89 that begin by making a plausible argument about the direction of dependence from a few examples, then attempt to incorporate other passages under this rubric in order to produce a coherent reading of all the data, such models have inherent limitations. The possibility of mutual literary dependence cannot be ruled out.

4 Concluding Reflections In what ways does modern scholarship on H and Ezekiel represent an advance over earlier scholarship, and what new trends can be detected? First, we no longer make arguments about the direction of literary dependence based on older assumptions about the development of Israelite religion. This is clearly visible in the move away from the literary-religious trajectory laid out by Wellhausen, a shift described in Christophe Nihanʼs insightful essay “The Holiness Code between D and P.” H is now widely viewed as a redactional adjustment of P rather than a pre-Priestly independent composition subsequently added to P – and this position is held by scholars on both sides of the debate about the direction of dependence between H and Ezekiel.90 Second, models of mutual literary dependence seem to be more common today than in the past.91 This may be due to the fact that such models have re Lyons, From Law to Prophecy (see n. 49), 62–63.  Nihan, Priestly Torah (see n. 39), 544. 89  Lyons, From Law to Prophecy (see n. 49), 61–67; see also Milgrom, Leviticus 23–27 (see n. 65), 2348–2363. 90  Nihan, “Holiness Code” (see n. 79), 81–122; also idem, Priestly Torah (see n. 39), 9–10. Cf. I. Knohl, The Sanctuary of Silence (Minneapolis: Fortress, 1995), 11–23; J. Milgrom, “HR in Leviticus and elsewhere in the Torah,” in The Book of Leviticus: Composition and Reception (ed. R. Rendtorff and R. A. Kugler; VTSup 93; Leiden: Brill, 2003), 24–40. 91  H. Sun, “Holiness Code,” ABD 3:254–257, here 256, takes this model as “the consensus of recent scholarship”; see also P. M. J oyce , Ezekiel: A Commentary (LHBOTS 482; New York: T&T Clark, 2009), 36; R. E. Clements, “The Ezekiel Tradition: Prophecy in a Time of 87 88

How Have We Changed?

1073

ceived broad exposure (e.g., in Zimmerliʼs influential commentary on Ezekiel). But it is probably also due to the fact that we have become increasingly aware that the relationship between H and Ezekiel continued as these two texts were transmitted into the Second Temple period. The manuscript evidence shows that scribes were aware of the existing literary relationships between H and Ezekiel, a fact that motivated them to create further relationships in both directions.92 This renders plausible the existence of complex, bidirectional textual interaction at the compositional and redactional stages.93 Still, further methodologically rigorous work is needed to demonstrate the exact points at which this may be occurring. Third, current studies reflect a greater appreciation for the mechanics and rhetorical function of literary allusion. We have developed more precise categories for analyzing text-handling techniques. Older assumptions about “prophetic originality” have been completely abandoned, and we are no longer reluctant to speak of prophets borrowing from earlier texts. In fact, literary allusion (often studied under the rubrics of innerbiblical interpretation or Schriftauslegung) is now seen as a sign of great creativity.94 This new emphasis on the study of literary borrowing and transformation also represents a significant move beyond older form- and tradition-critical approaches.95 Finally, regardless of their position on the direction of literary dependence, recent scholars of H and Ezekiel have paid much more attention to the function of borrowed material in its new context. This stands in contrast to older studies, many of which were content to merely list parallels between the two texts. The result is that our view of Ezekiel and H has been significantly complicated, since both must now be seen as texts that interact with earlier (and to some extent the same) legal material in complex ways.96 We are now in a position to investigate Crisis,” in Israelʼs Prophetic Tradition: Essays in Honour of Peter Ackroyd (ed. R. J. Coggins et al.; Cambridge: Cambridge University Press, 1982), 119–136, here 130. 92  Note the plusses in, e.g., MT Ezek 6:5a (← Lev 26:30); 36:11; 37:26 (← Lev 26:9); see also LXX Lev 19:26 (← Ezek 18:6; 22:9). 93  See the chapter by Nihan in this volume, pp. 1015–1039, which on this point offers a welcome corrective to my earlier work. 94  In many cases, this is due to the influence of the following groundbreaking works: M. F ishbane , Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985); B. M. Levinson, Deuteronomy and the Hermeneutics of Legal Innovation (New York: Oxford University Press, 1997); B. D. Sommer, A Prophet Reads Scripture: Allusion in Isaiah 40–66 (Stanford: Stanford University Press, 1998). 95  See Nihan, Priestly Torah (see n. 39), 6–7. 96  See N ihan , Priestly Torah (see n. 39), 616, who notes that H “was intended as a supplement to P, combining it with the systematic reception of other codes in the Pentateuch, the Decalogue, the ‘Covenant Code’ (CC), and the Deuteronomic code (D). These codes are not replaced, but rather harmonized, supplemented, and even revised in H. This means that H was actually conceived from the beginning in the prospect of a synoptic reading of the so-called ‘legal’ codes comprised within the Torah/Pentateuch.” For yet another perspective on Hʼs

1074

Michael A. Lyons

the similarities and differences in the exegetical techniques employed by both Ezekiel and H. Though the three models of literary dependence developed in the late 1800s are still in use today, and though some of the arguments used to support these models are also still being employed, recent scholarship on the relationship between H and Ezekiel has breathed new life into older positions. A careful consideration of past and present research shows how far we have progressed in explaining the nature of textual use in ancient Israel, as well as the social realities to which this textual use bears witness. But it is also instructive for examining the criteria we employ when creating models for explaining evident but ambiguous textual relationships, the implicit or explicit assumptions we bring to the study of texts, the varying ways in which we use notions of plausibility or likelihood, and our appeals to larger frameworks (and the extent to which they are helpful and/or necessary). Such consideration is obviously relevant not only for the study of the relationship between Ezekiel and H but also for explaining the formation, shape, and contents of other texts.

revision of earlier legal material, see J. Stackert, Rewriting the Torah: Literary Revision in Deuteronomy and the Holiness Legislation (FAT 52; Tübingen: Mohr Siebeck, 2007), esp. 209–225.

Ezekielʼs Prophetic Message in Light of Leviticus 26 Tova Ganzel and Risa Levitt Kohn When two biblical texts share identical terms and phrases, how is the direction of influence determined? Biblical scholars have been puzzling over just this issue for decades, especially when examining the writings of the prophet Ezekiel and the body of text within the Pentateuch known as the Holiness Code (H). In their respective contributions to the present volume, Christophe Nihan and Ariel Kopilovitz grapple with the presence of common terms and phrases found in these two sources and with the issue of how best to determine the direction of influence. This exercise is not new. Graf and Wellhausen were among the first to tackle the thematic and linguistic affinities between Ezekiel and the so-called Holiness Code.1 As early as 1866, Graf concluded that Ezekiel penned H.2 Wellhausen was more confounded. If Ezekiel had indeed composed H, how could the numerous differences between them be explained? Ezekiel never mentions Aaron or his sons, wine is not part of Ezekielʼs sacrifices, and the prophetʼs descriptions of festival observances show marked differences from those of H. Thus H must have been a corpus that, in his words, “inclines from Ezekiel towards the Priestly Code, but which stands, in terms of its content, closer to Ezekiel.” Later, he contended, H was revised to represent the views of P.3 Soon after Wellhausen made his argument, a flurry of scholarly research focused attention on Ezekiel and H, and the compilation of long lists of shared terms and expressions began. Some saw the legislation in H as more developed and complex than Ezekielʼs: Ezekiel does not mention the office of high priest; H does, and therefore Hʼs priesthood is more evolved than Ezekielʼs.4 Others noted that  K. H. Graf, Die geschichtlichen Bücher des Alten Testaments (Leipzig: Weigel, 1866), 81–83; J. Wellhausen, Prolegomena zur Geschichte Israels (5th ed.; Berlin: Reimer, 1899), 386–387. 2  Graf, Die Geschichtlichen Bücher (see n. 1), 81. 3  Wellhausen, Prolegomena (see n. 1), 386–387. 4  For lists, see T. N öldeke , Untersuchungen zur Kritik des Alten Testaments (Kiel: Schwers, 1869), 60; A. Klostermann, “Beiträge zur Entstehungsgeschichte des Pentateuchs,” Zeitschrift für die gesammte lutherische Theologie und Kirche 38 (1877), 417; R. S mend , Der Prophet Ezechiel (Leipzig: Hirzel, 1880), xxiv; and summaries in L. B. Paton , “The Holiness Code and Ezekiel,” Presbyterian and Reformed Review (January 1896), 98–115, and W. Zimmerli, Ezekiel 1 (trans. R. Clements; Hermeneia; Philadelphia: Fortress, 1979), 46. 1

1076

Tova Ganzel and Risa Levitt Kohn

the two sources exhibit different editorial formulae and hortatory passages. How could the author of H pen laws on sacrifice and other matters that could not possibly be obeyed in Ezekielʼs day and then append a list of terrible denunciations in case of disobedience? Is it not more probable that Ezekiel saw the moral and religious precepts of H as a useful manual of sorts in his own exilic context and, as a result, quoted from it frequently? That was the state of the field in 1896. The work of Nihan and Kopilovitz illustrates that there remains considerable debate as to whether the Holiness Code served as a source for Ezekiel or whether H was in fact influenced by the Ezekiel traditions themselves. Nihan underscores the cumulative weight of the linguistic parallels alongside their appearance in short textual segments, often in the same sequence, as an indicator that one text was clearly informed by the other. Determining the direction of influence, or borrowing, remains a separate issue for him. Here Nihan relies upon the use of ancient extrabiblical witnesses and “expansionism,” which he suggests can indicate a late addition to a text. He proposes, on this basis among others, that Lev 26 borrowed the concept of ‫ ברית‬from Ezekiel (37:26–27).5 In contrast, Kopilovitz uses Ezekiel to uncover the strata in Lev 26. His findings suggest that the direction of influence moves from Lev 26 to Ezekiel and that the linguistic affinities between Lev 26 and Ezekiel (especially chapter 33) reflect Ezekielʼs familiarity “with some written version of Lev 26” but without vv. 40–45. Kopilovitz further suggests that had Ezekiel been familiar with these verses, whose language he argues is not reflected in Ezek 33, he would certainly have used them in his restoration prophecies.6 As these contemporary discussions illustrate, establishing the priority of one text over another remains methodologically difficult. The collection and analysis of similar words, phrases, and expressions continues, but we are also acutely aware – perhaps to a greater degree than those investigating the issue in the late 1800s – that all biblical sources have undergone a complex process of composition, compilation, editing, and copying. As a result, they are linguistically stratified and difficult to untangle. We would like to suggest a different starting point. We now know and appreciate that before embarking on any consideration of possible influence and its direction, it is imperative to fully examine the individual contexts of all shared terms and expressions and, particularly in the case of the prophetic texts, to attempt to uncover the social matrix in which the prophet operated. In other words, there is literary, rhetorical, and theological evidence that can help elucidate the relationship between the Torah and the Prophets in general and between Ezekiel and the Holiness Code in particular. Rather than extending the long tradition of listing common phrases and making assumptions about who knew what first, we should choose instead to begin by noting the unique and, 5 6

 See his article, pp. 1015–1039, in the present volume.  See his article, pp. 1041–1054, in the present volume.

Ezekielʼs Prophetic Message in Light of Leviticus 26

1077

more important, documentable setting of the book of Ezekiel. It is impossible to examine the use of particular language in the book without reading it within the conceptual and contextual framework shaping the entire message and content of the book, namely, the reality of the Babylonian exile.7 We should be wary of imposing pentateuchal concepts on Ezekielʼs parables or metaphors without first considering their appearance in the context of the book of Ezekiel itself. Scholars have long recognized the role of the Israelite prophets as social activists and legal commentators. It seems clear that within their own social settings prophets served as mediators between biblical law and the actions of their own contemporary audience, with the prophetic function being one of lobbying for and upholding a standard of behavior and, often simultaneously, admonishing Israel for its failure to follow suit. Though most biblical prophets advocate for biblical law, it is less clear whether their ideal law is reflected directly in the pentateuchal sources or is more loosely derived from the biblical witness or their own interpretations of the biblical witness.8 Regardless, the prophets know the law and, as Y. Kaufmann noted long ago, they refer to it often.9 However, as Kaufmann also notes, the pentateuchal laws never refer to the prophets. An examination of Ezekielʼs use and interpretation of biblical law illustrates the way in which authoritative biblical texts are reinterpreted in the face of new historical circumstances, “when divine words had apparently gone unfulfilled as originally proclaimed (as in various promises and prophecies); or when new moral spiritual meanings were applied to texts which had lost their vitality.”10 As Moshe Greenberg notes, in Ezekiel “there is almost always a divergence large enough to raise the question, whether the prophet has purposely skewed the traditional material, or merely represents a version of it different from extant records.”11 7  For further discussion on this topic see, among others, T. Ganzel and S. E. Holtz, “Ezekielʼs Temple in Babylonian Context,” VT 64 (2014), 211–226; R. Levitt Kohn, “ ‘As Though You Yourself Came out of Egypt’: The Ethos of Exile in Ezekiel,” HBAI 3 (2014), 175–193; D. Vanderhooft, “Ezekiel in and on Babylon,” in “Bible et Proche-Orient: Mélanges André Lemaire III,” ed. J.-M. Durand and J. Elayi, special issue, Transeuphratène 46 (2014), 99–119; A. Winitzer , “Assyriology and Jewish Studies in Tel Aviv: Ezekiel among the Babylonian Literati,” in Encounters by the Rivers of Babylon: Scholarly Conversations between Jews, Iranians, and Babylonians in Antiquity (ed. U. Gabbay and S. Secunda; TSAJ; Tübingen: Mohr Siebeck), 163–216. Recently, Die Welt des Orients 45, no. 1 (2015) devoted the entire volume to this topic. 8  M. Greenberg, “Biblical Attitudes toward Power: Ideal and Reality in Law and Prophets,” in Religion and Law: Biblical-Judaic and Islamic Perspectives (ed. E. Firmage et al.; Winona Lake, IN: Eisenbrauns, 1990), 101–112. 9  Y. Kaufmann, The Religion of Israel (8 vols.; Tel Aviv: Mossad Bialik Dvir, 1937–1956), 3:564 (Hebrew). 10  M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985), 14. 11  M. G reenberg , “Notes of the Influence of Tradition on Ezekiel,” JANESCU 22 (1993), 29.

1078

Tova Ganzel and Risa Levitt Kohn

The book of Ezekiel is representative of Israelite theology at a crossroads, between preexilic and postexilic Israel. As a prophet of the exile, and as a prophet in exile, Ezekielʼs writings can help illuminate the way in which Israelite law and religious thought were understood at a time when following such precepts and practices was put to the ultimate test in terms of theology, sociology, and geography. Scholars are just beginning to comprehend the impact that the trauma of exile must have had upon Ezekielʼs language, theology, and visions.12 The rather recent trend of applying trauma and refugee studies to Ezekiel has not only helped contextualize the long-perceived eccentricity of the prophetʼs rhetoric and visions but also provided a useful lens through which to view Ezekielʼs stance vis-à-vis Israelite law and tradition. What can we learn through an examination of select biblical law as quoted in Ezekiel? How does this affect our understanding of H in Ezekiel (or Ezekiel in H)? What can we learn about the relationship of the law and the prophets? Ezekiel, in dealing with the trauma of exile and responding to what he perceived to be the apparent failure of Israelite covenantal promises and legal protections, seeks to reshape an older legal tradition in order to propose a new interpretation for his disillusioned and displaced audience. It is, contextually speaking, very difficult to suggest that the reverse might be the case: that is to say, that legislation was borne out of prophetic pronouncement. Certainly there is little precedent for this in other prophetic texts. Some of the ways in which the prophet uses unique literary and linguistic techniques in his reformulation have been noted elsewhere.13 In what follows, we would like to reexamine the prophecy of Jerusalemʼs doom in Ezek 5:5–17 to help illustrate this point. This unit divides into two: first the indictment of Jerusalem and then the sentencing of the city. Its opening verses (vv. 5–6) cite the main cause of the destruction, namely, Jerusalemʼs behavior, which is worse than that of all of the surrounding cities. She has rebelled against God and has broken his laws and statutes (‫)חקים ומשפטים‬. In the second unit (vv. 7–17), we learn that this behavior, namely, Jerusalemʼs wickedness (v. 7), has elicited a harsh divine response.14 The resulting punishment will take place in the presence of the 12  See, e.g., N. B owen , Ezekiel (AOTC; Nashville: Abingdon, 2010); D. S mith Christopher, A Biblical Theology of Exile (OBT; Minneapolis: Fortress, 2002); D. Garber Jr., “Traumatizing Ezekiel, the Exilic Prophet,” in Psychology and the Bible: A New Way to Read the Scriptures, from Genesis to Apocalyptic Vision (ed. J. Ellens and W. Rollins; 4 vols.; Westport: Praeger, 2004), 2:215–235. 13  See further discussions of reversals, prohibitions, and exodus/restoration language in R. Levitt Kohn, A New Heart and a New Soul: Ezekiel, the Exile and the Torah (JSOTSup 358; London: Sheffield, 2002), 76–81; T. G anzel , “The Transformation of Pentateuchal Descriptions of Idolatry,” in Transforming Visions: Transformations of Text, Tradition, and Theology in Ezekiel (ed. W. A. Tooman and M. A. Lyons; Princeton Theological Monograph Series 127; Eugene, OR: Pickwick, 2009), 33–49. 14  See M. Greenberg, Ezekiel 1–20: A New Translation with Introduction and Commentary

Ezekielʼs Prophetic Message in Light of Leviticus 26

1079

nations and is voiced by God in v. 9: “Doing to you what I have never done, and the likes of which I shall not do again because of all your abominations.” This verse connects the cityʼs deeds – all of her abominations – to an as yet unprecedented divine response, unique in its severity as compared to the most severe punishments found in the Torah. A survey of the terminology used in these verses sheds interesting light on the relationship between the prophet and the law and helps to demonstrate the claims laid out at the outset of this paper – namely, that we must go beyond detailed examination of linguistic affinities and pay greater attention to the broader contextual-conceptual framework within Ezekiel if we wish to better appreciate the significance of the linguistic affinities. Ezekiel 5:5–17 fuse elements found in divine threats of the Priestly literature and of the book of Deuteronomy, but they also exhibit terminology that does not appear in either source or elsewhere in the Torah. Moreover, Ezekiel chooses not to use certain terms with which he was clearly familiar, deciding instead to recast these phrases with his own unique emphasis.15 The phrases cited below illustrate the influences on Ezekielʼs use of language and how his reading of earlier Torah sources shaped the formulation of his own unique postexilic theology. These phrases are distinctive to Ezekiel and are not found with reference to the Israelites in any other biblical books. They include ‫אל‬/‫“( הנני על‬I am coming at [you]”);16 ‫“( עשה שפטים‬execute judgments”);17 (AB 22; New York: Doubleday, 1983), 119; D. Block, The Book of Ezekiel: Chapters 1–24 (NICOT; Grand Rapids, MI: Eerdmans, 1997), 196–217. 15  Levitt Kohn, New Heart (see n. 13). For the link between Ezekiel especially and the Holiness Code, see M. A. Lyons, From Law to Prophecy: Ezekielʼs Use of the Holiness Code (LHBOTS 507; New York: T&T Clark, 2009), esp. 117–122. 16  Greenberg, Ezekiel 1–20 (see n. 14), 113: this expression, which originally expresses divine anger at the nations (Nah 2:14), is directed against the Israelites in Ezekiel. 17  In Ezekiel, this phrase is directed to both the Israelites and the nations. As an address to the Israelites, it appears in Ezek 5:10, 15; 11:9; 14:5. With reference to the nations, it appears in Ezek 23:10; 25:11 (Moab); 28:22 (Sidon); 28:26 (the nations); 30:14 (No); 30:19 (Egypt). The use of ‫ שפטים‬is unique in an oracle that links the divine response to the acts of the people: the divine judgments respond to the peopleʼs judgments in vv. 6–7; Greenberg, Ezekiel 1–20 (see n. 14), 113. In the Pentateuch, this phrase appears with reference to Egypt alone; see, e.g., Exod 6:6; 7:4; 12:12; Num 33:4, and Levitt Kohn, New Heart (see n. 13), 43–44. As Block, Book of Ezekiel (see n. 14), 39 notes, “Ezekiel draws on the exodus narratives where Godʼs acts of judgment and salvation will lead both the Israelites and the Pharaoh and the Egyptians to recognize and know that he is the Lord.” This is its meaning in P (Exod 6:6; 7:4; 12:12; Num 33:4 [‫)]באלהיהם‬. ‫ שפטים‬appears nowhere outside P with the exception of 2 Chr 24:10, 24 and Prov 19:29. Put more precisely, the inherent message of this phrase is not restricted to a fate shared by the Israelites and the nations but lies in the use of terminology that, in the Pentateuch, is directed only toward Egypt but in Ezekiel is applied to Israel as well. This use of pentateuchal terminology heightens the harshness of the divine criticism of the Israelites. On Ezekielʼs use of the book of Exodus, see R. G. S. Idestrom, “Echoes of the Book of Exodus in Ezekiel,” JSOT 33 (2009), 489–510.

1080

Tova Ganzel and Risa Levitt Kohn

‫“( זרה לכל רוח‬scatter to every wind”);18 ‫לא חמל‬/‫“( לא חוס‬not sparing/not having pity”);19 ‫שלח אף‬/‫“( כלה‬spend anger and fury”);20 ‫שפך חמה‬/‫כלה‬/‫נוח‬ (“assuage/spend/pour out fury”) 21; ‫“( נתן לחרפה‬turn into a reproach”) 22; 18  This expression appears in Ezek 5:10, 12; 12:14, in contrast to the phrase ‫זרה בגוים‬, which is found in Ezekiel and in Lev 26:33. Other than these three occurrences with reference to the Israelites, ‫ אזרם‬is found in Jer 15:6 (‫)ואזרם במזרה בשערי הארץ‬, where it explains the parable in 5:2. ‫ זרה‬also appears in Isa 41:16, where it refers to the enemies of Israel (‫;)תזרם ורוח תשאם‬ in Jer 49:32 (‫)וזרתים לכל רוח קצוצי פאה‬, where it refers to Kedar and the kingdom of Hazor; and in Jer 49:36 (‫)וזרתים לכל הרחות האלה‬, where it refers to Elam. Here too, the uniqueness of Ezekielʼs use of this phrase is that it is directed to Israel, whereas Jeremiah and Isaiah direct it to the nations. 19  Ezekielʼs use of the phrase ‫( לא חוס‬Ezek 5:11; 24:14) is noted by B. J. S chwartz , “Ezekielʼs Dim View of Israelʼs Restoration,” in The Book of Ezekiel: Theological and Anthropological Perspectives (ed. M. S. Odell and J. T. Strong; Atlanta: Society of Biblical Literature, 2000), 53. Ezek 20:17 is another example in which God spares his people: ‫ותחס‬ ‫עיני עליהם משחתם‬. Here, however, this expression describes Godʼs past attitude toward the exodus generation in the wilderness and not his current attitude toward the generation of the destruction. The combination ‫ חוס עין‬appears in Deuteronomy, where it describes the Israelites. Only in Ezekiel does it describe the divine viewpoint; see L evitt K ohn , New Heart (see n. 13), 91. But it has five additional appearances in Ezekiel alongside the phrase ‫( לא חמל‬Ezek 7:4, 9; 8:18; 9:5, 10). The message of Ezek 5:11 is heightened by the repetition of this phrase six times in a limited range of chapters (5–9). Ezekiel does not repeat this expression after the description of the divine glory departing the temple in chapter 9. ‫ חוס‬and ‫ חמל‬do not appear individually in P and are paired only in Deut 13:9 with reference to someone who incites others to idolatry: ‫] ולא תחמל‬. . .[ ‫ולא תחוס‬. In addition, the phrase ‫ לא תחוס‬appears in Deut 7:16 with reference to the Canaanites, in Deut 19:13 with reference to a murderer, in Deut 19:21 in reference to a false witness, and in Deut 25:12 with reference to a woman who grabs male genitals; see M. Weinfeld, Deuteronomy and the Deuteronomic School (Oxford: Clarendon, 1972), 2. Outside of Ezekiel, the first-person description of God as not sparing and not showing pity appears only in Jer 13:14 and in a different context. As opposed to the harshness of the phrase in Ezekiel, the context in Jeremiah calls for repentance by the people and divine regret for the events of the destruction: ‫] ואם‬. . .[ ‫] תנו ליהוה אלהיכם כבוד בטרם יחשך‬. . .[ ‫שמעו והאזינו‬ ‫לא תשמעוה במסתרים תבכה נפשי‬, “Hear and heed [. . .] Give glory to Yahweh your God before he brings darkness [. . .] But if listen you will not, in secret my soul will weep” (Jer 13:15–17). 20  This expression appears in Ezek 5:13; 7:8; 20:8, 21; 43:8. The phrase ‫ באף ובחמה‬is found in Deuteronomy, the Deuteronomistic literature, and Ezekiel (to indicate wrath and great anger) but not in P; see Levitt Kohn, New Heart (see n. 13), 92. The use of ‫ כלה אף‬and ‫ שלח אף‬is specific to Ezekiel; see Block, Book of Ezekiel (see n. 14), 211, n. 77. Upon the execution of divine fury, its intended objectives will have been achieved. As Block notes, “Like Yahwehʼs mighty acts of deliverance centuries earlier, his acts of judgment on a rebellious people are intentionally designed to bring them to an acknowledgement of his presence, character, and claims on their lives.” 21  These phrases are found in Ezek 5:13; 6:12; 7:8; 9:8; 13:13; 14:19; 16:42; 20:8, 13, 21, 33–34; 21:22; 22:22; 24:13; 36:18. Like the phrases ‫שלח אף‬/‫כלה‬, these expressions and their frequency in Ezekiel convey the force of divine anger toward Israel. 22  These expressions appear in Ezek 5:14; 22:4; 36:15; 36:30. As G reenberg , Ezekiel 1–20 (see n. 14), 116, notes regarding Ezek 5:14: “These expressions recall Lev 26:31 and

Ezekielʼs Prophetic Message in Light of Leviticus 26

1081

‫נתן רעב‬/‫חרף‬/‫יסף‬/‫“( שלח‬send/add/reproach/inflict famine”).23 Ezekiel addresses the peopleʼs violation of the laws using these distinctive expressions. In addition, by describing the Israelitesʼ actions in much harsher terms than those found in the Priestly literature or in Deuteronomy, the prophet highlights and accentuates the seriousness of their actions. This prophetic unit also includes phrases found in the Priestly literature and Deuteronomy: ‫שבר‬ ‫מטה לחם‬, ‫ריק חרב‬, ‫בוא חרב‬, ‫נתן ארץ לשממה‬, ‫חיה רעה‬, ‫( גלה ערוה‬found in the Priestly literature [P/H] in the Pentateuch); ‫יתר‬, ‫בגוים‬/‫פוץ בארצות‬, ‫אבד‬, ‫כעס‬ (found in Deuteronomy); and ‫בגוים‬/‫זרה בארצות‬, ‫קנא‬, ‫כרת‬, ‫( דבר‬found in both P and in Deuteronomy as well as Ezekiel).24 Thus Ezekiel not only uses unique formulations but also draws upon terms from the ‫“( תוכחה‬warning”) verses in the Torah. It is our contention that Ezekiel references P/H laws in order to illustrate the way in which these laws have been disregarded by the contemporary inhabitants of Jerusalem and by his own community of exiles. Ezekiel evaluates Israelʼs successes and failures against the standard of the Torah in general and P/H in particular. As Michael Lyons also notes, Ezekiel treats Hʼs laws as authoritative instruction, using them to explain his own contemporary situation and in so doing transforms Hʼs laws into prophetic accusations and turns Hʼs conditional covenant punishment into descriptions of pending devastation.25 So too, conditional covenant blessings become guarantees of future restoration. Just as the doom prophecies of Ezek 1–24 draw on the curses of the covenant documents (and particularly Lev 26), the restoration visions in the latter part of the book turn the blessings promised for obeying the covenant laws into unconditional prophecies of future harmony. To imagine the reverse would be to suggest that Hʼs legal precepts were devised on the basis of specific infractions noted by the prophet and then formulated into law. Leviticus 26 and Ezekiel share passages and locutions apart from the terminology found in the Pentateuch and in Ezek 5:5–17. We briefly note some of the connections, focusing on Lev 26:40–45:

a Jeremianic commonplace. [. . .] The amazement of every passerby is mentioned in I Kings 9:8; Jer 18:16, 19:8. But the ‘combination in the sight of [l eʿene] every passerby’ is unique to Ezekiel (again at 36:34) and reflects his peculiar sensibility to the public, international humiliation inflicted on Israel.” 23  The threat of famine is widespread in the pentateuchal punishments. But the specific expressions ‫ חצי רעב‬and ‫ יסף רעב‬are found only in Ezekiel; cf. Deut 32:23 and the discussion in Greenberg, Ezekiel 1–20 (see n. 14), 116. On the connection between this phrase, Deuteronomy, and Lev 26:26, see Block, Book of Ezekiel (see n. 14), 213–214. 24  Likewise, the term for scattering (‫ )זרה‬appears in the curses of Lev 26:33, and Ezekielʼs reference to a remnant (‫ ;שאר‬Ezek 5:10) recalls Lev 26:36, 39. 25  Lyons, From Law to Prophecy (see n. 15), esp. chap. 4.

1082

Tova Ganzel and Risa Levitt Kohn

1. Lev 26:40 states: ‫ ;במעלם אשר מעלו בי‬in Ezek 17:20, ‫ מעלו אשר מעל בי‬is cited as the reason for Zedekiahʼs punishment for breaking the covenant with Babylonia. 2. Lev 26:41: ‫ ;לב ערל‬in Ezek 44:7, 9, ‫ ערל לב‬describes the foreigners who desecrated the temple. 3. Lev 26:42: ‫ ;וזכרתי את בריתי‬this phrase appears in the Prophets only in Ezekiel 16:60 (‫)וזכרתי אני את בריתי אותך‬, with the connotation of keeping an existing covenant. 4. Lev 26:43: ‫ ;יען וביען‬this emphatic expression for “because” appears in Scripture only in Lev 26 and Ezek 13:10, 36:3. 5. Lev 26:43: ‫ ;במשפטי מאסו‬this expression appears otherwise only in Ezek 5:6; 20:13, 16 (a similar expression appears in 11:12, 20; 20:19, 21, 24). 6. Lev 26:45: ‫ ;לעיני הגויים‬this expression is frequent in Ezekiel (5:8; 20:9, 14, 22, 41; 22:16; 28:25; 38:23; 39:27). Given that these phrases from Lev 26 appear elsewhere in Ezekiel, we cannot rule out the possibility that Ezekiel was familiar with some version of Leviticus, perhaps even in written form, and that it included vv. 40–45. Conceptually, Ezekiel emphasizes two things. First, the people did the exact opposite of what was commanded in Leviticus: “You did not walk according to my statutes and you did not do my judgments” (Ezek 5:7b). This is a variation of the positive and negative protases of the covenant blessings and curses in Lev 26:3: “If you walk according to my statutes and you keep my commandments and do them.” The curses are to befall the people “if you despise my statutes and disdain my judgments so as not to do all of my commandments” (Lev 26:15). The second thing Ezekiel emphasizes is that the results will be even harsher than the outcome in Lev 26. Ezekielʼs implications are grounded in terms of the Priestly covenant. The chastisement, which Yahweh tells Ezekiel is due, includes the outcome that “fathers will eat sons in your midst, and sons will eat their fathers, and I shall do judgments on you, and I shall scatter your whole remnant to all the winds” (Ezek 5:10). This reprimand, which echoes the covenant curse of Lev 26:29, is even harsher than the pentateuchal warning: “You shall eat the flesh of your sons, and the flesh of your daughters you shall eat” (see Deut 38:53–57; Jer 19:9; Isa 9:18–20;26 Lam 4:10; Zech 11:9). In Ezekiel, not only will parents eat the flesh of their children, but children will eat their parents, a prediction that is unmatched and without precedent in the Bible.27 This new spin on an older text must be understood in the larger context of the unusual depiction of God in this oracular unit, especially Ezek 5:9, where the fact that these actions are without precedent is specifically noted: “Doing to you what I have  Note that it is possible to read ‫ זַ ְרעו‬and not ‫זְ ר ֹעו‬.  See Radak on Ezek 5:10; for Radak, see M. Cohen (ed.), Mikra‍ʾot Gedolot ‘Haketer’: Ezekiel (Ramat Gan: Bar-Ilan University, 2000), 29. 26

27

Ezekielʼs Prophetic Message in Light of Leviticus 26

1083

never done, and the likes of which I shall not do again because of all your abominations.” Taken as a whole, these verses represent a new pinnacle in the description of a god of wrath, found only in Ezekiel and not in pentateuchal sources.28 Lyons also points to the differences between terminology in Leviticus and Ezekiel and suggests that Ezekiel presents an alternative remedy to that of the final verses of Lev 26.29 Thus, in determining the nature of the affinities between texts, differences, like similarities, must also be considered. It seems likely that Ezekiel based his prophetic message regarding Israelʼs restoration on Lev 26ʼs understanding of Israelʼs everlasting ‫ ברית‬with God.30 Ezekiel 37:26 reflects the prophetʼs interpretation of the ‫ ברית‬in Leviticus: “I will make a covenant of friendship with them – it shall be an everlasting covenant with them – I will establish them and multiply them, and I will place my sanctuary among them forever.” Given the strong conceptual similarity of the ‫ברית‬ as everlasting in both works, it is difficult to posit that Ezekiel was not familiar with Lev 26. As Jeffrey Stackert notes, “Leviticus 26 [. . .] does not perceive violation of ‫ ברית‬as the end of the relationship between God and Israel or an event requiring the establishment of a new ‫ברית‬.”31 This contextually grounded conclusion constitutes an even stronger connection to Ezekielʼs conception of ‫ ברית‬than simple examination of similar terms. Just as the pentateuchal texts treating ‫ ברית‬reflect a basic view of Israelʼs history and religion, consideration of Ezekiel evinces the prophetʼs view of Israelʼs history and religion. Note Ezekielʼs singular interpretation of Leviticus. There is an important difference between the verses in Leviticus, which open with a confession and the surrender of hearts, and Ezekielʼs restoration prophecies, which do not include regret or 28  Ezekiel does mention the covenant between God and his people in other prophecies, but Ezek 5:5–17 does not include this element. Although the absence of the covenant here could be explained, as A. Kopilovitz proposes in his article in this volume (see pp. 1041–1054), by suggesting that Ezekiel did not see the final verses of Lev 26, it seems that terminological relationships could lead to different conclusions. This observation regarding divine wrath can also augment Schwartzʼs discussion of Ezekielʼs dim view of the restoration of Israel (Schwartz, “Dim View” [see n. 19], 43–67). In particular, the harshness of the divine response as manifested in Ezekielʼs “new language” underscores the irreversible, one-time reality of the destruction of the temple. Alongside the terms and phrases that are unique to Ezekiel in this oracular unit, treated here, Ezekiel also shares terms with other biblical books. Thus the description of the famine (Ezek 5:16–17) resembles Lev 26:26 and Deut 32:23–24; see also R. Kasher, Ezekiel: Introduction and Commentary, Vol. 1: Chapters 1–24 (Mikra Leyisra‍ʾel; Tel Aviv: Am Oved, 2002), 209–212 (Hebrew). 29  Lyons, From Law to Prophecy (see n. 15), 85–88, 122–123. 30  See J. Milgrom, Leviticus 23–27: A New Translation with Introduction and Commentary (AB 3B; New York: Doubleday, 2000), 2348–2352; Block, Book of Ezekiel (see n. 14); and J. S tackert , “Distinguishing Innerbiblical Exegesis from Pentateuchal Redaction: Lev 26 as a Test Case,” in The Pentateuch; International Perspectives on Current Research (ed. T. B. Dozeman et al.; Tübingen: Mohr Siebeck, 2011), 369–386. 31  Stackert, “Distinguishing Innerbiblical Exegesis” (see n. 30), 384.

1084

Tova Ganzel and Risa Levitt Kohn

confession (just memory) and in which the change of action is attributed only to God (36:22, 25–27). We are just beginning to comprehend the impact that the trauma of exile must have had on Ezekielʼs language, theology, and visions. Ezekiel operated at a time when the interrelationship of oneʼs social identity, theological ideology, geographical location, and political reality demanded a complete reinterpretation of history and tradition. As a prophet of exile, Ezekiel was forced to reconcile Israelʼs past with its historical present. This interest in recounting, remembering, and appealing to past laws, traditions, and stories is a crucial step in the rebuilding and reestablishment of this newly exiled community. Ezekiel and his contemporaries, by virtue of their dislocation from Israel, had to rethink traditional notions of kinship in order to remap their newly formed communities. Indeed, they were forced to reconsider all traditional notions of what made Israel a people. As C. deGroot van Houten aptly notes, “There is no community without tradition as there is no tradition without community. [. . .] The community is shaped by the story it tells.”32 Part of the process of dealing with the reality of exile involves reshaping the traditions that define a community. Ezekielʼs contemporaries in exile are those Israelites who have yet to reach the promised land. Babylon is the wilderness where the presence of God appears without any of its traditional accouterments (tabernacle and ark) and is mediated by a prophet. The time of wandering precedes entry into the promised land, which is currently occupied by a sinful population who must be wiped out before the land can be rightfully inhabited. To return to an important observation noted earlier, it was Y. Kaufmann who first observed that the law (that is, the Torah) never refers to the prophets.33 This observation is key to the way in which we understand and approach the plethora of terms and expressions found in the Priestly source, the Holiness Code, and the Deuteronomic History. In addition, there is a continuously growing body of scholarly work that illustrates quite conclusively the way in which Ezekiel creatively reformulates Torah precepts in order to fit the context and needs of his contemporary audience living out their lives in the Babylonian diaspora.34 That said, the discussion of the relationship between H/P and Ezekiel must now, in our opinion, turn to a closer examination of the individual context in each source before considering textual mutuality, borrowing, or direction of influence. 32  C. de G root Van H outen , “Remember That You Were Aliens: A Traditio-Historical Study,” in Priest, Prophets and Scribes: Essays on the Formulation and Heritage of Second Temple Judaism in Honour of Joseph Blenkinsopp (ed. E. Ulrich et al.; JSOTSup 149; Sheffield: JSOT Press, 1992), 224–225. 33  See above, note 9. 34  See above, n. 11, and also the summaries in the works of Lyons, From Law to Prophecy (see n. 15); Levitt Kohn, “As Though You Yourself Came out of Egypt” (see n. 7), 175–193; and idem, “Ezekiel into the Twenty-First Century,” in Recent Research on the Major Prophets (ed. A. Hauser; Sheffield: Sheffield Phoenix, 2008), 260–278.

Part Ten Reading for Unity, Reading for Multiplicity Theological Implications of the Study of the Pentateuchʼs Composition

Introduction Benjamin D. Sommer How does the current debate among scholars of the Pentateuchʼs composition relate to other subfields within biblical studies? In a global academy marked by hyperspecialization and mutually incomprehensible discourses, this question is salutary. Specialists in the Pentateuch do well to ask whether a new theory, or even a new method, adds something meaningful to the field of biblical criticism as a whole. By relating data points to each other in new ways, does a new theory cause us to see some larger issue more clearly or for the first time, or has the rearrangement of these data points become an end in itself, which is to say, a dead end? The implications of the study of pentateuchal composition have been recognized as far-reaching since the middle of the nineteenth century. W. L. M. de Wette, Wilhelm Vatke, Karl Heinrich Graf, Abraham Kuenen, and above all Julius Wellhausen recognized that the close reading of the Pentateuchʼs constituent documents, coupled with attempts to date those documents, yielded nothing less than a picture of the history of the religion of Israel. Later scholars were motivated to take issue with Wellhausenʼs readings, or with his datings, or with his source analyses precisely because these disagreements altered their historiography of Israelite beliefs and practices. But it was not only history that the study of the Pentateuchʼs origins impacted decisively. Core assumptions about the nature of Scripture have been deeply undermined by the varied models that scholars of the Pentateuch developed to understand the composition and redaction of the Five Books of Moses. In short, compositional criticism subverted many religious thinkersʼ concept of Scripture (that is, their assumptions about what Scripture is and what it does, as well as how it should be interpreted and by whom). Frequently, religious thinkers responded to this subversion by rejecting the modern theories, whether through arguing against them or ignoring them. Fewer thinkers accepted the challenge of integrating new data into their understanding of how to read the Pentateuch as Scripture and their sense of what it means for a composite text to function as Scripture. Each of the essays from this session of the conference addresses some aspect of the theological-hermeneutic conundrum created by modern scholarship on the Pentateuch. At the core of this conundrum, whether one starts from a documentary, a supplementary, or a fragmentary model or from some combination of two or three of these, is the question of the Pentateuchʼs unity. Does the Pentateuch,

1088

Benjamin D. Sommer

in spite of the complexity of its composition history, nevertheless achieve a unity that allows it to be read as a whole? Or is a unity-oriented reading of the Pentateuch a hermeneutic chimera, so much striving after the wind? If the latter, is it necessarily the case that a nonunified Pentateuch cannot serve as Scripture? All the following essays wrestle with this set of questions, though in very different ways. Jean-Pierre Sonnet argues that the Pentateuch achieves a consistent closure that brings it together as a literary work. This complex but self-contained work is marked throughout by an emphasis on Israelʼs ultimately successful obedience to divine will as mediated not only by Moses but also by various spirit-filled co-mediators. This closure is dynamic, occurring not only at the end of the Pentateuch (for, Sonnet reminds us, narrative closure is not the same thing as a narrativeʼs ending) but throughout the work and in two different ways. A series of verses that repeat guiding vocabulary throughout the Pentateuch and especially at the beginnings and ends of books signal initially the hope for obedience and ultimately its presence. This theologically significant closure, going from expectation to fruition, is essentially narrative in nature. James Watts, in contrast, points out that we ought not assume that rhetorical unity, especially for a ritual and/or scriptural text, must be achieved on a narrative level. Indeed, he points out, the rich diversity of genres found in the Pentateuch suggests that narrative, although it may provide the framework of the Pentateuch, does not provide the persuasive power that allowed the Pentateuch to endure as a central text in many religious communities for millennia. Rather, the genres placed into that framework often exert their force over the surrounding narrative. The genres of law and list both promote and model the ritualization of the Pentateuch, a ritualization that grants the text a unity at the level of communal practice that does not depend on narrative coherence and thus is not threatened by its absence. In my own contribution, I too am concerned with the way the Pentateuch has been accepted as Scripture in a specific community, but I argue that the most common interpretive and lectionary practices of rabbinic Judaism put little emphasis on the narrative or even the thematic unity of the Five Books of Moses. Rather, in its fragmentary, verse-oriented mode of reading, classical midrash turns our attention toward new combinations of biblical verses that create coherent narratives based on snippets of Scripture brought together from afar. Modern source criticism, mutatis mutandis, does something similar: it draws several coherent narratives from the Pentateuch, each one self-consistent but often in tension or disagreement with the others. Such tensions or disagreements are a sacred norm in rabbinic literature. Thus composition-critical approaches allow modern Jews to embrace the Pentateuch as a religiously relevant anthology precisely by breaking down the illusion that the Pentateuch even tries to achieve narrative coherence. Markus Witte attends to these issues on a broad theoretical plane. He notes that a theology of the Pentateuch (or a reading of the Pentateuch as a theology)

Introduction

1089

requires several perspectives and approaches, both synchronic and diachronic – and the latter needs to encompass not only the prehistory of the Pentateuch as we know it but also its reception histories in various religious communities. Historical-critical understanding of the types encouraged by the present volume are necessary, but not sufficient, for an attempt to present a theology of the Pentateuch, whose narrative voice(s) and systematic implications must also be sought out.

Book or Anthology? The Pentateuch as Jewish Scripture Benjamin D. Sommer In this essay, I address some of the ways in which the current debate among specialists in the composition of the Pentateuch matters to biblical scholarship more broadly. Why should an exegete, a historian of Israelite religion, or a biblical theologian care whether the non-P, non-D material contains a more or less coherent block of material we can designate “J” or discrete traditions covering different eras that are not closely related to each other? Does it matter to our exegete, historian, or theologian whether the Pentateuch was put together by a compiler who worked diligently to preserve older texts largely untouched or by redactors and supplementers who treated their sources with great freedom? Is it important, from an exegetical point of view, that the Pentateuch has a composition history at all? In what follows, I suggest that these questions matter a great deal. I speak from the point of view of a theological interpreter; by referring to the Neo-Documentary Hypothesis associated with Menahem Haran, Baruch Schwartz, Joel Baden, and others, I hope to show how this model has important implications for biblical theology, which would disappear or at least be attenuated if we were to employ what David Carr calls the Sources-and-Supplements model for understanding the origins of the Pentateuch.1

Centripetal versus Centrifugal Reading I begin by contrasting two ways postbiblical interpreters read the Bible. We may characterize these two approaches to reading as centripetal and centrifugal. The centripetal mode strives for a center; it is holistic and seeks unity. The centrifugal mode flees from a center; it is atomistic and oriented toward multiplicity. Both forms of reading are known in Jewish biblical exegesis, but the centrifugal is especially prominent in rabbinic literature.  For this term, see D. M. C arr , “Data to Inform Ongoing Debates about the Formation of the Pentateuch: From Documented Cases of Transmission History to a Survey of Rabbinic Exegesis,” in this volume, pp. 87–106, esp. n. 39. 1

1092

Benjamin D. Sommer

Midrashic reading is more centrifugal than centripetal, in two respects. First, the unity or coherence of biblical books has traditionally not been important for rabbinic (or other ancient Jewish) approaches to Scripture. For the midrashic exegete, the next unit after the verse that matters is the Bible as a whole, or perhaps the section (Pentateuch or Neviʾim-Ketuvim), but certainly not the book. James Kugel explains: Midrash is an exegesis of biblical verses, not of books. [. . .] There simply is no boundary encountered beyond that of the verse until one comes to the borders of the canon itself – a situation analogous to certain political organizations in which there are no separate states, provinces, or the like but only the village and the Empire. [. . .] Each verse of the Bible is in principle as connected to its most distant fellow as to the one next door. [. . .] Indeed [the midrashist] sometimes delights in the remoter source.2

The rabbis believe that the main unit of expression in the Bible is the verse or a group of two or three successive verses. They are not interested in larger literary units, such as a whole poem or a complete story, much less a whole biblical book.3 In this respect, midrash is a fragmentary mode of reading: it examines words, verses, or small collections of verses independently of their context in a given biblical book. True, midrash is concerned in its own way with context (and in this respect midrash is centripetal as well as centrifugal), but the context into which a midrashist puts a given verse is that of the entire Bible. The rabbis are interested in the connections between a given verse and verses throughout the Bible. When discussing, say, Gen 22:1, midrashic interpreters were not particularly interested in the relationship of this verse to Gen 22:2 and 22:3 or even later verses from this chapter. More far-flung relationships interested them much more. Thus in Gen. Rab. § 55, the interpreters do not examine Gen 22:1 within what postmidrashic interpreters call the story of the binding of Isaac.4 Instead they overwhelmingly cite verses found far away: Pss 60:6; 11:5; and 110:4; Qoh 8:4; Deut 23:21; Lev 19:18; Mic 6:6; 2 Sam 7:18; and Prov 25:6 (to name but a few). Of the fifteen verses cited in the discussion of Gen 22:1 in Gen. Rab. § 55, only two are from Genesis, and even those (Gen 21:5 and 2  J. L. Kugel, “Two Introductions to Midrash,” in Midrash and Literature (ed. G. H. Hartman and S. Budick; New Haven, CT: Yale University Press, 1986), 77–103, here 93. See also S. F raade , “Moses and the Commandments: Can Hermeneutics, History, and Rhetoric Be Disentangled?,” in The Idea of Biblical Interpretation: Essays in Honor of James L. Kugel (ed. H. Najman and J. H. Newman; JSJSup 83; Leiden: Brill, 2004), 399–422, here 416. 3  See J. Barton, Oracles of God: Perception of Ancient Prophecy in Israel after the Exile (London: Darton, Longman and Todd, 1986), 150. 4  It is telling that (as my colleague Robert Harris has emphasized to me) the rabbis never use the term the Binding of Isaac to refer to a narrative unit. (The term appears only twice in midrashic literature, in Tanḥ. [Buber], Vayyera § 46 and ʾAg. Ber. § 38; in both it refers not to a particular narrative unit but to what happened to Isaac in v. 9.) Only in the Middle Ages, when Jewish interpreters began thinking about longer textual units, does this term appear in Jewish literature as a reference to a narrative-textual unit.

Book or Anthology?

1093

23:5) are from neighboring narratives but not from the story at hand itself. In the eyes of the rabbis, any verse has significant connections to other verses throughout the canon, not just in the passage in which the verse is found, and those cross-canonical connections produce meaning.5 For the rabbis, then, the Bible is not really a book at all. Rather, the Bible is a hypertext, a database with myriad internal connections spanning the whole canon. These connections link any one verse to many other verses, which in turn are linked to a large number of additional verses. Thus a given verse has several literary contexts, each of which implies several additional contexts. The physical data-storage technologies available to humanity in the midrashic era allowed a verse to be contextualized next to only a few other verses. But in reality as the classical rabbis conceived it, any one biblical verse is part of a matrix of verses, each of which invokes additional matrices and thus encourages new combinations of biblical verses and texts.6 Because midrash, as an atomizing form of reading, links words or verses from one part of Scripture with words or verses from elsewhere, the notion of book is of little significance for a rabbinic approach to Scripture. (A possible exception to this statement occurs in m. Meg. 4:4, which rules that hafṭarah readings can skip from section to section within a biblical book but not from one book to another.7 This shows that rabbinic law sees the unit of book as having at least some significance. The fact remains that rabbinic hermeneutics, like all ancient Jewish hermeneutics, never confronted the unit of book.) The relative unimportance of the concept of book and its textual flow also emerges from the lectionary practices of Judaism. Jews do not hear prophetic books chanted from start to finish in the synagogue but listen to discrete passages from the prophets, so that most synagogue-attending Jews have no sense of prophetic books as wholes. Even in the case of the Pentateuch, the context of the book as a whole is largely broken down liturgically. Although the Pentateuch is chanted sequentially in its entirety, this reading is accomplished piece by piece over the course of a year; and in some ancient communities (where a Palestinian rather than Babylonian lectionary cycle was in use) the reading was accomplished over three to three and a half years.8 Thus religious Jews both ancient and modern  On these intertextual linkages as the heart of midrashic exegesis, see esp. D. Boyarin, Intertextuality and the Reading of Midrash (Bloomington: Indiana University Press, 1990). 6  See M. Fishbane, The Garments of Torah: Essays in Biblical Hermeneutics (Bloomington: Indiana University Press, 1989), 124. 7  So also Mišneh Torah, Tefillah 14:13; Šulḥan ʿArukh, O.H. § 144. 8  For a brief description of these cycles, see G. S temberger and H. L. S track , Intro­ duction to the Talmud and Midrash (trans. M. Bockmuehl; Edinburgh: T&T Clark, 1991), 262–264, and I. Elbogen and J. Heinemann, Jewish Liturgy: A Comprehensive History (trans. R. P. Scheindlin; Philadelphia: Jewish Publication Society; New York: Jewish Theological Seminary of America, 1993), 129–138. Some modern synagogues have introduced a triennial cycle that is even more disjunctive than its ancient counterpart. See R. Eisenberg , “A 5

1094

Benjamin D. Sommer

have been encouraged by their liturgical practices to view Scripture as a series of short segments that link up with one another across local contexts rather than as whole scrolls. There is a second respect in which we may characterize midrash, and indeed the intellectual project of classical rabbinic Judaism as a whole, as centrifugal rather than centripetal. Midrashic collections string together alternate interpretations of a verse and often record exegetical controversies among rabbis.9 Classical editions of medieval rabbinic commentary on the Bible in widespread use (known as Miqra‍ʾot Gedolot) achieve a similar effect in the way they present debate among commentators in various columns.10 The setup of the page in the Miqra‍ʾot Gedolot is important when we consider Jewish ideas of scripturality, since from the medieval era until recently a rabbinic Bible was the only Bible most Jewish scholars used. Single-volume editions containing only the Bible itself were rare among Jews in the Middle Ages, and they remain somewhat rare even among contemporary ultra-Orthodox Jews.11 Argument and exchange, then, are the focus of traditional Jewish Bible study. Similarly, the Talmuds consist largely of debates about a wide variety of matters, whether they are records of actual discussions that took place in the academies or literary creations. The multiplicity that is essential to rabbinic culture becomes especially clear in a phrase that appears in b. ʿErub. 13b and b. Giṭ 6b: ‫אלו ואלו דברי א־להים חיים‬ Complete Triennial System for Reading the Torah,” https://www.rabbinicalassembly.org/sites/ default/files/public/halakhah/teshuvot/19861990/eisenberg_triennial.pdf (accessed 5/3/16). 9  Exegetical debate plays a much smaller role in nonrabbinic Judaisms of the ancient world. In Qumran, for example, the act of interpretation, at least by the Teacher of Righteousness, was regarded as inspired (see, e.g., 1QpHab 7:2–4), and therefore recording several possible interpretations was unnecessary. Precisely because the rabbis saw their readings as human in origin and fallible, they engaged in debate regarding correct understandings of biblical verses. See S. Fraade, From Tradition to Commentary: Torah and Its Interpretation in the Midrash Sifre to Deuteronomy (SUNY Series in Judaica; Albany: State University of New York Press, 1991), 3–5, 9, 13. 10  For evocative portrayals of the multiplicity of voices in traditional Jewish commentary, see E. L. G reenstein , “Medieval Bible Commentaries,” in Back to the Sources (ed. Barry Holtz; New York: Summit Books, 1984), 213–259, here 214–215, 256–257; E. L evinas , Beyond the Verse: Talmudic Readings and Lectures (trans. G. D. Mole; London: Athlone, 1994), 137; and A. Gillman, “Between Religion and Culture: Mendelssohn, Buber, Rosen­ zweig and the Enterprise of Biblical Translation,” in Biblical Translation in Context (ed. F. Knobloch; Bethesda, MD: University Press of Maryland, 2002), 92–114, here 101–102. As both Greenstein and Gillman show and Levinas hints, the physical layout of the printed page of a Miqra‍ʾot Gedolot or of a talmudic tractate conveys a sense of polyvocality that is essential to rabbinic cultures. 11  Even the earliest printed Hebrew Bibles usually included the commentaries; see M. J. M ulder , “The Transmission of the Biblical Text,” in Mikra: Text, Translation, Reading and Interpretation of the Hebrew Bible in Ancient Judaism and Early Christianity (ed. M. J. Mulder; CRINT; Philadelphia: Fortress, 1988), 116–121, and H. Z afren , “Bible Editions, Bible Study and the Early History of Hebrew Printing,” ErIsr 16 (1982), 240–251.

Book or Anthology?

1095

(“Both these and those are the living words of God”). This phrase bestows approval upon each side of a debate, however mutually exclusive the two sides are.12 Midrashic reading, then, differs in a fundamental way from holistic methods of reading that emphasize the integrity of literary units such as poems, stories, or whole books. This is not to deny that some elements of holistic reading occasionally appear in rabbinic commentary. In one respect, midrash has a strong harmonizing tendency: it often brings together two contradictory verses and proceeds to resolve the contradiction so that the two verses work together as a unity. Thus midrash is at once atomistic and holistic: it reads individual verses rather than longer units, but it often does so in order to harmonize them.13 A shift toward holistic reading occurs among some medieval Jewish interpreters of the Bible, especially from the French school associated with Rashi and Rashbam. Deeply influenced by their contemporaries, the Christian biblical scholars working in the Abbey of Saint Victor in Paris, Rashi and his followers emphasized what for rabbinic culture was the new idea of local literary context.14 12  On the use of this dictum in the Talmud itself, see A. Sagi, Elu Va-Elu: A Study on the Meaning of Halakhic Discourse (Tel Aviv: Ha-Kibbutz Ha-Meʾuḥad, 1996), 10–24 (Hebrew). In the remainder of his book, Sagi investigates varied understandings of this dictum in later rabbinic cultures, medieval and modern alike. See also the astute observations of Z. B raiterman, “Elu Ve-Elu: Textual Difference and Sublime Judgement in Eruvin and Lyotard,” in Textual Reasonings: Jewish Philosophy and Text Study at the End of the Twentieth Century (ed. P. Ochs and N. Levene; Grand Rapids, MI: Eerdmans, 2003), 206–213, who critiques the sloppy use of this dictum among postmodern thinkers and liberal Jewish theologians alike. See further E. S chweid , The Philosophy of the Bible as Foundation of Jewish Culture (trans. Leonard Levin; 2 vols.; Boston: Academic Studies Press, 2008), 1:35. For an especially interesting medieval discussion relevant to our concerns, see the section titled ‫מעלת אלו ואלו דברי א־להים‬ ‫ חיים‬in I. H orowitz , Šenei Luḥot Haberit (Amsterdam, 1648), 199a (Hebrew), or 1:43a of the 1930 Warsaw edition. On varied attitudes toward legal pluralism and its limits in classical rabbinic literature (e.g., its lesser prominence in the Jerusalem Talmud as opposed to the Babylonian), see R. Hidary, Dispute for the Sake of Heaven: Legal Pluralism in the Talmud (BJS; Providence, RI: Brown Judaic Studies, 2010). 13  On the interplay between the perception of Scripture as a unity and Scriptureʼs diversity in rabbinic interpretation, cf. Fishbane, Garments (see n. 6), 123–124. 14  See R. Harris, “Jewish Biblical Exegesis from Its Beginnings to the Twelfth Century,” in The New Cambridge History of the Bible, Volume 2: From 600 to 1450 (ed. R. Marsden and E. A. Matter; Cambridge: Cambridge University Press, 2003–2009), 596–615, here 604–612; idem , “Concepts of Scripture in the School of Rashi,” in Jewish Concepts of Scripture: A Comparative Introduction (ed. B. D. Sommer; New York: New York University Press, 2012), 102–122; S. Kamin, Rashiʼs Exegetical Categorization in Respect to the Distinction Between Peshat and Derash (Jerusalem: Magnes Press, 1986), 14–15 (Hebrew); E. Touitou, Exegesis in Perpetual Motion: Studies in the Pentateuchal Commentary of Rabbi Samuel Ben Meir (Ramat Gan: Bar Ilan University Press, 2005), 126–134 (Hebrew). Analogous trends emerged a few centuries earlier among Jews in the Arabic-speaking world, though without quite the

1096

Benjamin D. Sommer

Especially in the work of Rashiʼs grandson Rashbam and his circle, we find a centripetal approach closer to modern literary interpretation. But this approach developed late in the history of Jewish biblical interpretation, and it remained somewhat marginal. Rashbam and his emphasis on pəšaṭ were largely forgotten until Moses Mendelssohn began to emphasize his work in the late eighteenth century. Even since that time, pəšaṭ interpretations, though beloved by modern Jewish biblical scholars, have not taken center stage within the community of Jewish readers.15 Nevertheless, in the enormously popular and influential commentary of Rashi, we find a mode of reading that combines midrashic data with an emphasis on local literary context and thus qualifies as centripetal even though his readings do not exemplify pəšaṭ in the typical use of the term.16 In short, Jewish modes of reading Scripture have included both atomistic, multiplicity-seeking trends and holistic, unity-seeking forces; but the former have been more widespread.17 The seeds of both trends go back to the Bible itself. On the one hand, the redactors of the Pentateuch appear largely unconcerned with literary unity or ideational harmony. As Baruch Schwartz stresses, the Pentateuchʼs redactors created a text whose plot is full of undisguised self-contradictions and whose laws conflict with one another on questions that are both multiple and obvious.18 Joel Baden summarizes this position and its implications: The extent of contradiction that the compiler allowed to stand in his combined text is extraordinary. [. . .] The compilerʼs primary goal was not to resolve contradictions same emphasis on the unity of local textual context; see M. Polliack, “Concepts of Scripture among the Jews of the Medieval Islamic World,” in Jewish Concepts of Scripture: A Comparative Introduction (ed. B. D. Sommer; New York: New York University Press, 2012), 80–101. On precursors to a contextual approach in late talmudic sources, especially in Babylonia, see C. Hayes, “Displaced Self-Perception: The Deployment of Minim and Romans in B. Sanhedrin 90b–91a,” in Religious and Ethnic Communities in Later Roman Palestine (ed. H. Lapin; Bethesda, MD: University Press of Maryland, 2006), 249–289. 15  S. Japhet, “Major Trends in the Study of Medieval Jewish Exegesis in Northern France,” Trumah 9 (2000), 43–61, esp. 43–44, 57, and E. B reuer and C. G afni , “Jewish Biblical Scholarship between Tradition and Innovation,” in Hebrew Bible/Old Testament: The History of Its Interpretation (ed. M. Sæbø; Göttingen: Vandenhoeck & Ruprecht, 2012), 3:262–302, here 296–302. 16  On the differences between Rashi and Rashbam regarding pəšaṭ, see K amin , Rashiʼs Exegetical Categorization (see n. 14), 266–274. 17  For a fine example of both modes on a single page, see b. Šabb. 30b (‫ד״ה רב יהודה בריה‬ ‫)דרב שמואל‬, in which the earlier Hebrew comment about Qoheleth is centrifugal whereas the later Aramaic statement is centripetal. See further D. Frankel, The Land of Canaan and the Destiny of Israel: Theologies of Territory in the Hebrew Bible (Siphrut: Literature and Theology of the Hebrew Scriptures; Winona Lake, IN: Eisenbrauns, 2011), 387–388. 18  See esp. B. J. S chwartz , “The Torah: Its Five Books and Four Documents,” in The Literature of the Hebrew Bible: Introductions and Studies (ed. Z. Talshir; Jerusalem: Yad Ben-Zvi Press, 2011), 161–226, here 211–218 (Hebrew).

Book or Anthology?

1097

[. . .] [but] to retain as much of his source material as possible. [. . .] By preserving four discrete and distinct documents, each of which relates its own version of the early history of Israel and argues for a particular view of Israelite religion, the compiler has made an important theological statement. [. . .] No one viewpoint captures the entirety of the ancient Israelite religious experience. [. . .] The competing voices preserved in the Pentateuch are, in fact, complementary, even as they disagree. Only when they are read together is the picture complete. [. . .] To attempt to read the canonical Pentateuch as having a single theological message, be it that of the compiler or one of his sources, is to gravely misunderstand the meaning of the final form of the text, of the compilerʼs work.19

To be sure, counterexamples can be cited (such as death reports, which do not display redundancies we might expect if the entire text of each source underlying the Pentateuch were wholly preserved).20 It is possible that the redactor avoided especially ridiculous plot doublets, but from the many contradictions and doublets that remain it is clear that narrative coherence mattered much less to the redactor than fidelity to older source material. Thus the method of the Pentateuchʼs compiler marks the beginning within Jewish tradition of the attitude that both these and those are the living words of God. Openness to multiple viewpoints that are left as they are without harmonization is characteristic of rabbinic culture, but it dates back at least to the time of the Pentateuchʼs compilation.21 On the other hand, the centripetal approach is also known within the Bible. Though the Pentateuch gives no indication that its laws need to be reconciled, historiographic literature of the Second Temple period nevertheless strives to harmonize the various law codes. To take a famous example: Chronicles suggests how one can observe the contradictory laws of the Passover ritual meal as they appear in Exodus and Deuteronomy. Both Exod 12 and Deut 16 require all Israelite families to slaughter and consume an animal as a Passover ritual. But Exod 12:8–9 directs Israelites to roast the offering, whereas Deut 16:6–7 requires that the offering be boiled.22 Second Chronicles 35:13 maintains that  J. S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypothesis (AYBRL; New Haven, CT: Yale University Press, 2012), 221–228. Cf. Schweid, Philosophy of the Bible (see n. 12), 1:36, though Schweid makes the unwarranted assumption that the redactor wants the reader to harmonize what the redactor left contradictory. 20  See D. M. Carr, The Formation of the Hebrew Bible: A New Reconstruction (New York: Oxford University Press, 2011), 112. 21  On what we might call the logic of ‫ אלו ואלו‬in the biblical canon generally, see the remarks of J. B lenkinsopp , Prophecy and Canon: A Contribution to the Study of Jewish Origins (Notre Dame, IN: University of Notre Dame Press, 1977), 94–95. On biblical booksʼ lack of emphasis on their own textual unity, see also I. Heinemann , Darkhei Ha-Aggadah (Jerusalem: Magnes, 1970), 56–57 (Hebrew). 22  In Biblical Hebrew, the verb ‫ ִבּ ֵשל‬means not “cook” (as in modern Hebrew) but “boil, cook in water.” In order to harmonize the contradiction between two passages, traditionalist commentators have argued that the verb can mean merely “cook,” but such a reading requires 19

1098

Benjamin D. Sommer

the offering was somehow roasted and boiled simultaneously: “They boiled the Passover offering in fire, in accordance with the law.” Thus the Chronicler regards the Pentateuch as a legislative unity, not as a collection that preserves the wording of older sources within a broad if bumpy narrative framework.23 According to the Chronicler, one does not choose which law to observe; one attempts, through a somewhat forced exegesis, to observe a law that encompasses the wording of both. Similar exegeses are found in other passages of Chronicles and in Ezra–Nehemiah. Scholars have shown that the authors of both these works engage in halakic midrash to resolve contradictions in the Pentateuchʼs laws.24 This also occurs, albeit less frequently, with narrative passages. Take Neh 9:13: ‫ועל הר סיני ירדת ודבר עמהם משמים‬, “You came down to Mount Sinai to speak to them from heaven.” This verse values harmonization more than it fears oxymoron: it gives all the relevant passages from the Pentateuch their due: Exod 19 tells us that God descended to Sinai before the lawgiving, whereas Deut 4–5 insist that God remained in heaven throughout the lawgiving at Horeb (so also Exod 22:22b). On occasion, the centripetal point of view occurs even within the Pentateuch itself. Several scholars have shown that Exod 34:18–26 is the product of a postredactional addition to the Pentateuch – that is to say, it is one of the rare pentateuchal texts that even Neo-Documentary scholars do not attribute to J, E, P, or D. Michael Fishbane, Shimon Gesundheit, David Carr, and Bernard Levenson demonstrate that the festival calendar in Exod 34 represents a melding of legal passages from elsewhere in the Pentateuch.25 The passage represents a us to assume that the Pentateuch uses the verb in radically different senses in the Passover laws of Exod 12:9 and Deut 16:7. As M. Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Clarendon, 1985), 136–137, notes, “This argument is both tendentious and circular.” See also the useful treatment in S. R. Driver, Deuteronomy (3rd ed.; ICC; Edinburgh: T&T Clark, 1902), 192–193. 23  2 Chr 35:12 refers to “the book of Moses,” which, to judge by the several attempts Chronicles makes to harmonize between the P laws and D laws, combines both. In other words, that “book of Moses” is either the Pentateuch known to us or something very much like it. On the Chroniclerʼs harmonizing in this passage, see esp. Fishbane, Biblical Interpretation (see n. 22), 136–137. 24  Y. K aufmann , Toledot ha-emunah ha-yisraelit (4 vols.; Jerusalem: Bialik and Devir, 1937–1956), 4:291–293, 327–329, 331–338, 341–342, 346–350 (Hebrew). See further S. Japhet, From the Rivers of Babylon to the Highlands of Judah: Collected Studies on the Restoration Period (Winona Lake, IN: Eisenbrauns, 2006), 137–151, and multiple examples in Fishbane, Biblical Interpretation (see n. 22), 107–162. 25  See F ishbane , Biblical Interpretation (see n. 22), 194–197; D. M. C arr , “Method in Determination of Direction of Dependence: An Empirical Test of Criteria Applied to Exodus 34, 11–26 and Its Parallels,” in Gottes Volk am Sinai: Untersuchungen zu Ex 32–34 und Dtn 9–10 (ed. M. Köckert and E. Blum; Gütersloh: Chr. Kaiser Gütersloher Verlagshaus, 2001), 107–140; S. G esundheit , Three Times a Year (FAT 82; Tübingen: Mohr Siebeck, 2012), 12–43.

Book or Anthology?

1099

revision of the festival law in Exod 23 in light of festival and sacrificial laws in both P (Exod 12:10; Lev 23:15) and D (Deut 4:38; 9:4–5; 12:20; 14:8; 18:12). This addition to the Pentateuch attempts to create a comprehensive pentateuchal law of the festivals that borrows from and thus harmonizes the sometimes contradictory laws on the festivals elsewhere in the Pentateuch. In contrast to the compiler (who left intact several discrepancies among the festival laws in the Pentateuchʼs sources), the late interpolator responsible for Exod 34:19–26 attempted to demonstrate that it is possible to bring language from varied traditions together in a way that entails no contradictions or at least diverts our attention from them. Such late additions are rare in the Masoretic Text. They appear somewhat more frequently, however, in the Samaritan version of the Pentateuch – especially in passages that describe the revelation at Sinai. As a result, the Masoretic Pentateuch largely preserves a more pristine centrifugal voice in what for Jews is the most sacred part of their Scripture (the Pentateuch), even as other parts of their Scripture (Ezra–Nehemiah, Chronicles) foreshadow the harmonizing trends evident in some medieval rabbinic documents.

Rosenzweig, R, and Canon Criticism Having outlined these two modes of biblical reading and their forebears in the biblical texts itself, I would like to move on to the attitude toward biblical interpretation among some modern theologians. Religious readers of Scripture often look askance at biblical criticism because, they claim, religious readers seek unity in scriptural texts, whereas biblical critics atomize them. For example, both Franz Rosenzweig and Brevard Childs privilege the final form of the text as an integrated whole over the parts from which the whole was constituted. Neither Rosenzweig nor Childs denies the validity of compositional criticism, but they regard centrifugal modes of reading as carrying less religious relevance than centripetal ones; indeed, Rosenzweig claims not to see religious significance in centrifugal ones at all.26 Contrary to Rosenzweigʼs view, I think this critique has little traction for Jews who want to read the Bible in a Jewishly authentic way, because, as we have seen, the final form of scriptural books as wholes is not important for midrashic exegesis.

 Both Rosenzweig and Abraham Joshua Heschel acknowledge the intellectual legitimacy of biblical criticism while minimizing or effectively denying its religious relevance. Rosen­ zweig and Heschel thus can avoid the charge of fundamentalism while remaining essentially unaffected by modern biblical scholarship. See J. D. Levenson, “Religious Affirmation and Historical Criticism in Heschelʼs Biblical Interpretation,” AJSR 25 (2000–2001), 25–44, here 39–40, whose comments on Heschel could describe Rosenzweig as well. 26

1100

Benjamin D. Sommer

Rosenzweig famously expressed his views on the proper way for Jews to interpret Scripture in a letter to Jacob Rosenheim, a leader of Orthodox Jewry in Germany, concerning the German translation of the Bible that Rosenzweig wrote together with Martin Buber. Rosenzweig tells Rosenheim that he and Buber regard the Pentateuch as a literary unity. To be sure, he clarifies, they do not accept the view that God gave the whole Pentateuch to Moses, and in principle they are open to the possibility that the Documentary Hypothesis or some similar theory correctly describes the historical development of the Pentateuch. But, he explains, development is of little import for interpretation or theology. As a result, their mode of reading the Pentateuch is not so different from an Orthodox one: We too translate the Torah as one book. For us too it is the work of a single mind. We do not know who this mind was; we cannot believe that it was Moses. We name that mind among ourselves by the abbreviation with which the Higher Criticism of the Bible indicates its presumed final redactor of the text: R. We, however, take this R to stand not for redactor but for rabbenu [“our Rabbi”]. For whoever he was, and whatever text lay before him, he is our teacher, and his theology is our teaching. An example: let us suppose that Higher Criticism is right, and that Genesis 1 and Genesis 2 are in fact by different writers. [. . .] Even in that case, however, it would remain true that what we need to know from the account of creation is not to be learned from either chapter alone but only from the juxtaposition and reconciliation of the two. Indeed, it is to be learned only from the reconciliation of the apparent contradictions from which the critical distinction begins: the “cosmological” creation of the first chapter, which leads up to man, and the “anthropological” creation of the second chapter, which begins from man. Only this sof maʿaseh ba-maḥashabah teḥillah [“what was created last has conceptual priority,” a quote from the Sabbath hymn Lekhah Dodi] is the necessary teaching.27

A similar point of view plays a central role in the work of Brevard Childs. In a programmatic essay, he wrote, It is [a] [. . .] grievous error for Christian scholars to assume that the reconstruction of the literatureʼs historical development can now replace the study of the canonical shape of the Pentateuch. [. . .] Rather, the present shape of the Pentateuch offers a particular interpretation – indeed a confession – as to how the tradition was to be understood by the community of faith. Therefore, it seems to me important first of all to describe the

27  In M. Buber and F. Rosenzweig, Scripture and Translation (trans. L. Rosenwald; Indiana Studies in Biblical Literature; Bloomington: Indiana University Press, 1994), 23; I add various explanations in brackets. On Rosenzweigʼs unifying readings, see Y. Amir, A Small Still Voice: Theological Critical Reflections (Sifre Ḥemed; Tel Aviv: Yediʿot ʾaḥaronot, 2009), 181 (Hebrew), and idem, Reason Out of Faith: The Philosophy of Franz Rosenzweig (ʾAron Sefarim Yehudi; Tel Aviv: Am Oved, 2004), 284–285 (Hebrew), who shows that Rosenzweig believes the religious reading of the editorial unity recalls the approach of classical midrash. In believing this, Rosenzweig is, I think, fundamentally unaware of the atomistic nature of midrashic reading.

Book or Anthology?

1101

actual characteristics of the canonical shape and secondly to determine the theological significance of this shape.28

Childs does not deny the findings of modern biblical critics regarding multiple layers of authorship and supplementation in the Bible, but he insists that the final form of the text reflects the work of the redactors and canonizers who have created a whole rather than a jumble of parts, and that final form is what the religious interpreter needs to read. The religious interpreter is not required to ignore what we know about the multiplicity of the text; on the contrary, a central goal of religious reading for Childs is to see how the redactor shaped a unity from diverse materials.29 The quest for unity is fitting for Protestant readings of Scripture, because Protestant Christianity has often assumed that the Bible speaks as a unity.30 (In light of both the history of theology and comparative religions, this assumption is somewhat unusual.31) From the point of view of a religious community com28  B. S. Childs, “The Old Testament as Scripture of the Church,” CTM 43 (1972), 709–722, here 711, 715. 29  See C hilds , “Old Testament as Scripture” (see n. 28); idem , Introduction to the Old Testament as Scripture (Philadelphia: Fortress, 1979); idem , Biblical Theology of the Old and New Testaments: Theological Reflection on the Christian Bible (Minneapolis: Fortress, 1993). For a useful overview and contextualization of Childsʼs method, see L. Perdue, “Old Testament Theology Since Barthʼs Epistle to the Romans,” in Biblical Theology: Introducing the Conversation (ed. L. Perdue; Library of Biblical Theology; Nashville: Abingdon, 2009), 55–136, 285–299, here 109–114, 295. For a subtle and sympathetic discussion of Childsʼs exegetical and theological proposals, see C. S eitz , Word without End: The Old Testament as Abiding Theological Witness (Grand Rapids, MI: Eerdmans, 1998), 102–109. For critical appraisals, see J. Barton , Reading the Old Testament: Method in Biblical Study (rev. and enlarged ed.; Louisville: Westminster John Knox, 1996), 77–103; J. B arr , The Concept of Biblical Theology (Minneapolis: Fortress, 1999), 378–438; B. D. S ommer , “The Scroll of Isaiah as Jewish Scripture, Or, Why Jews Donʼt Read Books,” in Society of Biblical Literature Seminar Papers (Atlanta: Scholars Press, 1996), 225–242, idem, review of Isaiah: A Commentary, by B. Childs, Bib 83 (2002), 579–583. 30  On the tendency of Christian readers to harmonize and universalize and hence to disregard much of the available biblical evidence, see W. B rueggemann , Theology of the Old Testament: Testimony, Advocacy, Dispute (Minneapolis: Fortress, 1997), 327. However, this magnum opus by Brueggemann shows that such harmonizing readings are not the only possible ones for a Christian; his Theology represents a liberal Protestant attempt to read the Bibleʼs theological diversity as religiously instructive. For an evangelical attempt to do so, see P. Enns, Inspiration and Incarnation: Evangelicals and the Problem of the Old Testament (Grand Rapids, MI: Baker Academic, 2005), 71–112. See further J. Barton , “Unity and Diversity in the Biblical Canon,” in The Unity of Scripture and the Diversity of the Canon (ed. J. Barton and M. Wolter; Berlin: de Gruyter, 2003), 11–26; J. Goldingay, Theological Diversity and the Authority of the Old Testament (Grand Rapids, MI: Eerdmans, 1987). 31  On the evolution in Western Christianity, starting in the late first millennium and culminating in the Reformation, of the tendency to view the Bible as Scripture rather than as a collection of Scriptures, see W. C. Smith, What Is Scripture? A Comparative Approach

1102

Benjamin D. Sommer

mitted to the doctrine of sola scriptura, when one crosses the boundary of the canon, one moves into a different, and lesser, realm, and therefore what is within the boundary must have a conceptual integrity or singularity. But rabbinic forms of Judaism, like Catholic and Orthodox Christianity, emphasize the authoritative status of tradition alongside or even over that of Scripture.32 Thus the literary integrity of the entities on each side of the boundary can matter less. Because traditional Jewish interpreters of Scripture are at least as centrifugal as centripetal in orientation, the emphasis that Rosenzweig and Childs place on textual unity appears extraneous for a Jewish reader.33 As a result, I cannot see why, for a Jewish biblical theologian, the redactor of the Pentateuch should have a more important voice than the P authors or the D authors who came before him or than various commentators on the Pentateuch who came after.34 In saying this, I am not claiming that all voices carry the same weight or that all opinions are equally valid. Jewish tradition over the ages has made it clear, for example, that Rashiʼs commentaries on the Bible are more important than those of his rough contemporary Ḥazzequni and that the Tosefta is part of Torah whereas the book of Jubilees is not. (I use the term Torah throughout this essay in its broad sense, to include all sacred Jewish learning. To avoid confusion, when I refer in this essay to the Five Books of Moses, I (Minneapolis: Fortress, 1993), 13–14, 53–54. Smith demythologizes this view, showing how unusual it is in the history of religions (cf. 126–127). In light of Smithʼs work, it becomes clear that the assumption of Childs and his followers that canon entails first and foremost unity is just that: an assumption, not a natural or inevitable feature of Scripture. It is a historically conditioned view that reflects not the world of the canonizers (as J. B arr , Holy Scripture: Canon, Authority, Criticism [Philadelphia: Westminster, 1983], 49–74, 130–71, notes) but the post-Gutenberg, post-Reformation era in which Childs worked. 32  For the notion in Catholicism and Judaism that tradition is a more or less equal partner with Scripture, or even that Scripture is merely a subset of tradition, see B. D. Sommer, “Unity and Plurality in Jewish Canons: The Case of the Oral and Written Torahs,” in One Scripture or Many? Perspectives Historical, Theological and Philosophical (ed. C. Helmer and C. Landmesser; New York: Oxford University Press, 2004), 108–150, esp. 123–127; Y. Congar, The Meaning of Tradition (trans. A. N. Woodrow; San Francisco: Ignatius Press, 2004), esp. 13 (though see also the reservations at 99–102); A. Dulles, “Reflections on ‘Sola Scriptura,’ ” in Revelation and the Quest for Unity (Washington, DC: Corpus Books, 1968), 67–74. For a similar view in the work of an Anglican theologian, see D. Brown, Tradition and Imagination: Revelation and Change (Oxford: Oxford University Press, 1999), esp. the clear statement of the thesis on page 1 and the discussion at 106–167. 33  In defense of Rosenzweig, it should be noted that at least to some degree the unity of which he speaks is ultimately a unity of Scripture together with tradition and thus a unity that encompasses multivocality. See L. Batnitzky, Idolatry and Representation: The Philosophy of Franz Rosenzweig Reconsidered (Princeton: Princeton University Press, 2000), 125–127. 34  For an approach very similar to what I propose here, see Frankel, Land (see n. 17), 385. On the religious irrelevance of questions regarding authorship, see M. Z. Brettlerʼs remarks in M. Z. Brettler et al., The Bible and the Believer: How to Read the Bible Critically and Religiously (New York: Oxford University Press, 2012), 54–55.

Book or Anthology?

1103

use the term Pentateuch.) I am merely pointing out that once we realize that the Pentateuch is as much an anthology as a book, there is no reason to see the anthologizer as more sacred, more authoritative, or even more interesting than the anthologized. Further, the well-known passage about R that I quoted from Rosenzweigʼs letter is problematic from an exegetical point of view. Rosenzweig maintains that what is to be learned from Genesis emerges not from what either creation account says on its own but “only from the juxtaposition and reconciliation of the two.” Yet the interpretation he sketches out – that creation is anthropocentric – could in fact be arrived at by reading either account by itself. In the first account, which stems from P, humanity is the goal and climax of creation, whereas in the second account, from J, humanity is the starting point. Both accounts emphasize the importance of humanity within the scheme of creation, though they express that emphasis in different ways. The interpretation Rosen­ zweig produces involves no reconciliation, since P and J agree on that matter.35 By stressing this shared theme, however, Rosenzweig misses what may be most interesting in the redacted text: the two accountsʼ very different portrayals of human nature and of Godʼs relationship with humanity. In J, humanity is made from mud; in P, humanity is created in the image of God. In J, YHWH is on earth, personally blowing the breath of life into Adamʼs nostrils; in P, God is distant from the world God creates. The final form of Genesis gives us two theologies and also two anthropologies, and R provides no guidance as to how they might be reconciled or even any hint that they should be reconciled. Accepting that both these and those, however different, are Torah is a perfectly good Jewish practice; harmonizing, or pretending to harmonize by noting what the two accounts have in common while ignoring their differences, is not a better method of reading, Jewish or otherwise.36 35  One could just as well read the text source-critically to arrive at the conclusion Rosenzweig articulates. In fact, Amir, Reason (see n. 27), 303, n. 6, points out that in Star of Redemption (at the end of book 1 of part 2), Rosenzweig provided a lengthy reading of Gen 1 by itself. This means, in documentary terms, that Rosenzweig reads Pʼs account, without its J counterpart. (For the passage, see F. Rosenzweig, The Star of Redemption [trans. W. W. Hallo; Boston: Beacon Press, 1972], 151–155; idem, The Star of Redemption [trans. B. Galli; Madison: University of Wisconsin Press, 2005], 163–168.) 36  The tendency of Rosenzweigʼs biblical exegeses to ignore the religiously interesting distinctions that are highlighted when we differentiate among the Bibleʼs authors and schools is equally evident in A. J. Heschel, The Prophets (New York: HarperCollins, 2001). There, as Levenson, “Religious Affirmation” (see n. 26), 43–44, notes, “the thinking and experience of all the prophets tend to run together in a blur. [. . .] By paying close attention to the distinctive voice in each prophetic collection as recovered through critical study, Heschel could have [. . .] helped uncover yet another facet of the shivʿim panim la-torah, the seventy facets of the Torah, of which the Jewish tradition speaks.” But, Levenson shows, Heschel did not do so. This is ironic, since the unattempted project Levenson describes would have fit quite well with Heschelʼs theology.

1104

Benjamin D. Sommer

The problematic nature of Rosenzweigʼs approach to reading the Pentateuch is also evident in his review of the first volume of the German Encyclopaedia Judaica. Rosenzweig criticizes the encyclopedia for segregating discussions of the Bible itself from treatments of biblical interpretation among the classical and medieval sages. He calls for a new form of biblical scholarship that he hopes to see in some future publication: This new biblical scholarship will not avert its eyes from critical problems of modern scholarship, but it will apprehend all problems comprehensively [. . .] from the viewpoint of the final redactor – or, to put it differently, from the viewpoint of the first reader. This new scholarship will rediscover a connection to Jewish biblical scholarship of the past, which [. . .] started out from a similar point of view – only a connection; not more than that; not submission to it. [. . .] Then, perhaps, some day an encyclopaedia will be able to put aside that fearful distinction separating what is in the Bible, what our Sages said, and what modern biblical criticism says!37

Two problems mar Rosenzweigʼs comment. First, Rosenzweig equates the viewpoint of the final redactor with that of the first reader. He assumes that until R did his work, there was no Scripture, and therefore there could be no interpretation of Scripture. But one of the most important conclusions of modern biblical scholarship is that the Bibleʼs first readers lived generations before its final redactors. The Deuteronomist read both E and J and reacted to them throughout Deuteronomy.38 Priests supplemented a law of Passover with clarifying remarks that transformed an older law into the text of Exod 12 as we know it.39 Deutero-Isaiah read P and D – and reacted to them in very different ways that suggest this prophet knew them as separate documents.40 Only after the earliest interpreters produced these readings were the sources combined to produce the first edition of the Pentateuch. Even then, subsequent readers added the legal passage in Exod 34. Rosenzweigʼs decision to privilege the work of the redactor is based on his assumption that the most legitimate reading of the Pentateuch is one that returns to the experience of first readers; but to  F. R osenzweig , Zweistromland: Kleinere Schriften zu Glauben und Denken, vol. 3 of Der Mensch und Sein Werk (ed. R. Mayer and A. Mayer; Dordrecht: Franz Rosenzweig; Boston: M. Nijhoff, 1984), 734; my translation. 38  The literature is extensive. See, e.g., D river , Deuteronomy (see n. 14), iii–xix; M. W einfeld , Deuteronomy and the Deuteronomic School (Oxford: Clarendon, 1972), 179–243; B. M. L evinson , Deuteronomy and the Hermeneutics of Legal Innovation (New York: Oxford University Press, 1997); T. Veijola, Moses Erben: Studien zum Dekalog, zum Deuteronomismus, und zum Schriftgelehrtentum (BWANT 149; Stuttgart: Kohlhammer, 2000); E. Otto, Das Deuteronomium: Politische Theologie und Rechtsreform in Juda und Assyrien (BZAW 284; Berlin: de Gruyter, 1999), as well as many relevant sections in F ishbane , Biblical Interpretation (see n. 22). 39  See Gesundheit, Three Times (see n. 25), 46–58. 40  See B. D. S ommer , A Prophet Reads Scripture: Allusion in Isaiah 40–66 (Contraversions; Stanford: Stanford University Press, 1998), 132–151. 37

Book or Anthology?

1105

return to the first readers we must interpret preredacted sources underlying the Pentateuch and not the Pentateuch itself. Thus if we follow Rosenzweigʼs advice by acknowledging the findings of biblical criticism rather than averting our eyes from them, we have to recognize that there were Jewish readings of Scripture before there was a Bible, and those scriptural readings were produced by sages whose teachings are part of Torah. Second, here and in his letter to Rosenheim, Rosenzweig assumes that Jewish biblical scholarship should accentuate the unity of the text. Consequently, he regards modern atomizing readings as less relevant to Jewish concerns. These readings, Rosenzweig avers, are not meaningful for a Jewish reader who wants to find torah or guidance in the Bible, because the traditional interpreters bring the text together rather than splitting it apart. Thus for Rosenzweig, if we want to enter the age-old and ongoing conversation of Jewish learning, we would do best to read for unity rather than for the disunity characteristic of so much biblical criticism. In fact, however, we have seen that a great deal of Jewish religious reading is atomistic. Midrash typically takes verses out of their immediate contexts and connects them with verses elsewhere to form what is in effect another text altogether.41 Now, it should be evident that one could substitute the subject source criticism for midrash in the previous sentence, and the new sentence would remain true. A midrashic mode of reading is in significant ways closer to Julius Wellhausen and Baruch Schwartz than it is to Franz Rosenzweig or Brevard Childs.42

The Contemporary Debate I have argued that compositional criticism of biblical texts allows us to recover voices from Jewish tradition that precede the redaction of the Pentateuch. Just as it is religiously significant for a Jewish reader to study the position of Hillel as distinct from that of Shammai in the Mishnah, so too it can be relevant to consider the differing theological positions of P and D in the Pentateuch. By fostering a particular sort of atomistic reading, modern approaches to the study of the Pentateuch encourage an interpretive practice that suits Jewish modes of interpretation quite well in light of traditional Judaismʼs predilection toward atomistic reading.

 See esp. Boyarin, Intertextuality (see n. 5).  Further, the exegetical spur behind the decontextualization and recontextualization of biblical verses in both midrash and source criticism is the same: both midrash and source criticism attempt to create coherence in a text that is often incoherent as it stands. On this often-neglected aspect or source criticism, see J. B arton , The Nature of Biblical Criticism (Louisville, KY: Westminster John Knox, 2007), 43–44. 41

42

1106

Benjamin D. Sommer

A final question deserves attention. How does the contemporary debate between Neo-Documentary and Sources-and-Supplements models for reconstructing the composition of the Pentateuch affect a Jewish approach to reading the Pentateuch at once critically and theologically? Both types of models dovetail well with classical Jewish literature but in strikingly different ways. Work by recent scholars who emphasize a multiplicity of materials that were combined, revised, and supplemented to form the Pentateuch allows us to posit a parallel between the way that the Pentateuch came into existence and the way classical rabbinic texts, and especially the Babylonian Talmud, came into existence according to modern rabbinics scholarship.43 In both cases, later sages copied and adapted older sources even as they composed their own additions, some very brief and some longer, to those sources. The Pentateuch, like the Talmud, emerged as an evolving corpus. What Carr calls the Sources-and-Supplements model, then, allows us to find in the Pentateuch a very rabbinic book: it is authored by an interpretive tradition that appended it, reacted to it, and updated it over generations. This school of thought emphasizes a favorite theme of many modern Jews: the ever-unfolding and malleable nature of Torah. This core characteristic of Torah originated not in Germany of the nineteenth century nor in Babylonia of the mid-first millennium CE. Rather, it is found in the layering of the Pentateuch itself. To be sure, other aspects of biblical criticism have also focused our attention on the evolving and exegetical nature of biblical documents; this is especially the case in the study of innerbiblical exegesis.44 But for  For introductions to this sort of scholarship, see J. Kulp and J. Rogoff, Reconstructing the Talmud: An Introduction to the Academic Study of Rabbinic Literature (New York: Mechon Hadar, 2013), and D. W. Halivni, The Formation of the Babylonian Talmud (trans. and annotated by Jeffrey Rubenstein; Oxford: Oxford University Press, 2013). 44  On the continuity of innerbiblical exegesis and postbiblical exegesis, see F ishbane , Garments (see n. 6), 3–32, 64–78; Y. Zakovitch, An Introduction to Inner-Biblical Interpretation (Even-Yehuda: Reches, 1992), 9–11, 131–135 (Hebrew); J. L. Kugel, Traditions of the Bible: A Guide to the Bible as It Was at the Start of the Common Era (Cambridge, MA: Harvard University Press, 1998), 15–19, 895–896; idem, “The Bibleʼs Earliest Interpreters,” Prooftexts 7 (1987): 269–283. See further B. M. L evinson , Legal Revision and Religious Renewal in Ancient Israel (New York: Cambridge University Press, 2008), 18. These scholars embrace and expand an insight that goes back to the beginnings of die Wissenschaft des Judentums, whose founders – especially Leopold Zunz and Abraham Geiger – argued (as S. Schechter, Studies in Judaism, First Series [Philadelphia: Jewish Publication Society, 1896], xvi, put it in regard to the former) that “certain portions of the Bible [. . .] are, in fact, little more than a traditional interpretation of older portions of Scripture, adapted to the religious needs of the time.” Schechter further presaged the study of innerbiblical exegesis when he wrote, “the great fact remains that the best commentary on the Bible is the Bible itself” (S. Schechter, Studies in Judaism, Second Series [Philadelphia: Jewish Publication Society, 1908], 36). Thus it is not surprising that the first Jewish scholars who entered the field of biblical criticism almost immediately pioneered the study of innerbiblical exegesis. See, e.g., Kaufmann, Toledot (see n. 24), 4:291–93 (where he speaks of ‫)מדרש הלכה בעזרא‬, 327–329, 331–338, 341–342, 346–350 43

Book or Anthology?

1107

those scholars convinced by the Sources-and-Supplements model, the study of the origin of the Pentateuch can extend and deepen the analogy between biblical literature and classical Judaism. The Neo-Documentary school is also quite relevant from the point of view of a Jewish biblical theology, but in a new and surprising way. Scholars (such as myself) convinced by this school of thought need not deny the biblical origins of what we might call a classically rabbinic mindset that highlights exegesis and rejuvenating tradition as sacred tasks; the protorabbinic nature of biblical texts is clear from the study of innerbiblical exegesis, regardless of oneʼs position on the composition of the Pentateuch. But Neo-Documentarians are forced to note the conservatism that sometimes accompanies this process of rejuvenation, and they cannot avoid the realization that preservation was no less a core value among ancient Israelite sages than was modification. The Pentateuch is distinctive among literary texts of the ancient Near East in its blatant and constant self-contradiction. To be sure, the Pentateuch is a composite work, and in this regard it resembles other ancient Near Eastern texts like Enuma Elish, the Code of Hammurabi, and the various editions of the Gilgamesh Epic.45 But the Pentateuch is a composite work of a different type. These other works do not repeatedly juxtapose double accounts of a single event. They do not present contradictory narratives and clashing laws the way the Pentateuch does from its very first chapters. It follows, as Schwartz and Baden have emphasized, that the Pentateuchʼs redaction took place in a distinctive, indeed perhaps unique, manner. We saw earlier that the redactor (or, to use Schwartzʼs term, the compiler) treats his four sources with extraordinary fidelity, usually altering their wording as little as possible so as to preserve the original texts to a maximal extent, even when this creates conflicts – not only between two texts at some remove from each other (as is the case with the legal materials) but also in a single narrative and within the space of a few verses. It is for this reason that we (where he speaks of ‫ ;)מדרש התורה וראשית תורה שבעל פה‬I. L. Seeligmann, Voraussetzungen der Midraschexegese (VTSup 1; Leiden: Brill, 1953), 150–181; idem , “The Beginnings of Midrash in the Books of Chronicles,” Tarbiz 49 (1980), 14–32 (Hebrew); D. Y ellin , “Allusion,” in The Writings of David Yellin: Biblical Studies (ed. E. Z. Melammed; Jerusalem: Reuven Mass, 1983), 210–13 (Hebrew), originally published in idem , Selected Writings (2 vols.; Jerusalem: Qiryat Sefer, 1939), 2:22–27; A. Kaminka, Meḥqarim (Tel Aviv: Devir, 1938), 19–30, 52–56, 159–164 (Hebrew); M. Seidel, “Parallels between the Book of Isaiah and the Book of Psalms,” Sinai 38 (1955–1956), 149–172, 229–242, 272–280, 333–355 (Hebrew). 45  See esp. the classic studies in J. T igay (ed.), Empirical Models for Biblical Criticism (Philadelphia: University of Pennsylvania Press, 1985), as well as, more recently, A. R. George, The Babylonian Gilgamesh Epic: Introduction, Critical Edition, and Cuneiform Texts (Oxford: Oxford University Press, 2003), 3–54, esp. 39–54. The dependence of the narrative of the creation of man in Enuma Elish VI on Atraḫasis I is well-known (as are the later textʼs fascinating alterations). The similarity between the names of Marduk in Enuma Elish VII and god lists such as An-Anum evinces the dependence of this final tablet on older lists of that type.

1108

Benjamin D. Sommer

can regard the Pentateuch as the first work for which ‫אלו ואלו דברי א־להים חיים‬, both these and those can be the living words of God, no matter how different they are. For the compiler of the Pentateuch, preserving tradition more or less intact is of greater value than making basic narrative sense or laying forth clear legal norms. Conservation trumps coherence. The Pentateuch as understood by Neo-Documentary critics, then, cultivates a religious outlook that is surprisingly traditionalist. Rosenzweigʼs R turns out to be one of the earliest rabbis, but in a way Rosenzweig did not envision: we may call R our rabbi not because R created a smooth unity but because R was so beholden to tradition that R refused to create a unity at all.

Methodological Reflections on a Theology of the Pentateuch Markus Witte 1 The Pentateuch as Theology The Pentateuch not only contains theology or theologies but is itself theology. According to a simple medieval definition, theology can be understood as speech about God that, firstly, teaches about God (de deo docet); secondly, is taught by God himself (a deo docetur); and thirdly, leads to God (ad deum ducit).1 Hence, theology is speech about God in the mode of reflection, revelation, and inspiration, as well as doxology and paraenesis. In other words, theology is systematic, argumentative, interpretative, and worshipful religious speech.2 The speech about God in the Pentateuch meets these criteria in its different literary layers, especially in its most recent compositional and redactional layer. The frame of the flood narrative in Gen 6:5–7; 8:21–22 and Abrahamʼs dialogue with God in Gen 18:17–33 are examples of reflection. The numerous self-descriptions and speeches of God are examples of inspiration or revelation. The great prayers in Exod 15:1–21 and Deut 32:1–44 are examples of doxology. The whole of Deuteronomy, as an instruction to Israel by Moses, is an example of paraenesis. All of these aspects are present in a compressed form in the scene of revelation in Exod 34:5–8.

2 The Account of the Theology of the Pentateuch An account of the theology of the Pentateuch is descriptive, prescriptive, and constructive. On the one hand, an adequate account of the theology of the Pentateuch includes a description of the various theologies it contains and I warmly thank Tobias Tan of Oxford University for correcting the English version of this article. 1  Albertus Magnus, Commentarii in sententiarum, dist. I, art. ii, in vol. 25 of Opera omnia (ed. A. Borgnet; Paris: Vivés, 1893), 15–17; idem, Summae theologiae, tract. I, quaest. II, in vol. 31 of Opera omnia (ed. A. Borgnet; Paris: Vivés, 1895), 11–12. 2  See R. Smend, Die Mitte des Alten Testaments: Gesammelte Studien 1 (BEvT 99; Munich: Kaiser, 1986), 104–117, here 111–117.

1110

Markus Witte

their manifold interactions. On the other hand, even a simple description of the theology of the Pentateuch always has a normative aspect. For the God who is recounted in the Pentateuch, and in whose name laws are promulgated, is claimed to be creator of the whole world and lord of all history. Furthermore, an account of the theology of the Pentateuch developed within the context of a faith community and a community of interpretation that takes the Bible as the foundation of its speech about God should articulate a constructive and critical contribution to God-talk in the past and in the present. Where the community of faith and the wider society are no longer congruent with each other, the potential for a theology of the Pentateuch to interpret life and to orient communal life is called into question.

3 The Inquiry into the Theology of the Pentateuch An inquiry into the theology of the Pentateuch is not identical with a history of literature, a history of law, or a history of the religion of ancient Israel and early Judaism. Rather, it requires all three of these. The Pentateuch emerged alongside cultural, literary, and historical-religious developments. Therefore a theology of the Pentateuch must integrate the results of research into the history of the literature, law, and religion of ancient Israel and early Judaism. The history of literature contributes to the profiling of the literary layers in the Pentateuch. On a broad scale, a theology of the Pentateuch can have the character of a theological literary history. The history of law shows the forms and history of binding the law to YHWH as its founder, preserver, and supreme authority. Thus a theology of the Pentateuch can also appear as a theological history of law. The history of religion of ancient Israel and early Judaism, which occurs in the milieu of the religions of the ancient Near East, Egypt, the Levant, and ancient Greece, serves to determine the nature and development of YHWH as the central subject, basis, and aim of speech about God in the Pentateuch. Therefore a theology of the Pentateuch can exhibit the contours of a theological history of religion. Together, literary, legal, and religious history illuminate the character of the Pentateuch as a many-voiced discourse about God developed from the constant change of scribal production and interpretation, of tradition and redaction, of integration and adaption. However, as a dialogue produced by redaction, the theology of the Pentateuch is always more than the sum of the theologies of the different literary layers and their contextualization in a particular legal and religious history. Yet the theology of the Pentateuch is not simply identical with the theology of the so-called final redaction and its classification within the history of religions. On the one hand, the final redaction is understood as the conscious

Methodological Reflections on a Theology of the Pentateuc

1111

compromise between different Trägerkreise (schools) and their theologies. On the other hand, the final redaction represents a contemporary group and its theology that prevailed under certain religious and historical conditions.3 For an account of the theology of the Pentateuch is always influenced by the theology of the one giving the account. The presenter of the account (re)constructs the literary, legal, and religious history of ancient Israel and early Judaism under certain contemporary historical conditions and selects and interprets under certain hermeneutic presuppositions. In this respect, a theology of the Pentateuch certainly adopts a point of view “from the outside.”4 As with the production of the Pentateuch and its interpretation within itself, a theology of the Pentateuch is subject to the processes of historical development, reception, and adaption. It both conserves and creates and is in principle always open-ended.

4 Four Coordinates of a Theology of the Pentateuch A historical-critical understanding of the Pentateuch is inevitable, as there has been a point of no return since at least the exegesis of Wilhelm Martin Leberecht de Wette (1780–1849).5 Under these conditions, four fundamental coordinates of a theology of the Pentateuch must be taken into consideration: firstly, the names and epithets of God as the main character of the Pentateuch (the personage); secondly, the compositional and narrative profile of the Pentateuch (the dramaturgy); thirdly, the forms and traditions of the Pentateuch (the encyclopedia); and, fourthly, the appropriation of the theology of the Pentateuch by readers (the participation and reception). 4.1 The Personage A theology can be understood as an argumentative unfolding of the meanings contained in the name of a god, the horizon of meaning opened by this name, its historical background, and the nameʼs narrative and functional contexts. The 3  For extensive details, see E. B lum , Studien zur Komposition des Pentateuch (BZAW 189; Berlin: de Gruyter, 1989); T. Römer, “Hauptprobleme der gegenwärtigen Pentateuchforschung,” TZ 60 (2004), 289–307. 4  R. G. K ratz , “Noch einmal: Theologie im Alten Testament,” in Vergegenwärtigung des Alten Testaments: Beiträge zur biblischen Hermeneutik; Festschrift Rudolf Smend (ed. C. Bultmann et al.; Göttingen: Vandenhoeck & Ruprecht, 2002), 310–326, here 325. 5  “Glücklich waren unsere Alten, die, noch unkundig der kritischen Künste, treu und ehrlich alles das selbst glaubten, was sie lehrten! Die Geschichte verlor, aber die Religion gewann! – Ich habe die Kritik nicht angefangen; da sie einmal ihr gefährliches Spiel begonnen hatte, so mußte es durchgeführt werden, denn nur das Vollendete in seiner Art ist gut”; see W. M. L. de Wette, Beiträge zur Einleitung in das Alte Testament: Zweiter Band (Halle: Schimmelpfennig, 1807), 408.

1112

Markus Witte

beginnings of theology thus understood can be found in the combination of the divine designation ‫ אלהים‬with the tetragrammaton in Gen 2:4–3:24, the interpretation of the tetragrammaton by means of the root ‫ היה‬in Exod 3:14–15, the linking of the divine designations ‫ יהוה‬and ‫ אל שדי‬in accordance with the Priestly imagination of a graded revelation of YHWH in Exod 6:2–3 (cf. Gen 17:1), or in a complex manner in the formula of grace in Exod 34:6–7. Exodus 34:5–8 is an exemplary model of an innerbiblical theology that conforms to the criteria mentioned above and lends itself particularly well to developing coordinates for the theology of the Pentateuch described in this paper.6 Hence an account of the theology of the Pentateuch must contain a religious history of the names of God in the Pentateuch. The consideration of the epithets and titles attributed to God contributes to a further religious-historical development of the composition of speech about God. The phenomenon of the “Kompetenzausweitung JHWHs”7 (the expansions of YHWHʼs domain of competence), which characterizes the whole history of YHWH worship, can be illustrated by the attribution of certain epithets, such as the designation of YHWH as judge in Gen 18:25 or as king in Num 23:21. 4.2 The Dramaturgy The narrative and rhetorical profile of the Pentateuch is framed by the speeches of God in Gen 1:3 and Deut 34:4 and presents itself in long passages of great divine speech. According to this profile, we find different fundamental roles of God: as creator in the primeval history, as companion in the patriarchal history, as liberator in the exodus narrative, as the Holy One in the establishment of the cult, and, last but not least, as unequalled legislator and teacher in the testimony of Moses in Deuteronomy.8 These roles indicate fundamental theological concepts and entire theologies. We can assign different themes to the various roles that run through the whole Pentateuch and its theology, such as the human person, the land, the people, or the covenant. Finally, the narrative dialectic of life and death, blessing and curse, revelation and hiddenness, and punishment and forgiveness forms essential points of reference for an account of the theol See also R. S coralick , Gottes Güte und Gottes Zorn: Die Gottesprädikationen in Ex 34,6f. und ihre intertextuellen Beziehungen zum Zwölfprophetenbuch (HBS 33; Freiburg: Herder, 2002); M. F ranz , Der barmherzige und gnädige Gott: Die Gnadenrede vom Sinai (Exodus 34,6–7) und ihre Parallelen im Alten Testament und seiner Umwelt (BWANT 160; Stuttgart: Kohlhammer, 2003). 7  On this term, see B. J anowski , “Der Gott Israels und die Toten: Eine religions- und theologiegeschichtliche Skizze,” in JHWH und die Götter der Völker: Symposion zum 80. Geburtstag von Klaus Koch (ed. F. Hartenstein and M. Rösel, Neukirchen-Vluyn: Neukirchener Verlag, 2009), 99–138, here 104–105. 8  See also G. Fischer, Theologien des Alten Testaments (NSKAT 31; Stuttgart: Katholisches Bibelwerk, 2012). 6

Methodological Reflections on a Theology of the Pentateuc

1113

ogy of the Pentateuch. At the center of its content is the vitality and omnipotence of YHWH, which is both presupposed and proclaimed. 4.3 The Encyclopedia Individual theologies in the Pentateuch can be characterized as priestly, prophetic, sapiential, or legal on the basis of their forms (genres) and traditions. Even if sociological differentiation between the various schools of the Pentateuch tradition has proven difficult to demonstrate (so too for the wider Old Testament tradition), the classification of priestly, prophetic, and legal authors and transmitters introduced in the nineteenth century has fundamentally proven its worth by its ability to distinguish between the various theologies in the Pentateuch.9 In contrast, the binary opposition between the (priestly-)theocratic and the (prophetic-)eschatological10 is less suitable, as particular eschatological conceptions can also be theocratic. In light of the functional overlap of priests and prophets in ancient Israel, one might rather speak of cultic or cult and ritualistically oriented theology instead of priestly theology and eschatologically oriented theology instead of prophetic theology. Previous Pentateuch scholarship, especially Protestant research of the nineteenth century, favored the prophetic narrator over the priestly narrator. Retention of the old terminology need not imply such an evaluation. 4.4 The Participation and Reception The framing element of participation and reception arises from a threefold background: Firstly, the Pentateuch, or certain passages within it, are already quoted and interpreted in scriptures that only later became canonical.11 Secondly, the Pentateuch itself addresses its readers directly, particularly by means of the great prayers of Exod 15 and Deut 32, the blessings in Gen 49 and Num 23–24*, or the address to the current Israel in Deuteronomy. Thirdly, receptions of the Pentateuch, whether in early Jewish literature or medieval painting, 9  On the history of research, see M. Witte, Die biblische Urgeschichte: Redaktions- und theologiegeschichtliche Beobachtungen zu Genesis 1,1–11,26 (BZAW 265; Berlin: de Gruyter, 1998), 7–16. 10  See, e.g., H.-C. Schmitt, Theologie in Prophetie und Pentateuch: Gesammelte Schriften (ed. U. Schorn and M. Büttner; BZAW 310; Berlin: de Gruyter, 2001), 203–219, subsequent to the classical work of O. P löger, Theokratie und Eschatologie (WMANT 2; Neukir­chenVluyn: Neukirchener Verlag, 1959). 11  Here one is also reminded of the formula of grace from Exod 34:6–7, which appears in all sections of the Hebrew Bible and in the Dead Sea Scrolls and New Testament, or of the phenomenon of the (critical) reception of the Torah in late prophetic and wisdom texts (cf. Prov 2; Job 31); see B. S chipper and D. A. T eeter (eds.), Wisdom and Torah: The Reception of ‘Torah’ in the Wisdom Literature of the Second Temple Period (JSJSup 163; Leiden: Brill, 2013).

1114

Markus Witte

have themselves influenced later receptions and articulations of the theology of the Pentateuch. Here we can see the close systematic connection between an author-based and a reception-based hermeneutics. If the history of reception is deliberately integrated into accounts of the theological discourse conducted within the Pentateuch itself, we can, following Benjamin D. Sommer, speak of a “dialogical theology of the Pentateuch.”12

5 Models for the Presentation of the Theology of the Pentateuch The theology of the Pentateuch can be presented either as narrative or systematically or historically. These approaches can already be found within the Pentateuch and early Jewish literature. Given the multiplicity of theologies in the Pentateuch, it is also necessary to inquire into the unity of the theology of the Pentateuch. 5.1 A Narrative Presentation Nehemiah 9 offers an innerbiblical model for a primarily narrative presentation. Beginning with a doxology and the profession of YHWHʼs uniqueness, the poet proceeds in poetic-theological miniatures of the primeval history, the patriarchs, the exodus, Sinai, and the settlement of Israel. He then continues with the history of Israel in the land up until his own time. These prayers, composed in historical perspective, shape both the key words of the Pentateuch tradition13 and the hymnic predications of YHWH.14 Thus the narrative baseline of this presentation becomes enriched by systematic and doxological elements. From the choice of these key words and the individual hymnic predictions, a theology of the Pentateuch can be recognized in the garb of the theology of the author of Neh 9. This means that Neh 9 mediates a theology of the Pentateuch that is poetically renarrated, systematically composed, and doxologically structured with a focus on the uniqueness, justice, and mercy of God (cf. Exod 34:6–7). Given the constant address of God as “you,”15 the designation of YHWH as “our  B. D. Sommer, “Ein neues Modell für Biblische Theologie,” in Theologie und Exegese des Alten Testaments/der Hebräischen Bibel: Zwischenbilanz und Zukunftsperspektiven (ed. B. Janowski; SBS 200; Stuttgart: Katholisches Bibelwerk, 2005), 187–211, here 196–211. 13  Cf. ‫( בחר‬Neh 9:7; cf. Deut 4:37; 10:5), ‫( יצא‬Neh 9:7; cf. Gen 15:7), ‫( אמן‬Neh 9:8; cf. Gen 15:6), ‫( ברית‬Neh 9:8; cf. Gen 15:18; 17:2), ‫( אות‬Neh 9:10; cf. Exod 4:8–9, 28; 7:3; Deut 6:22), ‫( מופת‬Neh 9:10; cf. Exod 4:21; 7:3; Deut 6:22), ‫( נפלאות‬Neh 9:17; cf. Exod 3:20; 34:10), ‫( תורה‬Neh 9:13, 26; cf. Exod 24:12; Deut 4:8), ‫( צוה‬Neh 9:14; cf. Deut 4:5, 14, 40; 6:1, respectively), ‫( זכר‬Neh 9:17; cf. Ex 13:3; 7:18; 8:2). 14  Cf. Neh 9:6 (‫ ;)יהוה לבדך‬9:7 (‫ ;)אתה הוא יהוה האלהים‬9:8 (‫ ;)צדיק אתה‬9:10 (‫ ;)שם‬9:17 (// Exod 34:6–7); 9:19 (‫ ;רחם‬cf. 9:27, 31); 9:20 (‫ ;)רוחך הטובה‬9:32 (‫)האל הגדול‬. 15  Cf. Neh 9:6, 7, 8, 17, 19, 20. 12

Methodological Reflections on a Theology of the Pentateuc

1115

God,” and the conclusion in vv. 32–37 written in the first-person plural, this theology is directed toward those who read and pray the text of Neh 9. It thus aims to be applied and integrated into their life and faith. Within the scope of modern accounts of theology in the Pentateuch, Gerhard von Radʼs Theology of the Old Testament offers the classical articulation of a narrative approach, as prefigured by Neh 9 (and other so-called historical Psalms),16 in the form of “renarration” (Nacherzählung).17 The strength of a narrative approach lies, firstly, in its close alignment with the literary composition of the Pentateuch and the literary forms used in it; secondly, in its ability to take all texts of the Pentateuch into equal consideration; and, thirdly, in its accessibility to readers. For in this approach the “narrative thread” (Erzählfaden) of the Pentateuch constitutes the internal theological coherence. A weakness of the narrative approach arises with respect to its historical dimension, namely its systematic synchrony and taxonomy. To avoid excluding literary and religious-historical findings, the literary and religious-historical level on which the renarrator is located must be specified. Diachronic and longitudinal perspectives need to be incorporated alongside synchronic and systematic overviews. 5.2 A Systematic Presentation An innerbiblical precursor for a more systematically focused theology of the Pentateuch can be found in the description of the activity of divine wisdom (σοφία) in Wisdom of Solomon 10:1–11:1. Wisdom of Solomon 10 exhibits a narrative profile insofar as its author adheres to the chronology of the Pentateuch by selecting the righteous from Adam to Moses. However, the hypostasis-like σοφία constitutes a fundamental and systematic means of interpretation that is external to the text. It represents and interprets the activity of God as it is described from Genesis to Exodus. The narrative-systematic theology of the Pentateuch from Wis 10 is comparable to Ben Siraʼs praise of the fathers (Sir 44:1–49:16). Even this poetic summary of history basically follows the outline of the Pentateuch but focuses on chosen heroes from Enoch to Joshua (and, beyond him, up to Nehemiah).18 The overall theological concept internal to the text – that is to say, taken from the Pentateuch itself – is in this context the terms

16  Cf. Pss 78; 105; 106; 135; 136, and J. Gärtner, Die Geschichtspsalmen: Eine Studie zu den Psalmen 78, 105, 106, 135 und 136 als hermeneutische Schlüsseltexte im Psalter (FAT 84; Tübingen: Mohr Siebeck, 2012); S. Ramond, Les leçons et les énigmes du passé: Une exégèse intra-biblique des psaumes historiques (BZAW 459; Berlin: de Gruyter, 2014); and M. Witte, Von Ewigkeit zu Ewigkeit: Weisheit und Geschichte in den Psalmen (Biblisch-Theologische Studien 146; Neukirchen-Vluyn: Neukirchener Verlag, 2014). 17  G. von Rad, Theologie des Alten Testaments (2 vols.; Munich: Kaiser, 1957–1960). 18  In contrast to the Wisdom of Solomon, which consciously abstains from denominating the heroes, Ben Sira introduces the figures by name in his laus patrum.

1116

Markus Witte

‫‌ברית‬19 and διαθήκη20. A similar concept is found in the historical summary of the Damascus Document (CD III)21 or in the letter to the Hebrews (ch. 11), when they treat the theological view of time from the creation up to the prophets under the umbrella term πίστις. Wisdom of Solomon and Ben Sira anticipate a theological overview focused on the work of wisdom, the foundation of a covenant, or the notion of faith in modern Christian and Jewish presentations of the theology of the Pentateuch.22 The classic example of this is Walter Eichrodtʼs Theology of the Old Testament.23 Even accounts that, following the old dogmatic loci method, center less on a particular theological concept and more on an overview of themes such as God, human, creation, sin, election, salvation, or justification belong to the category of systematic theologies of the Pentateuch. There are many examples of this approach in recent research.24 Insofar as they attempt to locate the inner unity of the theology of the Pentateuch, they describe it in relation to terms such as justice or faith or by means of correlations such as electing and obligating or promising and fulfilling.25 The strength of such an approach lies in its ability to relate clearly individual texts of the Pentateuch to fundamental and overarching issues for theology as a whole and for other humanities and cultural studies. A theology that strives to be applicable usually also orients itself more readily around a systematic than a narrative presentation. The weakness of a systematic approach becomes evident in the application of nontextual categories for the classification and selection of the texts from the Pentateuch. The selection of  Sir (H) 44:12, 17, 20, 22; 45:15, 24, 25.  Sir (G) 44:11, 18, 20, 22; 45:5, 7, 17, 24, 25; 47:11. 21  In doing so, CD III not only uses the term ‫ ברית‬as a leitmotif in the historical overview from Noah to the present but also assigns the theologically programmatic titles “creator” (‫)עשה‬ and “teacher” (‫)יורה‬, by which the whole Pentateuch is framed, to God (CD III.8). 22  On the reception of the term faith as the title for a history of religion and a theology of the Hebrew Bible, see, e.g., M. Buber, Der Glaube der Propheten (Heidelberg: Lambert Schneider, 1950); S. Gesundheit, “Gibt es eine jüdische Theologie der hebräischen Bibel?,” in Theologie und Exegese des Alten Testaments/der Hebräischen Bibel: Zwischenbilanz und Zukunftsperspektiven (ed. B. Janowski; SBS 200; Stuttgart: Katholisches Bibelwerk, 2005), 73–86, here 86; and Schmitt, Theologie (see n. 10), 220–237. 23  W. Eichrodt, Theologie des Alten Testaments (3 vols.; Leipzig: Hinrichs, 1933, 1935, 1939). 24  See, e.g., R. E. Clements, Old Testament Theology: A Fresh Approach (London: Marshall, Morgan & Scott, 1978); J. Schreiner, Theologie des Alten Testaments (NEB 1; Würzburg: Echter, 1995); W. Brueggemann, Theology of the Old Testament: Testimony, Dispute, Advocacy (Minneapolis: Fortress, 1997); R. R endtorff , Theologie des Alten Testaments: Ein kanonischer Entwurf, Bd. 2; Thematische Entfaltung (Neukirchen-Vluyn: Neukirchener Verlag, 2001). 25  See, e.g., C. Westermann, Theologie des Alten Testaments in Grundzügen (GAT 6; Göttingen: Vandenhoeck & Ruprecht, 1978), and H. D. Preuss, Theologie des Alten Testaments (2 vols.; Stuttgart: Kohlhammer, 1991–1992). 19 20

Methodological Reflections on a Theology of the Pentateuc

1117

texts and the focus on individual systematic themes, which are at least implicit, often depend on dogmatic premises. Though the religious-historical dimension still can be raised in the treatment of the individual topoi, the systematic approach risks flattening out the literary-historical development and context. 5.3 A Historical Presentation The innerbiblical prototypes of a historical approach to the theology of the Pentateuch include the three-tiered concept of revelation in the Priestly writer, the confrontation between YHWH as the old and known God of Israel and the new and foreign gods in Deut 32 (cf. Josh 24), and the conversion narrative of Abraham in Jub. 12. These texts offer historical approaches to speech about God, though in the sense not of a genetic derivation but rather of a history of religious insight enabled by means of revelation. Modern examples of attempts to transfer this history of revelation in the Pentateuch to a history of religion of Israel include the theses concerning the God of the fathers26 and the origins of the worship of YHWH at Sinai.27 Where a theology of the Pentateuch is arranged primarily along historical lines, it must free itself from the equation of biblical Israel with historical Israel on the one hand and the narrative revelation of YHWH with the history of the religion of YHWH on the other.28 A theology of the Pentateuch that is primarily oriented according to history is – just as a theology of the Hebrew Bible or a theology of the Old Testament – neither identical with nor an alternative to a religious history of Israel and early Judaism. The essential characteristic of a historically oriented theology is the explication of the gradual development that characterizes the speech about God in the Pentateuch. Thus a historically oriented theology of the Pentateuch presents the development of speech about God at the level of critically reconstructed as well as literary and theological profiles. Thereby, the history of redaction becomes the history of theology on the basis of the literary-historical conditions in the Pentateuch; in particular, it becomes a history of theology of early Judaism.29 The works of Christoph Levin and John Van Seters are examples of this, even though they do not yet fully extend to the entire Pentateuch.30 26  Cf. A. Alt, Der Gott der Väter: Ein Beitrag zur Vorgeschichte der israelitischen Religion (BWANT 48; Stuttgart: Kohlhammer, 1929). For refutation of this thesis, see M. Köckert, Vätergott und Väterverheißungen: Eine Auseinandersetzung mit Albrecht Alt und seinen Erben (FRLANT 142; Göttingen: Vandenhoeck & Ruprecht, 1988). 27  See the issue entitled “Anfänge und Ursprünge der Jahwe-Verehrung,” BTZ 30/1 (2013). 28  See R. G. K ratz , Historisches und biblisches Israel: Drei Überblicke zum Alten Testament (Tübingen: Mohr Siebeck, 2013), 141–143. 29  See Kratz, “Noch einmal” (see n. 4), 322. 30  J. Van S eters , Prologue to History: The Yahwist as Historian in Genesis (Zürich: Theologischer Verlag, 1992); idem, The Life of Moses: The Yahwist as Historian in Exodus–

1118

Markus Witte

The strengths of a historical approach lie in the fact that the reader is included in the history of God-talk in the Pentateuch and that the theological emphasis of individual redactions can be brought into conversation with analogs in the parabiblical and postbiblical early Jewish history of theology.31 The history itself appears here as an inner unity of theology. A difficulty arises in this approach when determining individual layers of its literary scope and dating their origins. By way of example, this is evident when we consider the question whether J, E, D, or P is the oldest literary and theological layer identified in the Pentateuch (which is being revisited in current research) and in which redaction-historical, and thereby theological-historical, relation the supposed layers stand relative to each other.32 I recall in particular the epoch-making statement that the Priestly writer presumes the Deuteronomic demand for unity of the cult33 or that the concept of God in the so-called J source is latently monotheistic and post-Deuteronomic or post-Deuteronomistic.34 The possibility of transferring the history of theology into a theology that meets the aforementioned criteria remains open in a purely historical approach. However, a decidedly nontheological approach to history or an approach that consciously avoids the concept of theology is never without (at least an implicit) theology.35

Numbers (CBET 10; Kampen: Kok Pharos, 1994); C. L evin , Der Jahwist (FRLANT 157; Göttingen: Vandenhoeck & Ruprecht, 1993); idem, Re-Reading the Scriptures: Essays on the Literary History of the Old Testament (FAT 87; Tübingen: Mohr Siebeck, 2013), 1–23; for such a program, see also Kratz, “Noch einmal” (see n. 4), 320–325. 31  See Sommer, “Ein neues Modell” (see n. 12), 203–208, for the theory, and Witte, Die biblische Urgeschichte (see n. 9), for an example. 32  For the state of discussion, see C. H outman , Der Pentateuch: Die Geschichte seiner Erforschung neben einer Auswertung (CBET 9; Kampen: Kok Pharos, 1994); J. C. Gertz et al. (eds.), Abschied vom Jahwisten: Die Komposition des Hexateuch in der jüngsten Diskussion (BZAW 315; Berlin: de Gruyter, 2002); T. B. Dozeman and K. Schmid (eds.), A Farewell to the Yahwist? The Composition of the Pentateuch in Recent European Interpretation (Society of Biblical Literature Symposium Series 34; Atlanta: Society of Biblical Literature, 2006); T. B. Dozeman et al. (eds.), The Pentateuch: International Perspectives on Current Research (FAT 78; Tübingen: Mohr Siebeck, 2011); J. C. G ertz et al., T&T Clark Handbook of the Old Testament: An Introduction to the Literature, Religion and History of the Old Testament (London: T&T Clark, 2012), 252–263. 33  J. W ellhausen , Geschichte Israels in zwei Bänden (2 vols.; Berlin: Reimer, 1878), 1:106–107; after the second edition, titled Prolegomena zur Geschichte Israels (Berlin: Reimer, 1883). 34  See H. H. Schmid, Der sogenannte Jahwist: Beobachtungen und Fragen zur Pentateuch­ forschung (Zürich: Theologischer Verlag, 1976), 154–183. On the difference between Dtn and Dtr and their relationship to each other, see Van Seters, Prologue (see n. 30), 128–129; 255–257, 262–265, 332, and Levin, Der Jahwist (see n. 30), 430–436. 35  Cf. the short historical review on the juxtaposition of Julius Wellhausen and Bernhard Duhm by Smend, Die Mitte (see n. 2), 105–110.

Methodological Reflections on a Theology of the Pentateuc

1119

5.4 A Synthetic Presentation An ideal theology of the Pentateuch would combine the narrative, systematic, and historical approaches. In view of the four coordinates mentioned above, this means that a theology of the Pentateuch first of all describes the names and epithets of God in the prolegomena. It would then develop each role of God – God as creator, companion, liberator, holy one, and unequalled teacher and legislator – in a separate paragraph according to narrative development. By doing so, the theology would describe the theologies of the corresponding layers and redactions in accordance with literary-historical insights. Finally, it would systematically evaluate the motifs and themes that shape the so-called final form and the reception history of the text in terms of the underlying conceptions of the divine and the human and their potential to interpret present-day life. Thus a theology of the Pentateuch would amount to an interpretation of the meaning of life from the perspective of how the dramaturgy and encyclopedic elements (the forms and traditions) of the Pentateuch play out. This theology would be prescriptive not in the sense of a norm but in the sense of an interpretation and application. Consequently, each chapter of such a theology would contain three sections: a renarration, a historical layering, and a systematic synthesis. Narrative, historical, and systematic relationships among the blocks of tradition in the Pentateuch must be made explicit by appropriate cross-references within and among traditions.36 5.5 The Pentateuch as a Part of the Hebrew Bible and Old Testament The Pentateuch/Torah is but a part of the Hebrew Bible and Old Testament and framed by a larger canon. Therefore, a theology of the Pentateuch/Torah does not stand on its own but interacts with the theology in and the theology of both the Jewish Bible and the Christian Bible. Within the canon of the Hebrew Bible, the theologies of the Torah, Neviʾim, and Ketuvim complement each other, interpret each other, and ultimately constitute a multivoiced critical Jewish theology. The Hebrew Bible finds its direct continuation, interpretation, and transformation in early Jewish literature, in the Talmud, and in the midrash. As part of the Christian Old Testament, the theology of the Pentateuch is an essential component for understanding both the theology of the New Testament and Christian theology.37 As an alternative to early church and medieval hermeneutics, as well as modern attempts at christocentric or christotelic interpretations,38 one could employ a  Cf. O. Kaiser, Der eine Gott Israels und die Mächte der Welt: Der Weg Gottes im Alten Testament vom Herrn seines Volkes zum Herrn der ganzen Welt (FRLANT 249; Göttingen: Vandenhoeck & Ruprecht, 2013). 37  See Rendtorff, Theologie (see n. 24), 311. 38  See P. E nns , Ecclesiastes (Two Horizons Old Testament Commentary; Grand Rapids, MI: Eerdmans, 2011), 27–29. 36

1120

Markus Witte

hermeneutics in which the theology/theologies of the Pentateuch are outlined according to the potential permeability of New Testament understandings of God, the world, and humanity. Structural correspondences and parallels in concepts and motifs as well as tradition-historical connections could also be presented. Such a “christotransparent approach”39 aims to illuminate the various theologies of the Pentateuch that shape the New Testament understanding of Jesus as Christ, lord, and God. From the point of view of the New Testament, the theologies of creation, history, law, cult, and wisdom gathered in the Pentateuch reach their goal (τέλος) when they concretely enter space and time and become personal through Godʼs activity in Jesus Christ. In this case, the theology of the Pentateuch contributes to Christology.

6 Conclusion A survey of the theology of the Pentateuch belongs to the fundamental and enduring tasks of Pentateuch research. In the process of developing such a theology, the theological significance of the Pentateuch must be explicated, firstly, in its different canonical forms (with the Hebrew Bible and the Septuagint being the most important historical receptions and dialogue partners for the Torah); secondly, referring to its sources and redactions; and thirdly, referring to its role as Torah in the context of Judaism and as part of the Christian Bible consisting of Old and New Testaments. The strength of each theory of the development and meaning of the Pentateuch is also made evident by the elucidation of the theological currents that are reflected in the composition of the Pentateuch and by elucidation of the theological character of the Pentateuch.

39  See M. W itte , Jesus Christus im Alten Testament. Eine biblisch-theologische Skizze (Salzburger Exegetische Theologische Vorträge 4; Berlin: LIT, 2013).

The Dynamic of Closure in the Pentateuch Jean-Pierre Sonnet “The end writes the beginning and shapes the middle” P. Brooks1

Much of our understanding of the Pentateuch depends on our perception of its narrative integrity: is it a self-contained work, marked off by signs of closure, and, if so, by what kind of closure? In this essay, I intend to show that the Pentateuch is underlined and driven by a double dynamic of closure, which leads to the completion of the sanctuary (‫ )משכן‬in Exod 39–40 on the one hand (“the sons of Israel did according to all that YHWH had commanded Moses”; Exod 39:32; see 39:42–43), and, on the other hand, to the peopleʼs obedience to Godʼs commandments, reported by the narrator in Deut 34:9: “And the sons of Israel listened to him [Joshua] and did as YHWH had commanded Moses.” The Pentateuch does not close without stating, in two stages, that Godʼs offer has been appropriately received by the people in auspicious beginnings, whatever the uncertainties of the ensuing history. Biblical scholarship usually emphasizes the incomplete character of the pentateuchal plot. We all have in mind David Clinesʼs words about “the partial fulfillment [of the Pentateuch] – which implies the partial non-fulfillment of the promises to or blessing of the patriarchs”:2 the inheritance of the promised land is narrated beyond the limits of the Pentateuch, in the book of Joshua. Earlier observations of this fact led to the hypothesis of the Hexateuch,3 introduced by Ewald, canonized by von Rad,4 and revisited in many different (and reasonable) ways in current scholarship.5 For some scholars, the open-endedness of the  P. Brooks, Reading for the Plot: Design and Intention in Narrative (New York: Knopf, 1984), 22. 2  D. J. A. Clines, The Theme of the Pentateuch (2nd ed.; JSOTSup 10; Sheffield: Sheffield Academic, 1997), 30. 3  See G. H. A. Ewald, Die Geschichte des Volkes Israel (2 vols.; 3rd ed.; Göttingen: Vandenhoeck & Ruprecht, 1864). 4  See G. von Rad, The Problem of the Hexateuch and Other Essays (New York: McGrawHill, 1966), 1–78. 5  See, in particular, the essays collected in T. Römer and K. Schmid (eds.), Les dernières rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (BETL 203; Leuven: Peeters, 2007). 1

1122

Jean-Pierre Sonnet

Pentateuch mirrors the historical situation of the intended audience of the book, called to enter – or to reenter – the land after exile.6 According to others, the Pentateuch indeed embodies an idea of Judaism in exile or in diaspora, for which “it is possible to belong to the people of Israel without living in the Promised Land.”7 For still other scholars, the openness of the pentateuchal narrative is meant to generate suspense and existential expectations: “Will Israel live by this Torah or not? Will Israel choose life or death (Deut 11:26; 30:19)? The Pentateuchʼs inconclusive plot reinforces its rhetorical strategy of making readers decide its ending by their own actions.”8 The program defined in the Pentateuch, Thomas Mann writes, “exists only as a possibility in terms of what Israel will do.”9 Disturbingly enough, the possibility in question is turned into an impossibility by some Christian interpreters, as if the Pentateuch, with its Deuteronomic ending, were a narrative theory of Israelʼs inability to comply with the Torah.10

6  See, for instance, W. Brueggemann, Theology of the Old Testament: Testimony, Dispute, Advocacy (Minneapolis: Fortress, 1997), 210: “the community waiting at the brink of the land of promise in Joshua 1, just outside the Pentateuch, is not an ancient community of Moses; it is sixth-century Israel, waiting in anticipation for an entry (reentry) into a land from which they had been displaced by the geopolitics of a Babylonian world.” 7  See J. L. Ska, Introduction to Reading the Pentateuch (Winona Lake, IN: Eisenbrauns, 2006), 14: “The Promised Land, and not the possession of it, constitutes a fundamental element of Israelʼs faith. In other words, according to the Pentateuch, it is possible to belong to the people of Israel without living in the Promised Land. This sort of assertion is more understandable after the experience of the Exile and at the time of the diaspora.” 8  J. W. Watts , Reading Law: The Rhetorical Shaping of the Pentateuch (BibSem 59; Sheffield: Sheffield Academic, 1999), 154. In the same vein, W. J. Houston, The Pentateuch (London: SCM, 2013): “The Pentateuch ends without closure. It ends with an as yet unfulfilled expectation that Israel will enter the land, and with questions unanswered, especially the questions whether Israel is capable of being faithful to their God and whether their God will permit them to keep the land” (20), and “The readers or hearers are left with a question that they can only answer by themselves, whether they will be faithful to their God and obedient to the divine commands for the good life” (40). 9  T. M ann , The Book of the Torah: The Narrative Integrity of the Pentateuch (Atlanta: John Knox, 1988), 161. 10  P. B eauchamp , Lʼun et lʼautre Testament: Essai de lecture (Paris: Seuil, 1976), 72–73, discerns in Deuteronomy “lʼimpossibilité de lʼhomme à se conformer à la Loi du Dieu Saint [. . .]. Le Deutéronome approche (sans y atteindre) de la découverte dʼune incapacité radicale à observer la Loi.” Along the same lines, see R. Bultmann, Essays on Old Testament Hermeneutics (ed. C. Westermann; Richmond, VA: John Knox, 1963), 73–75: “How far, then, does Old Testament Jewish history represent prophecy fulfilled in the history of the New Testament community? It is fulfilled in its inner contradiction, its miscarriage” (73); “Faith requires the backward glance into Old Testament history as a history of failure, and so of promise” (75).

The Dynamic of Closure in the Pentateuch

1123

1 The Reception of the Covenantal Words in Deuteronomy and in the Pentateuch In my view, the Pentateuch calls for closer attention to the way it rounds off its inner plot. The closure of a narrative is much more than its ending – that is, than the completion of its final sentence. The closure influences the entire narrative. Peter Brooks writes, “The end writes the beginning and shapes the middle.”11 And Ann Jefferson echoes him: “Narrative is more than a mere collection of motifs, devices and functions, more than a simple sequence of actions; it is an end-dominated collection or sequence, and this domination of the end will affect the reading of all that precedes it.”12 This is very much the case in a fivefold book – the canonical Pentateuch – that opens with the words “In the beginning.” In a first step, I will consider the influence of a verse – actually a half verse – that is generally overlooked, Deut 34:9b: “And the sons of Israel listened to him [Joshua] and did as YHWH had commanded Moses.” The listening and compliance of the sons of Israel – that is, of Joshuaʼs generation – is the last action in Deuteronomy before the flashback of 34:10–12. As I have tried to show in recent articles, the report in question provides a decisive key in the understanding of Deuteronomyʼs narrative economy.13 The report by the narrator actually brings to a close Deuteronomyʼs overall plot of communication in a tight correspondence with Deut 1:3: Deuteronomy 1:3

Deuteronomy 34:9

Moses spoke to the sons of Israel according to all that YHWH had commanded him unto them (‫ככל‬ ‫)אשר צוה יהוה אתו אלהם‬.

And the sons of Israel listened to him [Joshua] and did as YHWH had commanded Moses (‫ויעשו‬ ‫)כאשר צוה יהוה את משה‬.

Deuteronomy 1:3 announces the success of Mosesʼs teaching, as far as the speaker is concerned: “Moses spoke to the sons of Israel according to all that YHWH had commanded him unto them” (‫)ככל אשר צוה יהוה אתו אלהם‬. In Deut 34:9, the narrator makes clear that the reception of Mosesʼs communication is equally successful: “And the sons of Israel listened to him [Joshua] and did as YHWH had commanded Moses” (‫וישמעו אליו בני ישראל ויעשו כאשר צוה‬  Brooks, Reading (see n. 1), 22.  A. J efferson , “Mise en abyme and the Prophetic of Narrative,” Style 17 (1983), 196–208, here 196. 13  See J.-P. Sonnet, “Redefining the Plot of Deuteronomy – From End to Beginning: The Import of Deut 34:9,” in Deuteronomium: Tora für eine neue Generation (ed. G. Fischer et al.; BZABR 17; Wiesbaden: Harrassowitz, 2011), 37–49; idem , “The Fifth Book of the Pentateuch: Deuteronomy in Its Narrative Dynamic,” Journal of Ancient Judaism 3 (2012), 197–234; idem, “If-Plots in Deuteronomy,” VT 63 (2013), 453–470. 11

12

1124

Jean-Pierre Sonnet

‫)יהוה את משה‬.14 Mosesʼs teaching of the Torah on the last day of his life, it is affirmed, has not been in vain.15 Both verses imply the narratorʼs authority and omniscience: “Moses spoke [. . .] according to all that YHWH had commanded him” (1:3); “The sons of Israel did as YHWH had commanded Moses” (34:9). In the epistemological model of biblical narrative, this report thus stands at the highest level of reliability.16 Yet what Deuteronomyʼs narrative arc most emphasizes is actually a major shift in the communication process: it is not Moses who is heeded but Joshua; it is through Joshuaʼs mediation that the Mosaic Torah is brought home. The transfer of agency in question brings out the core of Deuteronomyʼs overall plot as a story of succession and delegation. This plot of delegation and succession is evident from Mosesʼs first act, the appointment of wise judges in Deut 1:9–18, up to the provisions about the public offices (in Deut 17–18), including the promise of the prophet like Moses (17:9–22).17 The narratorʼs report in Deut 34:9 drives us back to the figure of the prophet like Moses. The report in question – “The sons of Israel listened to him [= Joshua] (‫ )וישמעו אליו‬and did as YHWH had commanded Moses” – actually represents the fulfillment of Mosesʼs injunction about the prophet in question mentioned in 18:15, “To him you will listen (‫)אליו תשמעון‬,”18 and this in a sharp contrast to the warning against the prophet apostate of Deut 13:1–5: “you will not listen (‫ )לא תשמע‬to the words of that prophet or dreamer of dreams” (v. 4). In his way of conveying Godʼs requirements and of bringing about the peopleʼs obedience, Joshua plays the role of the prophet in question – possibly as the first figure in a sequence.19 14  About the centrality of the representation of reception in biblical narrative, see M. Sternberg, “The World from the Addresseeʼs Viewpoint: Reception as Representation, Dialogue as Monologue,” Style 20 (1986), 295–318, here 308–309. 15  The modal “as” (‫ )כאשר‬in Deut 34:9 echoes the distributive “according to all” (‫)ככל אשר‬ in 1:3. What Moses had to spell out piecemeal can be implemented per modum unius by the people, as recorded by the narrator in his proleptic summary in 34:9. 16  See R. Alter, The Art of Biblical Narrative (2nd ed.; New York: Basic Books, 2011), 146–147; see also M. Sternberg, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading (Bloomington: Indiana University Press, 1985), 380–382. 17  See Sonnet, “Fifth Book” (see n. 13), 229–232. 18  Ibn Ezra (c. 1092–1167) has noticed the echoing effect between Deut 18:15 and 34:9, and has accordingly identified Joshua as the promised prophet (see his comment on Deut 18:15, taken up by Abraham Hazquni [born in 1627]); see also E. O tto , Die Tora: Studien zum Pentateuch; Gesammelte Aufsätze (BZABR 9; Wiesbaden: Harrassowitz, 2009), 209. 19  To go ahead in the historical books is to find out that Joshua is actually the first (and the exemplary) figure of a series of prophets sent by YHWH – “my servants, the prophets” (2 Kgs 17:13, 23; 21:10; 24:2) – in order to warn the people so that they might avoid the catastrophe of the exile; the last representative of the series is apparently Jeremiah (see Jer 7:25; 25:4; 26:5; 29:19 MT; 35:15; 44:4). On this issue, see J. B lenkinsopp , Prophecy and Canon: A Contribution to the Study of Jewish Origins (Notre Dame: University of Notre Dame Press, 1977), 39–53; J.-P. Sonnet, The Book within the Book: Writing in Deuteronomy (BibInt 14;

The Dynamic of Closure in the Pentateuch

Deuteronomy 1:3

1125

Deuteronomy 13:1–5 Deuteronomy 18:15–22 Deuteronomy 34:9 (the “prophet like (the prophet Moses”) apostate) “You will not listen (‫ )לא תשמע‬to the words of that prophet or dreamer of dreams.” 5 “To his [YHWHʼs] 15“To him voice you shall listen you will listen (‫אליו‬ (‫)ובקלו תשמעו‬.” ‫)תשמעון‬.” 20 “But the prophet who presumes to speak in my name a word that I have not commanded him (‫ )אשר לא צויתיו‬to speak [. . .] – that prophet shall die.” 4

Moses spoke to the sons of Israel according to all that YHWH had commanded him unto them (‫ככל אשר‬ ‫)צוה יהוה אתו אלהם‬.

And the sons of Israel listened (‫)וישמעו‬ to him [Joshua] and did as YHWH had commanded Moses (‫ויעשו כאשר צוה יהוה‬ ‫)את משה‬.

The peopleʼs obedience, however, comes as a surprise. The horizon of Mosesʼs teaching has been obscured by dire predictions about the peopleʼs unfaithfulness, which culminate in Deut 31: “For I know well how rebellious and stubborn you are. If you already have been so rebellious toward YHWH while I am still alive among you, how much more after my death!” (v. 27). Yet, to the readerʼs surprise, the last action by the people told of in Deuteronomy – listening to Joshua and compliance with the law taught by Moses – springs not from the peopleʼs unfaithfulness but from their faithfulness.20 Notwithstanding the gloomy future Leiden: Brill, 1997), 196–197; C. Nihan, “ ‘Un prophète comme Moïse’ (Deutéronome 18,15): Genèse et relectures dʼune construction deutéronomiste,” in La construction de la figure de Moïse: The Construction of the Figure of Moses (ed. T. Römer; TranseuSup 13; Paris: Gabalda, 2007), 43–76, here 56–59; D. Markl, “Moses Prophetenrolle in Dtn 5; 18; 34: Strukturelle Wendepunkte von rechtshermeneutischem Gewicht,” in Deuteronomium: Tora für eine neue Generation (ed. G. Fischer et al.; BZABR 17; Wiesbaden: Harrassowitz, 2011), 51–68, here 65. The narratorʼs summary in 2 Kgs 17, however, reveals that the further prophets met a fate that contrasted with Joshuaʼs: the people “did not listen” to them (2 Kgs 17:14). The contrast highlights the deliberate refusal of the people in their nonlistening. 20  The post-Mosaic apostasy announced by God in Deut 31:16 (“Here you are about to lie with your fathers. And this people will go astray after the alien gods”) did not come true in (or as) the immediate aftermath of Mosesʼs demise; it actually took place in post-Mosaic history, after the time of remission represented by Joshuaʼs generation; see S onnet , “Redefining the Plot” (see n. 13), 43. The correction of the readerʼs – and apparently of Mosesʼs (see 31:27–29) – dire assumption is illustrative of a general law in narrative. E. Segal, “Closure in Detective Fiction,” Poetics Today 31 (2010), 156, writes: “As long as the end has not been reached,” our judgments about the nature and degree of the textʼs closure “would always be open to readjustment (including a complete reversal), because what is left of the text may

1126

Jean-Pierre Sonnet

to follow, the people have initially chosen “life” . . . as had also been announced by Moses in Deut 30:19–20: “you will choose life.” Moses then made clear that compliance with the divine commandment is within reach, against any possible objection that the command in question is too high or too far away, “so that we may listen to it and do it” (‫( )וישמענו אתה ונעשנה‬Deut 30:12–13). This is precisely what the report in Deut 34:9 makes clear: “they listened [. . .] and they did” (‫] ויעשו‬. . .[ ‫)וישמעו‬. It is rarely noticed that what is fulfilled here is actually the pledge by the people in Deut 5:27, when they requested the mediation by Moses: “All that YHWH our God will say, you will tell us, and we will listen and do it” (‫ושמענו‬ ‫)ועשינו‬. And so it came to pass in 34:9: ‫] ויעשו‬. . .[ ‫וישמעו‬, “they listened and did as YHWH had commanded Moses.” The commitment by the people in Deut 5 may have been presumptuous or naïve; it is nonetheless achieved in Deut 34:9, made possible by Mosesʼs teaching and by Joshuaʼs mediation. The report by the narrator in Deut 34:9b is obviously proleptic, anticipating what will be detailed in the book of Joshua. The half verse thus implies the passage to the land, the inheritance of the land, and the compliance with the commandments on the land – it is indeed on the land that the commandments come into force.21 Yet this thorough acquiescence and conforming is told within Deuteronomy, which does not come to a close without telling of the felicitous reception of Godʼs requirements, however threatening the future may prove. Deuteronomy 34:9b

‫וישמעו אליו בני ישראל ויעשו כאשר צוה יהוה את משה‬ And the sons of Israel listened to him [Joshua] and did as YHWH had commanded Moses.

contain further developments: what appears open may yet be closed, and what appears closed may yet be reopened.” About the remission of Joshuaʼs generation, see Judg 2:7: “The people worshiped YHWH all the days of Joshua, and all the days of the elders who outlived Joshua, who had seen all the great work that YHWH had done for Israel”; about its end, see Judg 2:10–12. Furthermore, Judg 2:17 clearly signals the end of the respite in question, in a narratorʼs statement that not only negates the verbs of the summary in Deut 34:9b but also echoes the announcement of the peopleʼs unfaithfulness in Deut 31 (vv. 16, 18, 20, 29): “Yet they did not listen even to their judges; for they prostituted themselves with other gods and bowed down to them. They soon turned aside from the way in which their ancestors had walked, who had obeyed the commandments of YHWH; they did not do so”; see W. Gross, Richter (HThKAT; Freiburg: Herder, 2009), 208. 21  See the comment by Rabbi Levi ben Gershom (fourteenth century) about Deut 34:9 (with an eye to Josh 11:15) in Mikra‍ʾot Gedolot Haketer (ed. M. Cohen; Ramat Gan: Bar-Ilan University Press, 2011), 255: “Although Moses didnʼt have the privilege to lead Israel in the inheriting of the land, everything that was done there was done according to his word” (my translation).

The Dynamic of Closure in the Pentateuch

1127

Joshua 1:16–17 ]. . .[ ‫] ככל אשר שמענו אל משה כן נשמע אליך‬. . .[ ‫ויענו את יהושע לאמר כל אשר צויתנו נעשה‬ They answered Joshua: “All that you have commanded us we will do [. . .]. Just as we listened to Moses in all things, so we will listen to you. [. . .]” Joshua 8:35

‫לא היה דבר מכל אשר צוה משה אשר לא קרא יהושע נגד כל קהל ישראל‬ There was not a word of all that Moses commanded that Joshua did not read before all the assembly of Israel. Joshua 11:12 (see also 11:20, 23) As Moses the servant of YHWH had commanded.

‫] כאשר צוה משה עבד יהוה‬. . .[

Joshua 11:15 ‫כאשר צוה יהוה את משה עבדו כן צוה משה את יהושע וכן עשה יהושע לא הסיר דבר מכל אשר‬ ‫צוה יהוה את משה‬ As YHWH had commanded his servant Moses, so Moses commanded Joshua, and so Joshua did; he left nothing undone of all that YHWH had commanded Moses.

If the listening and the obedience of the people reported in Deut 34:9 work as a powerful fulfillment within Deuteronomy, they operate similarly within the pentateuchal macro-narrative. The report by the narrator in Deut 34:9 – ‫וישמע‬ ‫] ויעשו‬. . .[ – echoes not only the commitment by the people in Deut 5:25 – ‫ושמענו‬ ‫ – ועשינו‬but also the solemn pledge by the people in Exod 24:7, at the end of the Sinai covenant: “All that YHWH has said, we will do, and we will listen” (‫)נעשה ונשמע‬:22 Exodus 24:7

‫כל אשר דבר יהוה נעשה ונשמע‬ “All that YHWH has spoken we will do, and we will listen.” Deuteronomy 5:27

‫ואת תדבר אלינו את כל אשר ידבר יהוה אלהינו אליך ושמענו ועשינו‬ “All that YHWH our God will say you will tell us, and we will listen and do it.” Deuteronomy 34:9b

‫וישמעו אליו בני ישראל ויעשו כאשר צוה יהוה את משה‬ [They] listened to him and did as YHWH had commanded Moses.

The commitment in Exod 24:7 was clearly solemn, yet it was not followed (within the limits of the book of Exodus) by a report by the narrator “they did and they listened”/“they listened and they did.” No similar report is found either

 Apropos of the structuring role of the formula in Exod 19–24 (in 19:8; 20:19; 24:3, 7), see J.-P. Sonnet, “Le Sinaï dans lʼévénement de sa lecture: La dimension pragmatique dʼExode 19–24,” NRTh 111 (1989), 322–344, particularly 335–336, 341–343. 22

1128

Jean-Pierre Sonnet

in Leviticus or in Numbers.23 It is only at the end of Deuteronomy that the narrator authoritatively records the fulfilment of the peopleʼs pledge and, thus, their full compliance with Godʼs will as conveyed by Moses (and Joshua). A macro-plot thus spans Exodus through Deuteronomy, the four books centered on Mosesʼs prophetic mediation.

2 The Completion of the Sanctuary The macro-plot of the peopleʼs compliance with Godʼs legal revelation combines with another extended plot that unites the books of Genesis and Exodus, spanning Gen 1 and Exod 40. It leads to the other report of all-embracing completion in the Pentateuch – the completion of the sanctuary (‫)משכן‬. Though the achievement of the portable sanctuary is credited to Moses,24 to his chief artisan, and to his collaborators, it is also a popular achievement, ascribed to the people as a whole in Exod 39:32 – “the sons of Israel did according to all that YHWH had commanded Moses, so did they” (‫ויעשו בני ישראל ככל אשר צוה יהוה‬ ‫ ;את משה‬Exod 39:32) – and in Exod 39:42–43 – “According to all that YHWH commanded Moses, so the sons of Israel had done all the work. And Moses saw all the work, and, behold, they had done as YHWH had commanded” (‫כאשר‬ ‫)צוה יהוה‬. Exodus 39:32b

‫ויעשו בני ישראל ככל אשר צוה יהוה את משה‬ The sons of Israel did according to all that YHWH had commanded Moses.

 Fulfilments of individual commands are found in Lev 8:36; 10:7; Num 1:54; 2:34; 8:20, 22; 9:5; 17:26; 20:27; 31:31; 32:31; 36:10. Interestingly enough, the endings of Leviticus and Numbers somehow prepare for the completion of Deut 34:9. In both cases, the book ends with a retrospective summary, referring to commandments to be implemented: “These are the commandments which YHWH commanded Moses for the sons of Israel in Mount Sinai” (Lev 27:34); “These are the commandments and the judgments, which YHWH commanded by the hand of Moses unto the sons of Israel in the plains of Moab by Jordan near Jericho” (Num 36:13). The endings of Exodus and Deuteronomy, for their part, after mentioning the implementation of the commandments by the people, point to the “corroborative testimony of all Israel concerning Mosesʼs actions [‘in the eyes of all the house of Israel’ (Exod 40:38); ‘in the eyes of all Israel’ (Deut 34:12)], and therefore, further legitimize the Mosaic Torah”; see E. Ben Zvi, “The Closing Words of the Pentateuchal Books: A Clue for the Historical Status of the Book of Genesis within the Pentateuch,” BN 62 (1992), 7–10, here 9, n. 7. 24  As observed by G. F ischer and D. M arkl , Das Buch Exodus (NSKAT 2; Stuttgart: Katholisches Bibelwerk, 2009), 380, Exod 40:16–33 states seven times that Moses carried out everything “as YHWH had commanded him.” 23

The Dynamic of Closure in the Pentateuch

1129

Exodus 39:42–43 ‫ככל אשר צוה יהוה את משה כן עשו בני ישראל את כל העבדה וירא משה את כל המלאכה והנה‬ ‫עשו אתה כאשר צוה יהוה כן עשו ויברך אתם משה‬ According to all that YHWH had commanded Moses, so the sons of Israel did all the work. And Moses saw all the work, and, behold, they had done as YHWH had commanded, so they had done it: and Moses blessed them. Deuteronomy 34:9

‫וישמעו אליו בני ישראל ויעשו כאשר צוה יהוה את משה‬ And the sons of Israel listened to him and did as YHWH had commanded Moses.

An interesting analogy links the two completion reports. In both cases, the mediators – actually the mediators of Moses, the chief mediator – are presented as “full of a spirit of wisdom.” In Deut 34:9, Joshua has been described as “full of the spirit of wisdom” (‫)מלא רוח חכמה‬.25 In the context of the Pentateuch, this portrait refers to the appointment of Joshua as successor of Moses in Num 27:18, where Joshua is described as “a man in whom is spirit” (‫)איש אשר רוח בו‬. However, the association of spirit and wisdom is found only in Deut 34:9, and the combination of the two echoes the only previous use of the phrase “spirit of wisdom” (‫ )רוח חכמה‬in the Pentateuch, in Exod 28:3, in the context of the construction of the sanctuary.26 The assistants of Moses are described by God in Exod 28:3 as artisans “whom I filled with a spirit of wisdom” (‫אשר מלאתיו רוח‬ ‫)חכמה‬. They receive their iconic figure in Bezalel, of whom God says in Exod 31:3, “I have filled him with the spirit of God, in wisdom, and in understanding, and in knowledge, and in all manner of workmanship” (‫ואמלא אתו רוח אלהים‬ ‫ ;בחכמה‬see also 31:6; 35:31,35; 36:1–2).27 Thanks to Bezalelʼs wisdom and to his wise assistants, a comprehensive work is brought to completion – that of the portable sanctuary. When it comes to the completion of a complex work involving the entire people, wisdom is thus the most needed gift.28  Joshuaʼs excellence in wisdom echoes the first measure taken by Moses in Deuteronomy, his appointment of “wise men” (‫ )חכמים‬as leaders, officials, and judges (1:13; see also 1:15); wisdom was thus the chief quality expected by Moses of his lieutenants in the administration of the people. A significant link is thus created between the incipient mediation of Mosesʼs “wise” attendants and the arch-mediation of Joshua, “full of the spirit of wisdom.” 26  About the relationship between Num 27:18 and Deut 34:9 and the expression “filled with a spirit of wisdom” borrowed from Ex 28:3, see D. Nocquet, “La succession de Moïse et la place dʼÉléazar dans le livre des Nombres,” in The Books of Leviticus and Numbers (ed. T. Römer; BETL 215; Leuven: Peeters, 2008), 655–675, here 660–663. 27  See also 1 Kgs 7:14, about Hiram, the main artisan of Solomonʼs temple. 28  The lack of wisdom in the people, on the contrary, is stigmatized in Deut 32:6 (“O foolish people and unwise”). M. W einfeld , Deuteronomy and the Deuteronomic School (Oxford: Oxford University Press, 1972), emphasizes the different value of wisdom between the two texts: in Exod 28:3, as elsewhere in the Priestly literature, “it means technical wisdom (= skill)” (181, n. 3), whereas in Deut 34:9, where the Priestly material has been reworked by a Deuteronomic hand, it agrees “with Deuteronomyʼs conception of the required intellectual 25

1130

Jean-Pierre Sonnet

Genesis 1:1 The spirit of God (‫ )רוח אלהים‬was hovering over the waters. Exodus 28:3 “And you shall speak to all wise of heart (‫)חכמי לב‬, whom I have filled with a spirit of wisdom (‫)אשר מלאתיו רוח חכמה‬.” Exodus 31:3, 6 “And I have filled him (Bezalel) with the spirit of God (‫)ואמלא אתו רוח אלהים‬, with wisdom (‫)בחכמה‬, intelligence, and knowledge in every kind of craft. [. . .] And I have given wisdom to all the wise of heart, so that they may do all that I have commanded you.” Exodus 35:31, 35 (see also 36:1–2) “And (YHWH) has filled him (Bezalel) with the spirit of God (‫)וימלא אתו רוח אלהים‬, with wisdom (‫)בחכמה‬, intelligence, and knowledge in every kind of craft. [. . .] He has filled them with wisdom of heart (‫)מלא אתם חכמת לב‬.” Deuteronomy 1:13 (see also 1:15) “Choose for each of your tribes individuals who are wise (‫)חכמים‬, discerning, and reputable to be your leaders”. Deuteronomy 34:9 Now Joshua son of Nun was filled with a spirit of wisdom (‫)מלא רוח חכמה‬, because Moses had laid his hands upon him.

The motif of the completion of the sanctuary in Exod 39–40 creates an impressive arc that spans the first two books of the Pentateuch and echoes the completion of creation in Gen 1 and 2.29 In its achievement, the microcosm of the sanctuary matches the achievement of the macrocosm of creation. A series of lexical and syntagmatic recurrences marks out the two texts:

qualities of the leader” (181, n. 3). Whatever the genetic reconstruction of such nuances, the symbol of wisdom is operative in both texts, in similar structural junctures. 29  The congruence betrays all the signs of a P redactional phenomenon. For a recent review of literature, see D. C. T immer , Creation, Tabernacle and Sabbath: The Sabbath Frame of Exodus 31:12–17; 35:1–3 in Exegetical and Theological Perspective (FRLANT 227; Göttingen: Vandenhoeck & Ruprecht, 2009), 12–17. See also A. de P ury , “Pg as the Absolute Beginning,” in Dernières rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (ed. T. Römer and K. Schmid; BETL 203; Leuven: Peeters, 2007), 99–128, here 111: The structure of Pg “betrays a philosophically ambitious, audacious and strongwilled interpretation of the origins and destiny of mankind and of Israelʼs place in it. Within Elohimʼs nations of the earth, Israel has one mission only: to establish and keep the dwelling that will allow YHWH (i.e., Elohim in his ultimate identity) to reside among humanity in the midst of the Israelites. Where the sanctuary will finally be established is not important – the question is never explicitly addressed in Pg – but that it be installed, that is the Priestly perspective, the sense of mankindʼs history as a divine project!”

The Dynamic of Closure in the Pentateuch

1131

Genesis 1:1–2:4

Exodus 39–40

And God saw everything that he had made, and behold, it was very good (1:31). Thus the heavens and the earth were finished (2:1). God finished his work which he had done (2:2). So God blessed the seventh day (2:3).

And Moses saw all the work, and behold, they had done it (39:43). Thus all the work of the tabernacle of the tent of meeting was finished (39:32). So Moses finished the work (40:33). So Moses blessed them (39:43).

Moreover, the succession of divine injunctions (“You shall make [. . .]”) and of reports of completion (“And they made [. . .]”) in the narration of Exodus matches the algorithm of Gen 1:3: “Let it be light – and there was light”. What was immediately achieved by the power of Godʼs word now involves the mediation of human savoir-faire and of the “spirit of wisdom.”30 The characterization of Bezalel as filled “with the spirit of God” (‫)רוח אלהים‬, observes Peter Kearney, reminds us of “the ‫[ רוח אלהים‬that] swept over the waters (Gen 1:2)” at the beginning of creation.31 Something of the creation is playing throughout the construction of the ‫משכן‬, and the sacred time of the Shabbat eventually joins the sacred space of the sanctuary.32 Creation is not a mere background for history; it is a design that unfolds throughout history in Genesis and Exodus and that leads to the tabernacle – inhabited by God and surrounded by the community of the covenant. Yet the completion reported in Exod 39–40 comes along with a meaningful impulse: a divine and human achievement, the sanctuary of the desert is a sanctuary on the move, provided with all the necessary rings and poles to be carried away. It is meant to reach another space, the space of the promised land. The sanctuary is caught up in this mobility with the final vehicle of Mosesʼs teaching – the written book of the Torah (which equally enters a scheme of completion in Deut 31:24: “When Moses had finished writing the words of this Torah to the very end”). The two portable artifacts are physically combined in Mosesʼs order to the Levites in Deut 31:26: “Take this book of the law and put it beside the ark of the covenant of YHWH your God.” “Putting the torah beside the mobile ark of the covenant, which the Levites carried,” Dennis Olson writes, “underlines the ability of the written torah to travel and thus to

 The successful mediation by Bezalel and his collaborators in Exod 35–40 is anticipated by the sinful and idolatrous initiative of the people in Exod 32. The overcoming of the peopleʼs sin by Godʼs justice and mercy in Exod 33–34 (implying Mosesʼs prophetic mediation) is actually what leads to the successful construction of the sanctuary by the faithful artisans. 31  P. J. Kearney, “Creation and Liturgy: The P Redaction of Exod 25–40,” ZAW 89 (1977), 375–387, here 378. 32  See J. L. Ska, “La structure du Pentateuque dans sa forme canonique,” ZAW 113 (2001), 331–352, here 336–337. 30

1132

Jean-Pierre Sonnet

transcend physical limitations of space as well as time.”33 The Torah book will play a definitive role in the obedience of Joshua and his generation (see Judg 1:8; 8:30–35), anticipated in Deut 34:9. It is remarkable that the double closure of the Pentateuch is associated with mobile artifacts – the ark and the book – both of them meant to cross over and beyond the limit of the desert and the limit of the time frame of the Pentateuch. No closure could be more dynamic.

3 Conclusion The double dynamic of completion that underlies the Pentateuch challenges and emulates critical reflection about its redactional genesis. The first arc, which spans Gen 1 to Exod 40, most probably betrays the agenda of the Priestly redaction (whatever the actual ending of the P source).34 The second arc, which leads to Deut 34:9, apparently seeks to balance cultic presence with covenantal commitment.

Gen 1:1–2:4     Exod 19–24      Exod 39–40     Deut 34:9

The redactional origin of Deut 34 is a provocative conundrum, even if, as Serge Frolov has recently claimed, the text features a quite defensible narrative consistency.35 It is nevertheless legitimate to point to a combination of Deuteron33  D. T. Olson, Deuteronomy and the Death of Moses: A Theological Reading (OBT; Minneapolis: Fortress, 1994), 136; see also Sonnet, Book within the Book (see n. 19), 227–230. 34  About the moot issue of the ending of P, see the critical survey by J. L. Ska, “Le récit sacerdotal: Une ‘histoire sans fin,’ ” in The Books of Leviticus and Numbers (ed. T. Römer; BETL 215; Leuven: Peeters, 2008), 631–653. Exod 40 has been identified as the conclusion of P by T. P ola , Die ursprüngliche Priesterschrift: Beobachtungen zur Literarkritik und Traditionsgeschichte von Pg (WMANT 70; Neukirchen-Vluyn: Neukirchener Verlag, 1995), followed by M. Bauks, R. G. Kratz, C. Levin, and J. C. Gertz, among others; see Ska, “Le récit sacerdotal” (see n. 34), 635, n. 18, whose main objection to such an early ending is the fate of Godʼs promises and oaths concerning the gift of the land (639–643). 35  In his inductive reading of Deut 34, S. F rolov, “The Death of Moses and the Fate of Source Criticism,” JBL 133 (2014), 648–660, suggests that “the chapter is an entirely coherent, integral composition that fits well in its context on multiple levels and plays an important compositional and conceptual role in this context” (648). The article challenges the deductive analysis by P. Y. Y oo , “The Four Moses Death Accounts,” JBL 131 (2012), 423–41, which is built on preceding source-critical scholarship. See also the responses to Frolovʼs reading by P. Y. Y oo , “The Place of Deuteronomy 34 and Source Criticism: A Response to Serge

The Dynamic of Closure in the Pentateuch

1133

omistic and, apparently, Priestly elements,36 in particular in 34:9.37 In my view, the final product could represent a phenomenon of “assimilation” (I refer to the redactional concept introduced by Yair Zakovitch),38 contrived by pentateuchal redactors.39 A given and inceptive analogy between the two contingencies (the active reception of the sanctuary and the active reception of the Torah) could have been enhanced by a pentateuchal redactor, in particular through the reuse of the phrase ‫רוח חכמה‬, “spirit of wisdom,” so as to create a dynamic and analogical unfolding from the one to the other. Yet the first lesson is a lesson of narrative theology. The Pentateuch does not close without recording within itself the actuality of the peopleʼs reception: “the sons of Israel listened [. . .] and did just as YHWH had commanded Moses” Frolov,” JBL 131 (2012), 661–668; S. Dolansky, “Deuteronomy 34: The Death of Moses, Not of Source Criticism,” JBL 131 (2012), 669–676; and D. M. Carr, “Unified until Proven Disunified? Assumptions and Standards in Assessing the Literary Complexity of Ancient Biblical Texts,” JBL 131 (2012), 677–681. 36  See, in particular, T. R ömer and M. Z. B rettler , “Deuteronomy and the Case for a Persian Hexateuch,” JBL 119 (2000), 401–419, who speak of a “Priestly-Dtr joint venture” (408). 37  Deut 34:1a, 7–9 have been initially ascribed to P and have been considered by Theodor Nöldeke, Julius Wellhausen, and Martin Noth as the end of the Priestly document. L. Perlitt, Deuteronomium-Studien (FAT 8; Tübingen: Mohr Siebeck, 1994), 123–143, has, however, raised doubts challenging the traditional view and has described Deut 34 as a late text combining Priestly and Deuteronomistic elements. Perlittʼs critical view did not prevent C. Frevel, Mit Blick auf das Land die Schöpfung erinnern: Zum Ende der Priestergrundschrift (HBS 23; Freiburg: Herder, 2000), from coming back to the classical thesis; see also L. Schmidt, Studien zur Priesterschrift (BZAW 214; Berlin: de Gruyter, 1993), 241–251; and Yoo, “Four Moses Death Accounts” (see n. 35). 38  Y. Zakovitch, “Assimilation in Biblical Narratives,” in Empirical Models for Biblical Criticism (ed. J. H. Tigay; Philadelphia: University of Pennsylvania Press, 1985), 175–192, describes as “assimilation” a process of harmonization between narratives presenting a certain degree of analogy. 39  For further signs of pentateuchal redaction in Deut 34, see K. Schmid, “Der Pentateuch­ redaktor: Beobachtungen zum theologischen Profil des Toraschlusses in Dtn 34,” in Dernières rédactions du Pentateuque, de lʼHexateuque et de lʼEnnéateuque (ed. T. Römer and K. Schmid; BETL 203; Leuven: Peeters, 2007), 183–197. Schmid examines three motives: Mosesʼs vision of the land in Deut 34:1–4 as a framing device that refers to Abrahamʼs initial vision in Gen 13:10–15; Mosesʼs status as Erzprophet (Deut 34:10), in reference to Deut 18:15 and the prophets who will follow him beyond the Pentateuch; and Mosesʼs death at the age of 120 (Deut 34:7) as a reference to the limit set by God in Gen 6:3, with the consequence: “Moses Tod im Ostjordanland beruht nach Dtn 34,7 nicht auf einer Verschuldung Moses, sondern auf einem Verhängnis – der göttlich verfügten Limitierung der menschlichen Lebens­zeit” (139). In my view, such a general determination has to be combined with Mosesʼs personal sin, of which the prophet is reminded by God in Deut 32:48–52. For his part, E. Otto, Das Deuteronomium im Pentateuch und Hexateuch (FAT 30; Tübingen, Mohr Siebeck, 2000), 227, ascribes Deut 34:9 to “Ergänzungen aus der Schule der Pentateuchredaktion” (actually “einer prozadokidischen Ergänzung” parallel to Num 27:15–23); see also Otto, Die Tora (see n. 18), 208–209.

1134

Jean-Pierre Sonnet

(Deut 34:9). Such a closure in Deuteronomy echoes and extends a previous active reception at the end of Exodus: “According to all that YHWH had commanded Moses, so the sons of Israel had done all of the work. [. . .] They had done it just as YHWH had commanded” (Exod 39:42–43).40 The plot of the Pentateuch is driven by this implementation of Godʼs orders in two stages – in the construction of the sanctuary and in the implementation of the Torah – which dual implementation provides history with propitious tokens. The final note of the Torah is not that of a mere possibility, let alone of impossibility. Whatever the tribulations and the betrayals that will follow in history – and they will follow, authoritatively announced as they have been – a promising token has been given, authoritatively recorded in the Pentateuchʼs closure: the people have initially “chosen life” (Deut 30:19). They listened and they did.

40  The peopleʼs part is met by the divine counterpart: “the glory of YHWH filled the tabernacle” (Exod 40:34).

Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture James W. Watts The Pentateuchʼs juxtaposition of different genres within a narrative framework provides some of the evidence for building source- and redaction-critical theories of the Pentateuchʼs literary history. Rhetorical analysis suggests, however, that such genre juxtapositions are characteristic of an ancient Near Eastern strategy of persuasion. The Pentateuchʼs inset genres, especially its lists of instructions and laws, generated most of its normative force that, together with its ritualization, led to its scripturalization as Torah.

Narrative Genre Expectations The narrative framing of pentateuchal lists has dumbfounded modern biblical critics for a very long time, because the five books of the Torah together make for a very bad story. For two hundred years, the Pentateuchʼs poor narrative form has provided the fodder for reconstructing the history of the text. The Pentateuch suffers from at least three kinds of narrative difficulties. First, the story contains many repetitions and contradictions. The repetitions include famous examples, such as the two creation stories (Gen 1–2), the three wife-sister stories (Gen 12, 20, 26), the two calls of Moses (Exod 3, 6), and many more. Contradictions in the plot include different sequences for the creation of animals and humans (Gen 1:20–27; 2:4–25) and different counts for the numbers of each species of animal in the ark and for the length of time they stayed in the ark (Gen 6:19–20; 7:2–4, 8, 12, 15, 17, 24; 8:3, 6, 10, 12). Contradictions also appear at the level of motivation: YHWH decides to destroy the human race but then saves Noah and his family (Gen 6) and commissions Moses to free the Israelites from Egypt and then tries to kill him (Exod 3–4). The problems of repetition and contradiction also appear in the Pentateuchʼs lists of instructions. The legal collections duplicate and contradict each other on some issues (e.g., the altar laws of Exod 20:24; Lev 17:3–7; and Deut 12:13–14; the different calculations of reparations for theft in Exod 21:37–22:3 and Num 5:7; the three versions of the Ten Commandments in Exod 20:2–17; 34:11–26; and Deut 5:6–21) and omit topics that one might reasonably expect to be included (e.g., how to kill animal

1136

James W. Watts

offerings, rules for contract law). Nineteenth-century biblical scholars developed classical source criticism to explain the repetitions and contradictions. Yet the most convincing accounts did not succeed in eliminating all the repetitions and contradictions from their reconstructed sources.1 Though the problem of repetition and contradiction has drawn the most attention, the Pentateuch deviates from narrative conventions in two other ways as well. A second problem is that many lists interrupt the story, especially after Exod 19. The Ten Commandments (Exod 20) are followed by successively larger collections of laws and instructions: the Covenant Code (Exod 21–23); the instructions for building the tabernacle (Exod 25–31), which are followed by a detailed narrative of their fulfillment (Exod 35–40); the instructions for offerings in the tabernacle (Lev 1–7) and for cleansing impurities (Lev 11–16), which are followed by the Holiness Code (Lev 17–27); genealogies (Num 1–3) and lists of offerings (Num 7), including and followed by miscellaneous ritual and legal instructions interposed within a narrative of wilderness wandering in the rest of Numbers; all of which is concluded finally by a recapitulation of the whole exodus and wilderness story, including the laws and instructions, this time in the voice of Moses (Deuteronomy). As a result, more than half of the Pentateuchʼs contents consist of nonnarrative lists of one kind or another. Generations of biblical scholars have therefore attempted to “improve” the narrative by labeling the laws as secondary additions to one or more pristine stories (e.g., Pg supplemented by Ps).2 A third major deviation from narrative norms lies in the fact that the Pentateuchʼs story does not have a proper narrative ending. Moses completes his speech and then dies (Deut 34), which also ends the Pentateuch. The end of Deuteronomy leaves most of the Pentateuchʼs plot expectations unresolved, most obviously its repeated anticipation that Israel will occupy the land of Canaan. Scholars have therefore tried to provide it with a better ending by recreating older tetrateuchs (the narrative of Genesis through Numbers) and hexateuchs (Genesis through Joshua) or by simply accepting the addendum of the Deuteronomistic History to create an Enneateuch (Genesis through Kings).3 1  E.g., J. W ellhausen (Composition des Hexateuchs und der historischen Bücher des Alten Testaments [3rd ed.; Berlin: Georg Reimer, 1899], 207, 315) recognized supplementary layers in each of the reconstructed sources, which set the stage for most of the developments in twentieth-century pentateuchal criticism (so R. G. Kratz, The Composition of the Narrative Books of the Old Testament [trans. J. Bowden; London: T&T Clark, 2005; orig. German ed., 2000], 227–229). 2  E.g., Kratz still maintains that “there is largely agreement over excluding the law” from analysis of the separate, and only narrative, sources (K ratz , Composition of the Narrative Books [see n. 1], 226). 3  The prepentateuchal existence of some form of a Hexateuch ending with the conquest of the land somewhere in Joshua has been the most popular hypothesis from the time of Wellhausen up to today (a recent proponent is D. M. C arr , The Formation of the Hebrew

Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture

1137

These well-known issues show that modern expectations about what narrative genres require have driven a great many historical reconstructions of the Pentateuchʼs composition. These expectations about narrative genres are, for the most part, not anachronistic. The writers of biblical literature did know how to tell good stories with strong narrative climaxes. Modern readers usually regard the book of Samuelʼs Succession Narrative (2 Sam 11–20) as the best prose narrative writing in the Bible, but the pentateuchal writers also knew how to write well-plotted stories on a fairly large scale. The Joseph story is certainly one (Gen 37; 39–50). The exodus story, despite all its multiple layers and intrusions, manages to maintain narrative suspense over fourteen chapters (Exod 1–14), including a double climax in the death of the firstborn and the crossing of the Reed Sea, followed by the cathartic celebration of the Song of the Sea (Exod 15). Exodus 1–15 shows that pentateuchal redactors as well as authors could conform their materials to narrative plot conventions even while incorporating multiple sources, doublets, contradictions, and lists of instructions. The fame of Exod 1–15 as “the exodus story” throughout Jewish and Christian cultures shows that these problems need not prove fatal for narrative success. The exodus story, however, provides one thing that the Pentateuch as a whole does not: a satisfying narrative climax and conclusion. We should therefore consider carefully the failure of the Pentateuch overall to abide by this narrative convention that is attested in its own pages. The skills that biblical writers can deploy in plotting narratives should warn us that their failure to meet narrative expectations in combining stories with instructions in the composite Pentateuch, as well as probably in P and Deuteronomy, was the product not of authorial or editorial incompetence but of deliberate choices. These choices were clearly literary and rhetorical choices, but they were not narrative choices. They made no attempt to conform to the standards of how to conclude a story, then or now. Nevertheless, the choices they made were successful. The fact is that it is this form of the text, this genre-breaking amalgam of stories with lists of all sorts, this Pentateuch, that has survived. Indeed, it has more than survived: it succeeded to an unprecedented degree compared to all other Jewish literature – or, frankly, all other ancient literature until the development of the Christian canon that imitated Bible [Oxford: Oxford University Press, 2011], 290–291, 296). That the current Pentateuch was preceded by a Tetrateuch ending in Numbers was maintained most famously by M. Noth, A History of Pentateuchal Traditions (trans. B. W. Anderson; Englewood Cliffs, NJ: Prentice-Hall, 1972 [orig. German ed., 1948]). That it was preceded by a continuous narrative to the end of the Judean kingdom has been suggested by, among others, J. B lenkinsopp , The Pentateuch: An Introduction to the First Five Books of the Bible (ABRL; New York: Doubleday, 1992), 41–42, and by H.-C. S chmitt, “Die Suche nach der Identität des Jahweglaubens im nachexilischen Israel: Bemerkungen zur theologischen Intention der Endredaktion des Pentateuch,” in Pluralismus und Identität (ed. Joachim Mehlhausen; Gütersloh: Kaiser, 1995), 262.

1138

James W. Watts

and incorporated the Pentateuch while trying to supersede it. And it seems to me that it is this success – the fact that it was this Pentateuch that was scripturalized as Torah and became the first scripture of Western religious history – that we biblical scholars most need to explain about it. Our primary attention should not be directed at its earlier versions and stages and traditions. Even if we reconstruct them accurately, we only succeed in reproducing more conventional forms of ancient literature than are exhibited by the extant Pentateuch. We already have such conventional forms in abundance from other ancient cultures, but these forms of literature did not experience the religious and cultural success of the Torah as we have it. The pressing question, then, for me at least, is: Why this text and in this form rather than any other? Compositional history can play a role in answering this question, but not if we insist on working with narrative genre expectations that the Pentateuch itself refuses to respect. We need to rethink what the Pentateuch does as religious literature and how it does it.4 Though narrative coherence plays a role in most reconstructions of the Pentateuchʼs composition, some scholars think it governed only the writing of the individual sources. The editing of the sources together may have been governed by other considerations.5 Some point to the ancient rabbis as recognizing the incoherence of the Pentateuchʼs narrative and the incompatibility of its laws.6 4  Similarly D. J. A. Clines, “Does the Pentateuch Exist? Seven Questions We Should Be Asking if It Does,” in A Pillar of Cloud to Guide: Text-Critical, Redactional, and Linguistic Perspectives on the Old Testament in Honour of Marc Vervenne (ed. H. Ausloos and B. Lemmelijn; BETL 269; Leuven: Peeters, 2014), 45–54; idem , “Putting Source Criticism in Its Place: The Flood Story as a Test Case,” in Biblical Interpretation and Method: Essays in Honour of John Barton (ed. K. J. Dell and P. M. Joyce; Oxford: Oxford University Press, 2013), 3–14. 5  So B. J. Schwartz, “The Pentateuchal Sources and the Former Prophets: A Neo-Documentarianʼs Perspective,” in this volume, pp. 783–793, and also those who think the finished Pentateuch was the product of a political compromise between rival groups in the Persian period, e.g., P. Frei, “Zentralgewalt und Lokalautonomie im Achämenidenreich,” in P. Frei and K. Koch, Reichsidee und Reichsorganisation im Perserreich (2nd ed.; OBO 55; Fribourg: Universitätsverlag, 1996 [1st ed., 1984]), 8–131; idem , “Persian Imperial Authorization: A Summary,” in Persia and Torah (ed. and trans. J. W. Watts; Atlanta: Society of Biblical Literature, 2001), 5–40; F. C rüsemann , Die Tora: Theologie und Sozialgeschichte des alttestamentlichen Gesetzes (Munich: Kaiser Verlag, 1992), 260–261; K. S chmid , “The Persian Imperial Authorization as a Historical Problem and as a Biblical Construct: A Plea for Distinctions in the Current Debate,” in The Pentateuch as Torah (ed. G. N. Knoppers and B. M. Levinson; Winona Lake, IN: Eisenbrauns, 2007), 23–38; K.-J. Lee, The Authority and Authorization of Torah in the Persian Period (Leuven: Peeters, 2011); and my own conclusions in J. W. Watts, Reading Law: The Rhetorical Shaping of the Pentateuch (Sheffield: Sheffield Academic Press, 1999), 137–143. 6  B. J. Schwartz, in his summation of the conference Convergence and Divergence in Pentateuchal Theory, distinguished the approaches of Neo-Documentarian theories of pentateuchal composition from nondocumentarian theories based on their treatment of the presence

Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture

1139

But it is not clear that rabbinic recognition of the need to harmonize disparate laws is the same thing as recognizing narrative incoherence. It may instead reflect the rabbisʼ different expectations of the genres of both pentateuchal narratives and laws.7 The assumptions about narrative coherence that govern critical models of pentateuchal composition consider a range of possible genres and their conventions that is too limited. They almost never take into account that juxtaposed genres each influence how the other gets read or heard.

Inset Genres The Torahʼs storyline encompasses collections of laws, lists of ritual instructions, collections of tabernacle building instructions, and genealogies that amount to considerably more than half of the Pentateuchʼs total word count. Biblical reception history shows that narratively inset genres tend to influence how the narratives around them get read, rather than vice versa. The frame genre – in the Bible usually narrative prose – does not lend its reading conventions to the inset genres,8 but rather the inset genres lead readers to read the frame narrative by the inset genreʼs conventions. For example, the history of chanting and singing scriptural texts of all genres takes the reading conventions of song genres and extends them to the prose genres contained in scriptures. We do not know when the practice of chanting the Torahʼs text began, but it is clear that editors of biblical books were already inserting psalms so that psalmic themes and conventions of performance would influence the interpretation of the surrounding narratives. Their use of hymns to provide thematic emphasis as conclusions or brackets to narrative blocks (Exod 15; Deut 32; Judg 5; 1 Sam 2; 2 Sam 22) shows that this tendency is not just a postbiblical development but or absence of narrative coherence. He argued that Neo-Documentarians like himself find narrative coherence, or any other kind of literary coherence, only in the Pentateuchʼs sources, not in its final shape, whereas supplementary theories of redactional growth try to find at least theological coherence in the narrative whole; see S chwartz , “Pentateuchal Sources” (see n. 5). 7  So B. M. Levinson: “The fact that the three law collections are set in a narrative frame is descriptively accurate. I believe that it goes beyond the evidence, however, to claim that redactional strategy mandates the exclusive validity of one particular form of reading. Indeed, classical rabbinic exegesis, which is certainly synchronic, does not privilege narrative over law, and is certainly not beholden to reading from beginning to end in a unilinear way” (B. M. Levinson, “The Bibleʼs Break with Ancient Political Thought to Promote Equality – ‘It Ainʼt Necessarily So,’ ” JTS 61 [2010], 685–694, here 693). 8  W. J. Houston, among others, has recently defended reading the whole Pentateuch as narrative: “In that its overall structure is narrative, it must be possible to make overall sense of it as narrative”; W. J. H ouston, The Pentateuch (SCM Core Texts; London: SCM Press, 2013), 15, and see 17.

1140

James W. Watts

that it drove the actions of the editors who placed the psalms here to have that thematic effect.9 Contrary to long-standing modern assumptions about “learned psalmography” diverting and degrading the piety of oral psalmody, biblical editors placed psalms within narratives in order to bring the pious enthusiasm of victory hymns to the reading of historiographic stories.10 That is also the case for laws and regulations transmitted by divine oracle to a prophet: their presence within pentateuchal narrative led to the entire fivebook collection being read as law (even when differentiating haggadah from halakah)11 and as oracle. One place where this traditional effect still prevails over modern criticism is in Deuteronomy. Despite its narrative framework, that bookʼs hortatory cast dominates its interpretation in contemporary criticism, as it should. The history of the Pentateuchʼs interpretation in Jewish and Christian traditions shows clearly that the legal and instructional conventions of these 9  See J. W. Watts , Psalm and Story: Inset Hymns in Hebrew Narrative (JSOTSup 139; Sheffield: Sheffield Academic Press, 1992), and K. F insterbusch , “Integrating the Song of Moses into Deuteronomy and Reshaping the Narrative: Different Solutions in MT Deut 31:1–32:47 and (the Hebrew Vorlage of) LXX Deut 31:1–32:47,” in this volume, pp. 631–650. 10  There is an old idea in Psalms research, going back at least to S. Mowinckel but still repeated in recent surveys of biblical psalmody, that the vivid traditions of oral psalmody were constricted and repressed when scribes reduced them to writing. As so-called learned psalmography wrote psalms for narratives and acrostic alphabetical exercises in the Second Temple period, it lost a vital connection to the worshiping community. On the contrary, all the evidence we have for how psalms were actually used by writers and worshiping communities indicates the opposite development. Including psalms in narratives did not lead to psalms being read like stories; it rather led to stories being sung like psalms. Both Jewish and Christian traditions developed traditions of chanting the narrative texts that frequently led to full-scale melodic and even choral treatments. We do not know how far back in time such traditions of musical performance go, but it is plausible that most inset hymns have been strategically placed in biblical narratives to stimulate a sung response praising God for the stories that precede them. See J. W. Watts, “Biblical Psalms outside the Psalter,” in The Book of Psalms: Composition and Reception (ed. P. W. Flint and P. D. Miller; VTSup 99; Leiden: Brill, 2005), 288–309, here 304–308. 11  For example, J. Neusner described halakic use of the narrative of creation in Gen 1: “The Halakhic starting point then is self-evident. [. . .] If we begin at the end and reconstruct that process, we find ourselves in the very heart of the narrative that yielded the principles realized in the Halakhic rules. For the premise of the Halakhic enterprise – narratives bear within themselves laws that define correct conduct in the Israelite social order – leaves no alternative. From the Pentateuchal narratives, exhortations, and ad hoc rules, the Rabbinic sages undertook to define a system and design a structure. That would define how the behavior of an entire society would realize in everyday conduct the implications of the Torahʼs story. So from the outcome, the shape and structure of the Halakhah itself, we find our way back to the starting point: the point in the story that precipitated thought of one sort, rather than of another, on this topic, not on that”; J. Neusner, Judaismʼs Story of Creation: Scripture, Halakhah, Aggadah (Leiden: Brill, 2000), 53.

Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture

1141

inset genres spread to the surrounding narratives, not vice versa. The stories of Genesis, Exodus, and Numbers have been excavated as precedents for religious doctrines and legal rulings throughout Jewish and Christian history. By contrast, narrative concerns for plot, theme, and characterization emerge hardly at all in the history of interpretation until modern times.12 It is, of course, the case that the two religious traditions differ dramatically on the value they place on pentateuchal law. Whereas halakah based on Torah became fundamental to rabbinic and subsequent Judaisms and was already dominant in various groups of the late Second Temple period, Christiansʼ displacement of Torah with Gospel led them to emphasize the stories of the Pentateuch more than its instructions and laws. The Christian canon subsumes the Pentateuch into a longer Dodecateuch (or Pentadecateuch, by splitting Samuel, Kings, and Chronicles into two books each), a sweeping narrative history that runs all the way through the book of Nehemiah. But Christiansʼ preferred forms of interpretation for most of their religionʼs history did not focus on contextualized narrative plot and context but rather applied typology and allegory to both narrative and legal texts. The choice made in the Second Temple period to define Torah as five books ending with Deuteronomy cannot be explained by either halakah or typology. The literary shape of the Pentateuch requires another explanation. I believe this explanation can be found through rhetoric, by focusing on how the Pentateuch addresses its listeners and readers to persuade them of its claims and of its own authority.

Rhetoric The rhetorical strategy of juxtaposing genres in a sequence of (1) stories followed by (2) lists concluded by (3) positive and negative sanctions shapes the overall structure of the Pentateuch as well as its most important component parts, P and Deuteronomy.13 This basic claim about pentateuchal rhetoric does not rest on the description, much less the definition, of any particular genres 12  Even the traditional distinction between halakah and haggadah did not lead the latter to emphasize conventions of narrative genre, as Neusner observed: “But in the Rabbinic recapitulation in the cause of reconstruction, the Halakhah generalizes the Pentateuchal Halakhah upward. By contrast, the Rabbinic Aggadah particularizes the Pentateuchal Aggadah downward. The one takes specific laws and makes of them governing principles. That is what I mean by generalizing ‘upward.’ The other reads the narratives verse by verse, dismantling the flow of narrative and taking up each constituent piece in turn, thus particularizing ‘downward.’ Consequently, they are not commonly understood to tell a single continuous story. They are rarely viewed as writing, each its own chapters in its own modes of discourse, while collaborating in a common, continuous story”; Neusner, Judaismʼs Story of Creation (see n. 11), 6. See also Levinson, “Bibleʼs Break” (see n. 7). 13  For detailed discussion, see Watts, Reading Law (see n. 5).

1142

James W. Watts

within it. It is rather a claim about the persuasive effect of the combination as it stands in the five-book collection. It is true that the rhetoric of story followed by lists concluded by sanctions is composed of different genre building blocks. But its force depends on an underlying rhetoric of past, present, and future: the rhetoric of how we got here, what we have to do now, and what will happen if we do and if we donʼt. Modern political rhetoric uses this form of persuasion a great deal. Documents from across the ancient Near East show that ancient peoples heard a lot of this kind of rhetoric too.14 It is this rhetorical pattern – story, list, sanction – that explains the persuasive force of Torah and that contributed to its success in making the Pentateuch the first scripture of Western religious history. The history of scripturalization in the Second Temple period shows clearly that the Torah was scripturalized before the Prophets, despite the oracular nature of many of the prophetic books. It also shows that written Torah functioned as a norm first as the ritual law of the Jerusalem temple and was only gradually extended to social and criminal matters in and after the second century BCE.15 It was the Pentateuchʼs lists of ritual instructions that generated its growing authority as scripture. Of the three components of the story-list-sanctions rhetorical convention, it is especially the rhetoric of lists that requires more extensive analysis from Pentateuch scholars than it has received to date.

The Rhetoric of Lists Writing seems to have been invented in Mesopotamia for recording lists. Though other genres of ancient literature draw greater attention from modern readers, the writing and interpretation of lists remained the most basic and also the most prestigious scribal activity in Mesopotamian cultures for more than two thousand years. Historians of the ancient Near East have frequently noted that its development and interpretation constituted a “science of lists” (Listenwissenschaft).16 The various omen series, such as the Old Babylonian collection of extispicy omens, Bārutû, and the Neo-Assyrian collection of celestial omens, Enūma  J. W. Watts, “Story, List, Sanction: A Cross-Cultural Strategy of Ancient Persuasion,” in Rhetoric before and beyond the Greeks (ed. C. Lipson and R. Binkley, Albany: State University of New York Press, 2004), 197–212. 15  See J. W. Watts , “Ritual Legitimacy and Scriptural Authority,” JBL 124 (2005), 401–417; also M. L e F ebvre , Collections, Codes, and Torah: The Re-Characterization of Israelʼs Written Law (LHBOTS 451; New York: T&T Clark, 2006), 18–23, 146–182. 16  See W. von Soden, “Leistung und Grenze sumerischer und babylonischer Wissenschaft,” in Die Welt als Geschichte 2 (1936), 411–464, 509–557; A. A lt , “Die Weisheit Salomos,” TLZ 76 (1951), 139–44, reprinted in idem, Kleine Schriften zur Geschichte des Volkes Israel (2 vols.; Munich: Beck, 1953), 2:90–99; P. W. Coxon, “The ‘List’ Genre and Narrative Style in the Court Tales of Daniel,” JSOT 35 (1986), 95–121. 14

Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture

1143

Anu Enlil, represented the pinnacle of scribal expertise. Mastery of omens still grounded the reputation of Babylonian sages in the Roman period. The power of lists has not waned since antiquity. Lists continue to govern modern bureaucracies in the forms of instructions for routinized procedures and filing systems.17 Though lists have not received as much scholarly attention as have narrative and lyric poetry, they play a determinative role in human behavior. As the theorist of rhetoric J. D. OʼBanion has remarked, Rendered as tallies, recordings of the movements of the stars, word lists, dictionaries, or codified laws, the list is a powerful tool for arranging and disseminating isolated pieces of information. It also comes to arrange and, to a considerable degree, dictate the nature of the lives of those who are affected by lists.18

As tools for economic and legal control, lists dictate modes of exchange and social standing. The historian of law Cornelia Vismann observes that “[l]ists do not communicate; they control transfer operations.”19 The original and continuing dominance of lists in literate cultures fully justifies J. Z. Smithʼs description of them as “the most archaic and pervasive of genres.”20 Because of the social impact of lists, readers use lists in distinctively different ways than they do stories or poems. Lists, by their nature, invite readers and listeners to choose items relevant to themselves and ignore the rest. Whether the list contains omens, ritual instructions, or recipes, readers choose or feel obliged to act on only those elements they regard as appropriate to their situation. Collections of laws, instructions, and genealogies all invite listeners and readers to find those details relevant to their circumstances, their problems, and their identities. When a list appears in a scripture, people frequently presume that all of it must be relevant somehow, but they still pick and choose as their own wishes, time, and circumstances require or allow, leaving the rest for another occasion. I said above that readers tend to spread the conventions of inset genres to the larger narrative contexts that contain them. Nowhere is this more obvious than with the pick-and-choose characteristics of lists. Jewish and Christian interpreters have extended that practice to every verse of Torah and Bible. James Kugel identifies the criteria of relevance to the reader and of equal significance of any 17  J. Goody, The Logic of Writing and the Organization of Society (Cambridge: Cambridge University Press, 1986), 42–121. 18  J. D. OʼBanion, Reorienting Rhetoric: The Dialectic of List and Story (University Park: Pennsylvania State University Press, 1992), 12. 19  C. Vismann, Files: Law and Media Technology (trans. G. Winthrop-Young; Stanford: Stanford University Press, 2008; orig. German ed., 2000), 6; also Goody, Logic (see n. 17), 54–86. 20  J. Z. S mith , “Sacred Persistence: Toward a Redescription of Canon,” in Imagining Religion: From Babylon to Jonestown (Chicago: University of Chicago Press, 1982), 36–52, here 44.

1144

James W. Watts

verse of Scripture for interpreting any other as two of four key characteristics of scriptural interpretation that developed in antiquity.21 On the criteria of relevance and equal significance, let me add that the belief that somewhere in the Torah or the Bible some verse of Scripture must be relevant to me, now, extends and develops an inherent feature of lists of laws and instructions, whether they be ritual instructions, omen lists, or cookbooks, and applies them to other aspects of life. Lists invite readers to find the parts relevant to themselves and ignore the rest, at least for now. This characteristic practice of reading lists has become a distinguishing feature of how religious individuals and communities read scriptures.

Ritualizing Texts Even when rhetorical interpretation includes a robust analysis of lists, rhetoric alone cannot fully explain the scripturalization of the Pentateuch, at least if we limit it to the bookʼs words and their meanings. Comparative studies of scriptures show that they get ritualized in three different dimensions. Communities that venerate a scripture ritualize its rhetoric, that is, the meaning of its words, through sermon, study, and commentary. But such communities also ritualize the reading process itself through recitation, memorization, and song or chant, as well as performing its contents through various artistic and theatrical media. In addition, communities that venerate a scripture ritualize the physical book by its iconic display, manipulation, decoration, and storage in elaborate book boxes and libraries that serve as reliquaries for the sacred texts.22 In Exodus and Deuteronomy, the Pentateuch commands its own ritualization in iconic (tablets, ark, scroll) and performative (read aloud) dimensions, and its exhortations presuppose semantic study of at least its laws. The vocabulary of these commands characterizes the Pentateuch in covenantal, legal, and instructional terms as tôrâ, ʿēdût, and bərît.23 These terms indicate that its writers intended the legal and, especially, ritual material to generate the Pentateuchʼs dominant, what we might call “scriptural,” effect. Thus rhetorical analysis, the history of interpretation, and the Pentateuchʼs own instructions for its ritualization all point to the lists of laws and instructions as the main engine for its scripturalization. If that is the case, the narrative 21  J. L. K ugel , Traditions of the Bible: A Guide to the Bible as It Was at the Start of the Common Era (Cambridge, MA: Harvard University Press, 1998), 14–19. 22  J. W. Watts , “The Three Dimensions of Scriptures,” Postscripts 2/2–3 (2006–2008), 135–159, reprinted in Iconic Books and Texts (ed. J. W. Watts; London: Equinox, 2013), 9–32. 23  See J. W. Watts, “From Ark of the Covenant to Torah Scroll: Ritualizing Israelʼs Iconic Texts,” in Ritual Innovation in the Hebrew Bible and Early Judaism (ed. N. MacDonald; BZAW 468; Berlin: de Gruyter, forthcoming).

Narratives, Lists, Rhetoric, Ritual, and the Pentateuch as a Scripture

1145

framework may have been utilized simply as a capacious container for the instructional contents. Discussions of literary genres routinely comment on narrativeʼs capacious ability to incorporate other genres, whether narrative takes the form of classical epic or modern novel.24 Therefore, we must reckon with the possibility that neither the writers nor the editors of the Pentateuch felt bound to respect conventions of narrative genre composition. If that possibility be admitted, than biblical criticism should abandon compositional theories that presuppose conformity to narrative conventions as evidence for reconstructing one stage or another.

24  E.g., “A commonplace of Renaissance literary theory and practice held that epic was a sufficiently capacious genre to include elements of other forms of literature” (C. B ond , Spenser, Milton, and the Redemption of the Epic Hero [Newark: University of Delaware Press, 2011], 80), and “The novel is by all accounts a remarkably capacious genre, and it seems able to embrace just about every other literary form within its bounds. Among contemporary genres, only film has a similar capacity” (A. J. Cascardi, “The Novel,” in The Oxford Handbook of Philosophy and Literature [ed. R. T. Eldridge; Oxford: Oxford University Press, 2009], 162–179, here 162); similarly A. Fowler, Kinds of Literature: An Introduction to the Theory of Genres and Modes (Cambridge, MA: Harvard University Press, 1982), 180–181.

Contributors Reinhard Achenbach is professor of Old Testament at the Westfälische Wilhelms-Universität Münster, Germany. Rainer Albertz is professor emeritus of Old Testament at the Westfälische Wilhelms-Universität Münster, Germany. Yairah Amit is professor emerita of Bible at Tel Aviv University, Israel. Joel S. Baden is professor of Hebrew Bible at Yale University, New Haven, Connecticut, United States. Richard J. Bautch is professor of religious studies at St. Edward’s University, Austin, Texas, United States. David Ben-Gad HaCohen is an independent researcher living in Israel. Erhard Blum is professor of Old Testament at the Eberhard Karls Universität, Tübingen, Germany. Mark J. Boda is professor of Old Testament at McMaster Divinity College, Hamilton, Ontario, Canada. David M. Carr is professor of Old Testament/Hebrew Bible at Union Theological Seminary, New York, United States. Sidnie White Crawford is the Willa Cather Professor of Classics and Religious Studies at the University of Nebraska–Lincoln, United States. Thomas B. Dozeman is professor of Hebrew Bible at United Theological Seminary, Dayton, Ohio, United States. Cynthia Edenburg is a research fellow and instructor at the Open University of Israel, Raʿanana, Israel. Angela Roskop Erisman is an associate fellow at Hebrew Union College–Jewish Institute of Religion, Cincinnati, Ohio, United States. Israel Finkelstein is professor of archaeology at Tel Aviv University, Israel. Karin Finsterbusch is professor of Old Testament at the University of Koblenz–Landau, Germany.

1148

Contributors

Georg Fischer is professor of Old Testament at the University of Innsbruck, Austria. Tova Ganzel is midrashah director at Bar-Ilan University, Ramat Gan, Israel. Jan Christian Gertz is professor of Old Testament at the University of Heidelberg, Germany. Shimon Gesundheit is senior lecturer in Bible at the Hebrew University of Jerusalem, Israel. Sara Japhet is the Yehezkel Kaufman Professor of Bible emerita at the Hebrew University of Jerusalem, Israel. Jan Joosten is the Regius Professor of Hebrew at the University of Oxford, United Kingdom. John Kessler is professor of Old Testament at Tyndale Seminary, Toronto, Ontario, Canada. Itamar Kislev is senior lecturer in Bible at the University of Haifa, Israel. Ariel Kopilovitz is a doctoral candidate in Bible at the Hebrew University of Jerusalem, Israel. Reinhard G. Kratz is professor of Old Testament at the Georg-August-Universität, Göttingen, Germany. Armin Lange is professor of Second Temple Judaism and director of the Institute for Jewish Studies at the University of Vienna, Austria. Christoph Levin is professor emeritus of Old Testament at the Ludwig-Maximilians-Universität in Munich, Germany. Bernard M. Levinson is the Berman Family Chair of Jewish Studies and Hebrew Bible and professor of classical and Near Eastern studies and of law at the University of Minnesota, Minneapolis, United States. Risa Levitt Kohn is professor of Hebrew Bible and Judaism at San Diego State University, California, United States. Michael A. Lyons is associate professor of Old Testament at Simpson University, Redding, California, United States. Noam Mizrahi is senior lecturer in Bible at Tel Aviv University, Israel. Christophe L. Nihan is associate professor of Hebrew Bible and the history of ancient Israel at the University of Lausanne, Switzerland. Frank H. Polak is professor emeritus of Hebrew Bible at Tel Aviv University, Israel. Christopher A. Rollston is associate professor of Northwest Semitic languages and literatures at George Washington University, Washington, D. C., United States.

Contributors

1149

Dalit Rom-Shiloni is associate professor of Hebrew Bible at Tel Aviv University, Israel. Thomas Römer is professor of Hebrew Bible at the University of Lausanne, Switzerland, and professor at the Collège de France, Paris. Konrad Schmid is professor of Hebrew Bible and ancient Judaism at the University of Zurich, Switzerland. William M. Schniedewind is the Kershaw Chair of Ancient Eastern Mediterranean Studies at the University of California, Los Angeles, United States. Baruch J. Schwartz is associate professor of Bible at the Hebrew University of Jerusalem, Israel. Jean Louis Ska is professor of Old Testament exegesis at the Pontifical Biblical Institute, Rome, Italy. Benjamin D. Sommer is professor of Bible and ancient Semitic languages at the Jewish Theological Seminary, New York, United States. Jean-Pierre Sonnet is professor of Bible at the Pontifical Gregorian University, Rome, Italy. Jeffrey Stackert is associate professor of Hebrew Bible at the University of Chicago, Illinois, United States. Marvin A. Sweeney is professor of Hebrew Bible at the Claremont School of Theology and professor of Tanak at the Academy for Jewish Religion, Los Angeles, California, United States. James W. Watts is professor of Hebrew Bible at Syracuse University, New York, United States. Markus Witte is professor of Old Testament at the Humboldt-Universität, Berlin, Germany. Jakob Wöhrle is professor of Old Testament at Carl von Ossietzky Universität, Oldenburg, Germany. David P. Wright is professor of Hebrew Bible and ancient Near East at Brandeis University, Waltham, Massachusetts, United States. Molly M. Zahn is associate professor and graduate director of religious studies at the University of Kansas, Lawrence, United States.

Ancient Sources Index Hebrew Bible Genesis 1 249, 250, 522, 547–548, 815, 821, 901, 903, 904, 910, 991, 1100, 1132, 1140 1–2 901, 1130–1131, 1135 1–9 529 1–11 212, 213, 509, 601, 863 1:1 1130 1:1–2:4 257, 1131, 1132 1:2 1131 1:3 270, 1112, 1131 1:4 541 1:7 541 1:8 548 1:11–12 548 1:14–18 548 1:20–22 548 1:20–27 1135 1:24–25 548 1:26 280 1:26–28 548 1:28 272, 277, 625, 821, 822, 901 1:29 270–271 1:31 1131 2 249, 250, 1100 2–3 567 2:1 1131 2:2–3 874–875, 876 2:3 1131 2:4 816 2:4–25 257, 1135 2:4–3:24 1112 2:6 332 2:10–14 729–730, 731, 742–743 3:5 535 3:7 418 3:14 952 3:20 567 4 567 4:2 988 4:11 952

4:21 892 5:5 319 6 1135 6–9 205, 399, 522 6:3 1133 6:5–7 1109 6:13 848 6:18 399 6:19 105 6:19–20 1135 7:1 399 7:2 105 7:2–4 1135 7:8 1135 7:9 105 7:11 400 7:12 400, 1135 7:13 400 7:15 1135 7:17 1135 7:17–20 461 7:24 1135 8:3 105, 1135 8:5 380 8:5–11 380–381 8:6 380, 1135 8:8 381 8:9 381 8:10 381, 1135 8:11 381 8:12 1135 8:21 272 8:21–22 1109 9 270, 272, 277, 655, 849, 991, 1031 9:1 272 9:1–5 821, 822 9:3 270–271, 272, 276 9:3–5 271 9:4 277, 495 9:4–5 271

1152

Ancient Sources Index

Genesis (continued) 9:5 272, 655 9:6 655 9:9 272 9:9–10 277 9:25–26 952 10 366 10:7 731 10:8–10 205 11 862 11–25 863 11:3 279 11:7 279 11:26–32 229 11:26–36:43 223 11:27–30 467 11:27–25:10 405 11:28 366, 548 11:29–30 410, 415 11:30 408, 409, 410 11:31 229, 322, 366, 548 11:31–32 224, 393, 394 11:32 395, 396 12 318, 420, 862, 1135 12–13 512 12–25 212, 213, 214 12–50 205 12:1 230, 237, 318, 410 12:1–3 467, 509 12:1–4 229, 360, 363 12:1–9 414 12:2 409 12:2–3 230, 413–414 12:3 944 12:4 395, 396 12:4–5 229, 393, 394, 396 12:5 396, 410 12:6 233, 568–570, 570 12:6–7 233 12:7 410, 535, 805 12:8 233 12:10 415, 570 12:10–20 336–338, 340, 344, 411, 414, 472, 570–571, 585 12:10–13:2 238 12:11 337 12:12 337 12:15 338 12:16 237, 318, 411, 570–571

12:17 570 12:20 571 13 420 13:1–2 237 13:1–17 413 13:2 318 13:5 396 13:6 393, 396, 416, 417 13:10 417 13:10–15 1133 13:12 416 13:14 417, 425, 436 13:14–17 410, 411 13:15 417 13:17 232, 568–569 13:18 233, 323 14 209, 321, 324, 585, 594, 595 14–15 420 14:1–24 413 14:13 323 14:14 232, 321 15 233, 319, 320, 321, 322, 323–324, 338, 410, 548, 549, 550, 585 15:1 321, 322, 324, 408 15:1–12 414 15:2 321, 439 15:2–3 321, 408, 409 15:3 321, 324, 439 15:3–5 410 15:4–7 322 15:5 321, 438, 447 15:6 322, 324, 440, 549 15:7 229, 321, 366, 463, 548, 805 15:8 410 15:9 411 15:9–10 411 15:12 58, 92, 93 15:13 238 15:13–14 550 15:13–16 92–93, 94 15:14 324 15:17 92, 93, 411 15:17–18 414 15:18 320, 549, 805 15:18–19 321 15:19–21 549 15:20 549 16 414, 420 16:1–2 409

Ancient Sources Index

Genesis (continued) 16:7–11 71 16:10 535 16:13 417 16:15 395 16:16 395 17 105, 420, 512, 514, 550, 1031 17:1 319, 395, 463 17:1–6 414 17:1–22 233 17:2 549 17:4 549 17:4–5 550 17:7 549, 1031 17:7–14 413 17:8 549 17:13 105, 1031 17:15–22 414 17:17 322, 409 17:19 105, 1031 17:23 395 17:23–27 413 17:24–25 395 18 318, 903 18:1–15 414, 420, 430 18:1–33 233 18:2 174, 417, 423, 424 18:4–5 441 18:7 416 18:8 416, 441 18:9–15 411 18:11 409 18:12 409 18:13–14 463 18:16–33 414 18:17–33 1109 18:21 417 18:22–33 594 18:24 417 18:25 901, 1112 18:26 417 18:28–32 440 18:32 901 19 209, 901 19–20 420 19:1–22 414 19:3 413 19:21 417 20 361, 411, 472, 1135

20–21 88 20–22 584, 585 20:1 237 20:1–18 238, 414 20:9 845 20:14 237 20:15 237 21 105, 209, 318, 420 21:1 217 21:3 395 21:3–5 413 21:5 395 21:6–21 414 21:8 417 21:9–10 187 21:16 417–418 21:17 71 21:19 418 21:22–24 209 21:22–32 413 21:24 105 21:33 232 21:34 237 22 203, 319, 320, 321, 363, 903, 904 22:1 420, 1092 22:1–19 213, 420 22:3 411, 934 22:4 417, 423 22:7 420, 438 22:8 417, 438 22:9 1092 22:13 417, 423, 424 22:14 180, 304, 417 22:16–18 320, 324 22:17 321 22:18 900 22:19 232 22:20 229 22:23 229 22:30 417 23 105, 203, 233, 354, 413, 420, 585 23:5 354, 418 23:5–6 354–355 23:9 412, 939 23:10 418 23:11 354, 417 23:12 418 23:13 354 23:14 354, 355, 418

1153

1154

Ancient Sources Index

Genesis (continued) 23:18 417 24 230, 245, 315–320, 321, 322, 324, 338, 420, 599 24:1–21 413, 414 24:2–3 430 24:3 315, 317, 319 24:3–8 317 24:4 317, 318 24:5 318 24:6 317, 318 24:7 71, 75, 315, 317, 318, 319, 320, 805 24:8 317, 318 24:9 430 24:10 229 24:10–11 355 24:13 355 24:14–15 355 24:15 229 24:17 315 24:21 315 24:22–67 413, 414 24:24 229 24:32–33 440 24:35 318 24:36 318, 319 24:37–41 317 24:38 315 24:40 71, 75, 319 24:47 229 24:55 174 24:59 319 24:62 319 24:63 417 24:64 417 24:67 105, 319, 410 25–35 212, 214, 851, 863, 865 25–36 213 25:1–12 413 25:10 412 25:11 234, 319 25:19–26 396–398, 857 25:19–36:43 223 25:20 396–398 25:21–26 398 25:24 1002 25:25 987 25:26 395, 397–398, 1002 25:27–34 857

25:29–34 420 25:30–31 438 25:31 333 25:33 333 26 320, 321, 361, 866, 1135 26:1 237 26:1–14 238 26:2 237 26:2–5 214 26:3 320, 321 26:3–5 319, 320, 324 26:4 321, 900 26:5 321 26:12 234, 318 26:12–22 237 26:13 318 26:18 105 26:23–25 234 26:24 321 26:34 105, 395 26:34–35 230 27 214, 987, 1005 27–28 866 27:1–45 857 27:1–33:17 420 27:25–35 430 27:33 157 27:36 1002 27:37 157 27:38 418 27:41–42 991 27:41–28:5 237 27:43 229, 394, 1005 27:43–45 237 27:46–28:9 230, 857 28 319, 584, 864, 865 28:1 319 28:2 319 28:3 507, 508, 521 28:4 322 28:6 319 28:9 105 28:10 394 28:10–22 234, 857, 858, 1003–1004, 1005 28:11–22 304 28:13 463 28:13–15 474 29 1005 29–31 860

Ancient Sources Index

Genesis (continued) 29:4 394 29:5 229 29:6 495 29:11 418 29:13 420 29:15 1005 29:16–17 415 29:18 1005 29:20 1005 29:25 1005 29:27 1005 29:30 1005 31 856 31:3 239 31:5–13 239 31:6 239 31:10 423, 424 31:11 71 31:11–13 239 31:12 424 31:13 318 31:14–16 239 31:20 239 31:22–54 859–860 31:24 239 31:25–54 238 31:29–30 239 31:34–35 239 31:38 237 31:41 237 31:42 239 31:43–32:1 214 31:45 234 31:46 616 31:46–51 430 31:47 239, 860 31:49 239 31:51–52 239 31:53 239, 430 31:54 234 32 507, 508, 521 32:2 71 32:2–3 234 32:4–33 857 32:11 991 32:22–33 234 32:23 373 32:23–27 1003

32:23–33 999, 1002–1003 32:25 522, 858, 1003 32:25–32 507 32:25–33 492–496, 507 32:26 1003 32:29 507, 858, 1003 32:30 494, 508, 1003 32:30–31 508 32:31 508 32:31–33 1003 32:32 495 32:33 508 33 991 33:1 423 33:1–20 209 33:4 418, 420 33:5 423 33:14 991 33:17 214 33:18–19 235 33:19 233, 818 34 209, 586, 866 35 209, 507, 864 35:1–7 598 35:1–8 1004 35:1–15 858 35:2–4 235, 239 35:7 235 35:8 235, 319 35:9–12 507, 508, 521 35:9–13 1004, 1005 35:9–15 235 35:11 463 35:13 1005 35:15 1005 35:16–21 235 35:19 180 35:23 180 35:25 180 35:26 180 35:27 261 35:27–29 235, 866 36 224, 225 36:2–3 105 37 215, 1137 37–50 212, 213, 365, 866, 994 37:1 224, 260 37:1–2 224 37:1–50:26 223

1155

1156

Ancient Sources Index

Genesis (continued) 37:2 224, 260 37:3–11 260 37:9 373 37:10 105 37:11–18 260 37:12 234 37:13–17 234 37:14 260 37:18–36 866 37:19–20 260 37:20 989 37:21–22 260 37:22 989 37:23 260, 423 37:24 260, 989 37:25 260 37:26–27 260, 261, 866 37:26–28 262 37:27–28 105 37:28 105, 237, 257, 260–262, 266 37:28–29 989 37:29–30 260 37:31–35 260 37:35 105, 901 37:36 105, 257, 260–262, 260–262 38 210, 230, 245, 586, 866 38:1 105 39 246, 438 39–50 1137 39:1 105, 237 39:4 246 39:6 246 39:7 246, 425 39:20 990 39:22 990 40:13 241 40:15 989 40:20 241 41:14 989 41:38–57 237 41:46 395 42:1–47:27 235 42:35–38 866 43:3 994 43:8–9 994 43:8–14 866 43:11 157, 353 43:29 423

44:14 994 44:16 994 44:18–34 994 45:4 262 45:5 211 45:6 215 45:10 574 45:16–20 237 46–50 224 46:1–4 235, 237 46:1–27 224 46:2–4 463 46:3 217 46:4 463 46:10 230 46:27 180, 181 46:28 574 46:34 574 47:1 573, 574 47:1–5 571 47:1–6 573 47:1–12 571 47:2–4 573 47:4 572, 574 47:5 571–574, 573 47:5–6 572–573 47:5–12 572 47:6 573, 574 47:11 572 47:19 535 47:23 535 47:27 574 47:29–30 523 47:29–31 237 48 586 48:1–19 224 48:4 805 48:10 420 48:13–14 319 48:15–16 319 48:16 71 48:17–20 319 48:21–22 224 49 586, 607, 866, 1113 49:1 507, 523 49:29–32 237 49:29–33 238 49:29–50:13 239 49:33 224

Ancient Sources Index

Genesis (continued) 50 225 50:8 574 50:15–21 992 50:19–21 211, 215 50:24 320 50:24–25 237 50:24–26 239 50:25 598, 817, 818 50:26 598 Exodus 1 217, 219 1–4 611 1–14 1137 1–15 216, 220, 361, 565, 603, 1137 1:1 215, 617 1:1–6 600 1:1–22 457, 471 1:5 180–181 1:5–6 617 1:7 217, 466 1:8 217, 617 1:9 217, 466, 617 1:9–12 610, 613 1:9–34:32 617 1:11 574 1:12 466 1:12–13 181 1:13 181, 603 1:13–14 472 1:14 180, 181 1:15 471, 610 1:15–22 610 1:17 610 1:20 707 1:20–21 610 1:21 471, 610 2 68, 209, 219, 368, 610, 822 2–4 474, 565, 581 2:1–10 367, 456, 610 2:4–8 707 2:11–14 210 2:11–15 610 2:11–25 456 2:15–23 611 2:16–22 610 2:23 467, 610 2:23–25 610

2:24–25 463 3 584, 749, 820, 1135 3–4 509–514, 611, 1135 3:1 512, 513 3:1–6 457 3:1–16 457, 458, 471 3:1–22 611 3:1–4:18 610 3:2 74, 473, 512 3:5 819 3:5–7 510 3:6 457, 462, 463, 464, 472 3:6–8 462 3:6–12 463 3:7 461, 462, 463, 465, 551 3:7–8 461 3:8 143, 457, 461, 463, 749 3:9 461, 462, 463, 551 3:9–14 513 3:10 462, 463 3:11–12 513 3:12 472, 511, 512, 513, 514, 522 3:13 180, 464 3:13–16 510 3:14 464 3:14–15 74, 218, 462, 1112 3:15 180, 457, 464, 466, 474 3:16 461, 462, 464 3:16–18 513, 514 3:16–20 512 3:17–22 457, 458, 469 3:18 513 3:18–22 611 3:20–21 612 3:21–22 512 4 507, 598, 611 4:1 257 4:1–2 446 4:1–17 456, 457 4:2 473 4:4–6 511 4:5 509 4:6 180 4:14–16 265 4:17 257 4:18 467, 611 4:18–23 456 4:19 467, 610 4:19–20 610, 611

1157

1158 Exodus (continued) 4:19–23 456 4:20 459, 473, 512, 513, 611 4:20–26 508 4:24–26 459, 496, 507, 508, 611 4:25 467 4:26 467, 611 4:27 494, 512, 513 4:27–28 492–496, 507, 508, 514 4:27–31 512–513 4:28 494, 513 4:28–31 513 5 565, 612 5:1–2 513 5:1–6:1 458, 469 5:3 458 5:3–19 612 6 509, 511, 512, 523, 1135 6–7 565 6:2 463 6:2–3 462, 1112 6:2–8 218, 220, 221, 222 6:2–12 509 6:2–13 454, 469 6:3–8 523 6:4 507 6:6 221 6:8 180, 552, 821 6:12 339 6:20 707 6:26–7:6 454, 469 7–9 565 7–12 399, 612 7:8–13 264, 455, 469 7:10 264 7:10–11 265 7:11 264 7:11–12 267 7:13 264 7:14–15 612 7:14–18 262, 457 7:14–25 262 7:14–12:39 612 7:19 458 7:19–20 262, 459 7:19–24 458 7:20 262, 264, 265, 458 7:21 262, 263, 264 7:21–22 262

Ancient Sources Index

7:22 264, 265, 267 7:23–25 262 7:24 262–263 7:25–8:11 457 7:26–29 263, 265 7:26–8:11 263 7:27 267 8:1–3 263, 266 8:2 265, 266 8:3 263–267, 264, 265, 268 8:3–4 266, 267 8:3–11 263 8:4 265, 266 8:11 263, 264, 265 8:12–15 265, 267, 456 8:13 264 8:13–19 187 8:14 264, 265, 267 8:15 264 8:16–28 457 8:18 574 9 470 9–10 612 9–11 565 9:1–7 456 9:1–12 469 9:1–10:2 470 9:8–12 455 9:9–12 187 9:11 267 9:12 264 9:13–21 456 9:13–26 469 9:14–15 469 9:14–17 470 9:16 470 9:19 470 9:22 470 9:22–26 456 9:24 470 9:25 470 9:26 574 9:27–35 456 9:35 264 10 513 10:1–2 470 10:3–7 456 10:7 263 10:8–20 456

Ancient Sources Index

Exodus (continued) 10:19 187 10:20 187, 264 10:21–29 457 10:27 264 11 867 11:1–10 456, 469, 470 11:2–3 612 11:10 264 12 218, 565, 680, 764, 867, 868, 1097 12–13 868 12:1–14 455 12:1–28 820 12:8–9 529, 680, 1097 12:9 917, 1098 12:10 680, 1099 12:11 680 12:14–17 600 12:15 680 12:19 680 12:20 680 12:21–23 457 12:25 457 12:29 989 12:29–34 457 12:33–34 612 12:35–38 512 12:37 400–101, 574, 612, 700 12:37–38 205 12:37–42 885 12:38 570, 612 12:38–39 470 12:39 612 12:40–41 401 12:41 401 12:42 455, 470 12:43–50 819, 820 13–14 565 13:1–2 274 13:1–10 612 13:1–16 867 13:2 555–556, 867 13:3–4 613 13:3–10 455, 470 13:5–10 613 13:11–15 274 13:12–14 556 13:17–19 455, 470 13:17–15:22 704

1159

13:19 239, 598, 817, 818 13:20 612 13:21 73, 74, 75 13:21–22 71, 74, 455, 470, 552 13:22 803 14 399, 597, 599, 749, 819–820, 820, 906 14:1–4 454, 470 14:4 551 14:5–7 457 14:7 749 14:8 551 14:8–10 454, 470 14:9 551 14:10 187, 424, 457, 551 14:11 749 14:11–12 457 14:11–14 457 14:12–17 187 14:13–14 457 14:14 822, 823 14:15–18 454, 470 14:16 551 14:19 71, 73, 74, 75, 598 14:19–20 457 14:19–21 187 14:20 74 14:21 457, 551 14:22 551 14:22–23 454, 470 14:23 551 14:24 71 14:24–25 457 14:25 74 14:27 399, 457, 551 14:27–29 454, 470 14:28 399 14:29 551 14:30–31 457 14:31 700, 750 15 523, 607, 756, 757, 1113, 1137 15:1–18 609 15:1–19 595 15:1–21 1109 15:5 551 15:9 551 15:11 75 15:14–15 1054 15:15 693 15:16 58, 75

1160 Exodus (continued) 15:16–20 187 15:21 609 15:22 188, 612 15:22–24 750 15:22–26 187 15:22–27 457, 473 15:24 749 15:25 473, 598 15:26 598, 756, 757 15:27 473, 612 16 596, 616, 819, 885 16:1 612, 698, 707 16:1–3 454, 470 16:3 749 16:3–4 552 16:4–5 457, 596 16:6–12 454, 470 16:13–14 616 16:13–15 457 16:15–21 454, 470 16:19–36 886 16:21 457 16:22–26 454, 470 16:23 616 16:25–28 886 16:26 143 16:27–31 457 16:28 895 16:31 616 16:32–33 454 16:32–35 470 16:34 454, 458 16:35 454, 819 17–19 455 17:1 612 17:1–7 458, 473 17:3–5 187 17:5 459, 473 17:6 473 17:8–13 473, 798 17:8–16 458 17:9 459, 473, 700 17:9–10 473 17:11 699 17:12 473 17:14 281 17:16 459 18 470, 471, 598, 615

Ancient Sources Index

18:1 1054 18:1–12 455 18:13–15 187 18:13–27 455 18:20 916 18:21 699 18:25 699 18:27 615 19 514, 517, 518, 903, 1098, 1136 19–20 614 19–22 511 19–24 69, 468, 471, 511–512, 513, 613, 1132 19:1 698, 707 19:2 612 19:3–6 465, 468 19:3–8 595 19:3–19 458 19:4 473 19:5 901 19:6 276 19:9 455, 469 19:9–10 455 19:9–19 517 19:10 455 19:18 520 19:20 552 19:20–25 455, 469 19:21 520 19:25 517, 518 20 112, 113, 506, 518, 519, 521, 901, 1136 20–23 369, 844 20–24 862 20:1 517, 518 20:2 321 20:2–3 473 20:2–17 141, 142 20:5 906 20:5–6 69 20:7 670 20:8 876 20:8–11 874–875, 885 20:10 880 20:11 142, 547, 876 20:12–17 515 20:17 141 20:18 516, 517, 519 20:18–19 515, 517 20:18–21 458, 516, 518–519, 521, 522

Ancient Sources Index

Exodus (continued) 20:19 278, 279, 515, 516, 519 20:19–21 484, 515, 517, 519, 520 20:20 517, 519, 521 20:21 488, 516, 519 20:22 515, 516, 517, 518, 522, 552 20:22–23 515 20:22–24 518 20:22–21:25 515 20:22–23:19 609 20:23 72, 515 20:23–24 52 20:23–26 49, 50, 51, 75, 662, 664 20:24 50, 51, 52, 63, 64, 664, 666, 678 20:24–25 279 20:25 52, 666 20:26 52 21–22 455 21–23 279, 280, 1136 21:1 50, 447 21:1–22:19 52 21:2 888 21:2–6 49, 495, 671 21:2–11 48, 49, 51, 110, 651, 654 21:7–11 49 21:12 75, 654, 656, 667, 668, 672, 678 21:12–14 654, 655, 656, 657, 658, 672 21:12–17 48, 668 21:13–14 50 21:15–16 667 21:15–17 75, 654, 655 21:16 668 21:17 667, 668, 681 21:18 668, 672 21:18–19 654 21:18–21 657, 658 21:18–32 48, 49 21:19 657 21:20–21 654 21:21 147 21:22 657, 664, 666, 667, 668 21:22–23 667, 678 21:22–25 654, 656, 657 21:23 655, 656 21:23–24 653 21:23–25 653, 654, 667, 668, 678 21:24 659 21:24–25 656, 659 21:25 660

1161

21:26–27 654 21:28 655 21:28–32 654 21:29 655 21:30 657, 672 21:31 657 21:32 655, 657 21:33–34 655 21:33–22:14 678 21:35–36 48, 655, 665 21:36 656 21:37 656, 665 21:37–22:3 48 22:2–3 655 22:4–5 655 22:5–6 187 22:6–7 48 22:6–8 655 22:9–12 48, 656 22:13–14 48, 656 22:15–16 48, 49 22:17–19 75, 655 22:20–23 52, 662 22:20–26 51 22:20–30 51, 662–663, 664 22:20–23:8 51 22:20–23:19 49, 50, 75 22:22–23 64, 669 22:22–24 952 22:23–25 666 22:24 52 22:25–26 52, 662 22:27 51, 52, 63, 661, 663, 664, 666, 667, 668, 669, 670, 676, 678 22:28–29 555, 556, 663, 681 22:28–30 51, 52, 663 22:30 276, 280, 664 23 1099 23:1–8 51, 53, 56, 64, 66, 662 23:1–19 455, 470 23:3–8 72 23:6–7 888 23:9 52, 662 23:9–12 51 23:9–19 51, 662–663, 664 23:10–11 52, 556, 662, 666, 667, 671, 873 23:10–12 666 23:11 271, 666 23:12 52, 662, 873–874, 876, 884

1162

Ancient Sources Index

Exodus (continued) 23:13 51, 52, 63, 72, 661, 662, 663, 664, 666, 668, 678 23:14 666 23:14–19 51, 52, 273, 680 23:15 273, 274 23:17 274, 275 23:17–19 663 23:19 276, 279, 663, 664, 681 23:20 58, 60, 63, 66, 71, 72, 73, 598, 820 23:20–21 66 23:20–28 61 23:20–29 66 23:20–33 47–85 23:21 63, 66, 73 23:21–22 55–56, 66, 72 23:22 60, 61, 66 23:23 58, 59, 60, 61, 66, 72, 73 23:24 61, 63, 66 23:25 53, 54, 63, 66 23:25–26 53, 66 23:26 53, 54 23:27 59, 60, 66, 75 23:27–28 58, 59, 66, 73 23:28 61, 66, 72, 789 23:29 70 23:29–30 61, 66, 68, 699 23:31 61 23:32 62, 66, 70 23:32–33 61 23:33 66, 70 24 281, 511 24:1–2 101 24:1–11 457 24:3 281 24:3–8 65, 101, 281, 507, 511 24:4–6 522 24:4–7 282 24:4–8 468, 511, 512, 514 24:7 281, 806, 1127 24:9–11 101, 468, 613 24:12 281, 646–647, 901 24:13 798 24:16 876 25–29 592 25–31 192, 341, 592, 613, 907, 1136 25–40 522, 592 25:1–31:17 647 25:8 1032

25:10 473 25:10–22 615, 906 26:3 341 26:7 376 26:8 376 26:33 158 26:34 158 26:34–36 187 28:3 1129, 1130 28:16–17 187 28:19–20 187 29 514 29:19–20 522 29:20–22 187 30 511 30:37–38 187 31:1–3 187 31:3 1129, 1130 31:6 1130 31:12–17 877, 883, 890 31:13 885 31:14 886 31:17 885 32 72, 74, 363–364, 613, 748, 1131 32–34 273, 274, 613, 614, 615, 869 32:1 71, 72 32:1–8 458 32:4 274, 275, 363–364 32:5 273 32:6 273 32:7 274 32:8 273, 274, 275 32:10–14 93, 94 32:11–12 613 32:13 93, 274, 275, 805 32:15–30 458 32:16 175, 459 32:23 71, 72 32:29–33 613 32:34 71, 72, 73, 598 32:34–35 75 33 616 33–34 74, 614, 1131 33:1 74, 75, 275, 614 33:1–22 458 33:2 71, 72, 73, 75, 76, 598, 614 33:3 71, 74, 75, 574, 598, 613 33:3–6 74, 75, 614 33:4 71, 75, 598, 613

Ancient Sources Index

Exodus (continued) 33:5 274, 275, 574, 613 33:6 275, 613 33:6–11 75 33:7 276, 706 33:7–11 76, 275, 276, 614, 616, 617, 706 33:8 616 33:10 76, 616 33:11 798, 803 33:12–17 613, 614 33:12–23 74 33:16 157 33:18 74, 472 33:22 74 34 70, 273, 274, 275, 277, 613, 614, 844, 868, 877, 903 34:1 75, 281, 459 34:1–4 455, 470 34:1–5 458 34:2–3 74 34:4 74, 459 34:5 75 34:5–8 1109, 1112 34:5–9 74 34:5–10 458 34:6 901 34:6–7 74, 458, 613, 1112, 1113 34:6–16 534 34:7 248, 906 34:9 74, 613, 901 34:10 614 34:10–16 69 34:11–16 69 34:11–27 71, 598 34:11–28 458, 470 34:12 70 34:12–15 70 34:14–26 273, 274 34:15 70 34:16 69 34:17–26 69 34:18–26 273, 874, 1098–1099 34:19–20 273, 867–868 34:19–26 1099 34:20 273 34:21 874, 876, 877 34:23 274, 275, 877 34:26 279 34:27 459, 614

1163

34:27–28 455 34:28 281 34:28–29 75 34:29–32 613 34:29–35 454 34:32 614, 615, 617 34:33–35 614 35–39 341 35–40 192, 193, 592, 613, 907, 1131, 1136 35–49 592 35:1–36:7 455 35:2 143 35:3–5 187 35:4–5 445 35:6–9 445 35:11 1033 35:31 1130 35:35 1130 36:10 341, 342 36:14 376 36:15 376 36:32–38 187 36:34 310 37:29 187 38:1–7 187, 277 38:26 700 39–40 1121, 1130–1131, 1132 39:8–19 187 39:21 189–190 39:32 1121, 1128, 1131 39:42–43 1128, 1129, 1134 39:43 1131 40 514 40:16–33 1128 40:17 193, 707 40:18 190 40:18–33 1033 40:20 193 40:22 193 40:29 277 40:33 1131 40:34 1134 40:34–35 1033 Leviticus 1–3 602 1–7 607, 876, 1136 1–15 370 1–16 1015

1164

Ancient Sources Index

Leviticus (continued) 1:1 276 1:1–8 126 1:4 968 4:13 926 5:1 952 5:1–5 916 5:6 926 5:7 926 5:12 157 5:15 926 7:18 968 8–9 602 8–10 592 8:36 1128 10:7 1128 11 276, 277, 280, 539 11–15 602 11:1–2 187 11:17 187 11:21–24 187 11:32 187 11:40–42 187 11:44–45 187, 527, 539 13–14 659 13–15 495, 876 13:6–8 187 13:15 187 13:17–18 187 13:51–52 187 15:31 916 16 602 16:11–12 187 16:20–22 677 16:21 679 17 276, 277, 1015, 1056 17–26 1055–1074 17–25 976, 1065 17–26 276, 370, 533, 602, 680, 861, 954, 1015–1039 17:4 1034 17:10–13 495 17:11 277, 278 17:14 277 17:15 280 18 538, 932, 933 18–20 1056 18–22 532, 541–542 18–26 1015

18:6–23 541 18:20 1023, 1047 18:24 541 18:24–25 555 18:26–28 187 18:26–30 541 18:27 541 18:28 541 19:3 890 19:7 968 19:8 669 19:12 670 19:19 534 19:26 1023, 1025 19:30 890 20 538, 932, 933 20:2 535 20:2–5 669 20:7 669 20:9 681 20:10 932 20:20 1047 20:22–26 527, 531, 532–535, 537–542 20:24 533, 538, 539, 541, 934 20:24–25 278 20:25 533 20:26 533, 541 21 658, 660, 661, 681 21–22 1056 21:2 933 21:6 670 21:6–9 542 21:17–23 658 21:18 659, 660 21:19 659, 661 21:20 659 21:21 659 21:21–23 659 21:22 659 21:23 659 22 681 22:1–9 926 22:1–16 924, 926 22:4 535 22:10 926 22:11 926 22:12 926 22:13 926 22:14 926

Ancient Sources Index

Leviticus (continued) 22:14–16 542, 924, 925, 926, 927 22:15–16 926 22:16 542, 927 22:20 659 22:20–25 658 22:21 659 22:22 659 22:23 659, 661, 968 22:24 659 22:25 968 22:27 968 22:27–28 663 22:27–29 680–681 22:29 890 23 680, 889 23–25 680, 1056 23:3 877, 888, 890 23:4 750 23:8 680 23:10 680 23:10–14 679 23:10–21 679–680 23:11 680 23:14 680 23:15 680, 1099 23:15–16 679 23:16 680 23:17 680 23:17–22 679 23:18 680 23:19 680 23:20 680 23:21 680 23:23 679 23:30 1042 23:32 874 23:34 526 23:35–36 554 23:36 917 23:37 890 23:39 554 23:39–43 554 23:42 526 23:42–44 187 24 70, 651–682, 681 24:1–2 187 24:2 149, 188 24:10–11 664–665, 667, 670, 678

1165

24:10–15 679 24:10–23 652, 678, 681 24:11 670 24:14 655, 674, 676, 677, 679 24:15 665, 666, 667, 668, 670, 676 24:15–16 661, 666, 667, 668, 670 24:15–17 668, 680 24:15–22 679 24:16 659, 661, 665, 666, 667, 668, 669, 670, 671, 678, 682 24:17 654, 655 24:17–18 654, 665, 668 24:17–21 654, 659, 667, 675 24:17–22 653, 667, 677 24:18 654, 655, 656, 657 24:19 658, 672, 675, 679 24:19–20 654, 672 24:20 655, 658, 679 24:21 665 24:22 682 24:23 674, 679 24:29 655 25 603, 667, 680, 889 25:2–7 666, 672 25:3–4 666 25:4 666 25:7–9 187 25:8 109, 110, 111 25:36 1024 25:36–37 1024 25:37 1024 25:38 322 25:39–46 671 25:43 1069 25:44–46 603 25:46 935 26 358, 369, 370, 590, 831, 902, 903, 943–984, 955, 956, 958, 1026, 1027–1029, 1032, 1035, 1041–1054, 1047, 1055–1074, 1056, 1075–1084 26:1–2 1041 26:1–39 945 26:2 890 26:2–35 979 26:3 1043, 1069, 1082 26:3–13 962, 1021, 1026, 1027, 1029– 1035, 1034, 1038, 1041–1042 26:3–33 978, 979 26:3–39 1044, 1046

1166

Ancient Sources Index

Leviticus (continued) 26:3–45 979, 980 26:4 902, 957, 962, 1027, 1028, 1031, 1043 26:4–5 1027 26:4–6 1069, 1070 26:5 1027, 1029, 1043 26:6 957, 962, 1027, 1043 26:6–8 962 26:7 1043 26:9 969, 970, 1030, 1031, 1043, 1044 26:9–12 1030, 1033–1034, 1038 26:11 902, 957, 962, 970, 1030, 1033, 1034, 1043, 1071 26:11–12 1034, 1043 26:12 276, 902, 971, 1030, 1034, 1044, 1071 26:13 1021, 1022, 1027, 1029, 1069, 1070 26:13–39 1067 26:14 962, 964, 979, 1042 26:14–23 962 26:14–33 949 26:14–38 1067 26:14–39 965, 969, 1042, 1065, 1066, 1067 26:15 902, 951, 962, 969, 981, 1043, 1082 26:16 1043 26:17 957, 962, 964, 1043 26:17–28 187 26:18 962, 969, 979, 1042 26:19 957, 962, 1043, 1047, 1069 26:20 962, 1043 26:21 962, 969, 1042 26:22 1043 26:23 902, 962, 969, 1042 26:23–24 962, 964 26:24 962, 964 26:25 962, 964, 969, 971, 1043, 1047, 1071 26:26 962, 1043, 1068, 1083 26:27 962, 969, 1042 26:27–28 962, 964 26:27–39 971, 972 26:28 962, 964 26:29 1082 26:30 902, 964, 981, 1022, 1042, 1043, 1047, 1053, 1071, 1072 26:30–31 957, 962, 1071 26:30–32 187 26:30–33 1022 26:31 977, 1043, 1061, 1080 26:31–32 1067

26:31–33 965, 1043, 1047 26:33 902, 962, 1068, 1080, 1081 26:33–34 979 26:33–35 978, 1067 26:33–37 978, 1067 26:34 968 26:34–35 977–978, 980, 983, 1066, 1067 26:34–43 886 26:35–39 978 26:36 1043, 1081 26:36–37 962, 979 26:36–39 965, 979 26:36–45 978 26:37 1044 26:37–38 978 26:38 1042 26:38–39 1042, 1043, 1044, 1046 26:39 982, 1042, 1044, 1048, 1052, 1058, 1059, 1066, 1067, 1068, 1081 26:39–40 978, 1067 26:39–41 1052 26:39–45 972 26:40 962, 971, 972, 979, 982, 1082 26:40–41 962, 971, 972, 973, 982, 1050, 1052 26:40–43 971, 1067 26:40–45 903, 967, 971, 972, 977–978, 979, 980, 981, 983, 1042, 1048–1054, 1065, 1066, 1067, 1076, 1081–1082 26:41 962, 964, 965, 968, 971, 972, 1082 26:42 969, 971, 1082 26:43 889, 902, 971, 981, 982, 1082 26:43–44 978, 1067 26:44 973, 977, 981, 1042, 1051, 1067 26:44–45 969, 971, 973, 978, 1067 26:45 971, 1054, 1082 26:46 615, 982 27:11 934 27:34 187, 615, 1128 Numbers 1–3 1136 1–10 868, 869 1:1 276, 707 1:1–5 187 1:46 700 1:53 916 1:54 1128 2 622

Ancient Sources Index

Numbers (continued) 2:34 1128 3:11–13 868 3:26–30 187 4:47–49 187 4:49 188 5:11–31 938, 940, 952 5:28 535 5:30 432 6:22–26 368 7 1136 7:1 187, 188 7:72 375 7:78–80 187 8:11–12 187 8:19 868 8:20 1128 8:22 1128 9:1 707 9:5 1128 9:13 669 9:15 187 9:15–23 576 9:17–19 187 9:22–23 187 10:2–3 187 10:11 698, 707 10:12 691 10:29 614, 615 10:29–32 615 10:29–12:8 475 10:32 615 10:33 73, 575, 576 10:33–35 473 10:33–36 615, 617 10:34 575, 576 10:34–36 574–576 10:35 574, 576 10:35–36 574, 575, 576 10:36 574, 575 11 616, 747, 749 11:1 575 11:1–3 747, 750 11:1–35 617 11:4–6 748 11:4–17 615 11:4–34 615, 616, 700 11:6 750 11:7–8 616

11:7–9 616 11:8 616 11:9 616 11:10 615, 616 11:11 615 11:11–12 75, 93 11:11–14 471 11:12 93, 615 11:13 615 11:14 615 11:14–17 75, 93 11:16 76, 636, 699 11:16–17 471, 616 11:20 74 11:20–21 614 11:21 700 11:24 699 11:24–27 76 11:24–29 616 11:24–30 75, 93, 471 11:28 700, 798 11:30 76 12 1009 12:1–8 789 12:1–16 706 12:2 706 12:4–5 76 12:7 549, 700 12:9–15 76 12:16 691 13 569 13–14 206, 220, 818–819 13:2 569 13:3 691 13:12 187 13:13 187, 188 13:14–15 187 13:15 188 13:16–25 187 13:17 569 13:21 569 13:22 361 13:26 691, 703 13:29–30 187 13:30 934 14 819 14:4 552 14:8 805 14:11–21 94

1167

1168 Numbers (continued) 14:11–25 93 14:14 74 14:16 74, 93 14:18 74 14:19 74 14:21–22 74 14:22–23 74 14:23 74, 93 14:23–24 818 14:24 807, 934 14:25 707, 708 14:31 96 14:34 220 14:44 473, 803 15 278, 540 15:20 706 15:20–21 706 15:26–28 187 15:32–25 883, 890 15:37–41 278 15:40 278 15:41 278 16 249, 250, 496 16–17 696, 705 16:14 934 17–18 859, 868, 869 17:20–24 187 17:24–25 697 17:26 1128 17:27–28 697 18 1063 18:5 916 18:8–12 923, 924 18:15 555, 556 18:19 924 19:1–22 707 20 702, 748 20–21 693, 695, 705, 709, 859 20:1 698, 703, 706, 707, 708 20:1–13 748 20:3 697 20:4 697 20:4–5 748 20:5 746–747, 749 20:6 697 20:8 697 20:8–9 697 20:8–11 552

Ancient Sources Index

20:10 697 20:12 697, 748 20:14 692, 703 20:14–16 703 20:14–21 692, 701, 746 20:16 71, 73, 692 20:17 699 20:18 702 20:20–21 702 20:21 702, 747 20:22 693, 703, 706 20:22–27 693 20:22–29 697, 746 20:22–22:1 689 20:24 697 20:27 1128 21 688, 717, 720, 745–760, 748, 751, 758, 759, 761–776, 902 21–22 724 21–24 475 21:1–3 219, 691, 769 21:4 690, 693, 694, 701, 746, 747 21:4–5 747 21:4–9 746, 747, 748, 750, 753–758 21:5 746, 748, 749 21:5–9 694 21:6 746, 747, 750, 753–754 21:7 747 21:8 746, 754 21:8–9 747 21:9 746, 754 21:10 690, 694 21:10–11 693, 694, 764, 765, 769, 774–775 21:10–30 219 21:11 690, 693, 694, 766, 768 21:11–12 768 21:11–13 695 21:12 691, 693, 748 21:12–13 690, 695, 764, 765, 768 21:13 691, 693, 695, 705, 761, 766, 768 21:13–15 696 21:16 689, 691 21:16–19 693 21:18 693 21:18–20 690, 691, 764, 765, 768 21:20 698, 761, 766, 1010 21:21 695, 699 21:21–25 692 21:21–30 692

Ancient Sources Index

Numbers (continued) 21:22 699 21:23 691, 693, 723 21:23–26 718 21:24 695, 719, 761, 767 21:25 695, 717, 719 21:26 692, 695, 699, 719 21:27–30 692, 718, 771 21:28 695 21:28–29 902 21:29 38, 699, 719 21:32 695 21:33 1010 21:33–35 94, 626, 699, 1010 21:34 699 21:42 698 22 720 22–24 219, 581, 698, 721, 723, 944, 1009 22:1 761, 1010 22:1–2 766 22:2 699, 1010 22:2–24:25 1010 22:3 699 22:4 699 22:5 699 22:6 699 22:22–35 71 22:36 720–721 22:41 698 23–24 1113 23:10 433 23:13–14 693 23:14 694, 698 23:21 1112 23:23 756 23:28 698 23:63 694 24:2 423 24:7 535 24:15–17 147 24:17 902 25 629, 1013 25:1 797 25:9–34 918 25:12 1012, 1031 25:12–13 1012, 1013 26:3 694 26:23–24 99 26:51 700

26:59 707 27 339, 769 27:1–11 340, 627, 933 27:3 933 27:7 934 27:8 934 27:8–11 933, 934 27:11 187, 188, 933, 935 27:12 689, 769 27:14 703, 706 27:18 1129 27:18–23 798 27:20 614 28–29 554 28:9–10 877 29 554–555 29:20 375 30:1 279 31:2 174 31:8 703 31:12 694 31:31 1128 32 619–629, 726, 774, 805 32–33 902 32:1 613, 624 32:1–38 620, 621, 623 32:2 622, 624, 626, 627 32:3 38 32:4 626 32:4–5 626 32:5 626, 627, 805 32:6 624 32:7 805 32:7–15 623, 628 32:9 805 32:13 816–817, 817 32:16 621 32:16–19 622 32:17 621, 622, 805–806 32:18 626, 628 32:19 626 32:20 628 32:20–21 627–628 32:20–23 628 32:20–24 622, 626 32:21 627, 628 32:22 625, 626, 627, 628 32:22–29 821, 822 32:23 621, 623

1169

1170

Ancient Sources Index

Numbers (continued) 32:24 621 32:25 621 32:25–27 622 32:26 805 32:27 621, 627 32:28 625 32:28–30 622 32:28–32 625, 626 32:29 621, 622, 625, 627–628 32:29–30 621, 622, 628 32:30 621, 623, 625 32:31 621, 625, 626, 1128 32:31–32 621–622, 625 32:32 625 32:33 620, 621, 624, 626, 628, 716 32:33–38 714 32:34–38 621, 624, 715–716, 773 32:38 38 32:39 620 32:39–42 620, 626 32:40 620 32:41–42 620 33 205, 688, 691, 693, 697, 709, 772 33:3 698 33:3–37 689 33:5–48 690 33:8 551 33:12–14 691 33:18–30 691 33:36 706 33:36–37 703 33:37 689 33:37–49 693 33:38 698, 707 33:38–39 697 33:41–42 691 33:41–48 689 33:41–49 689 33:44 769 33:46–47 902 33:47 38 33:50–56 1042 33:53 935 34–35 785 34:1–38 629 34:16–29 625 34:20 319 35 672, 918, 933

35:9–34 672, 785 35:11 672 35:11–15 919 35:15 672 35:16–18 672 35:17 672 35:21 672 35:30 672, 933 35:31 672 35:31–32 672 35:33 931 35:33–34 928, 931 36 341 36:1 339, 342 36:1–2 187, 188 36:1–12 339–340 36:2 339 36:3 339 36:10 1128 36:13 1128 36:26 805 Deuteronomy 1 377–378, 819 1–3 876 1–4 448 1–11 902 1:1 638, 642, 698, 704 1:1–2 784 1:1–5 205 1:1–31:13 639 1:1–34:12 784 1:2 374, 378 1:3 278, 375, 378, 1123–1125, 1129 1:4–4:44 784 1:6–8 552 1:7 808 1:7–8 801 1:8 934 1:8–17 471 1:9–18 1124 1:13 1130 1:21 934 1:30–32 71 1:33 552 1:35–36 818–819, 819 1:38 798 1:39 96, 97, 934 1:40 707, 708

Ancient Sources Index

Deuteronomy (continued) 1:46 692, 703 2 765–766, 770–771, 774 2–3 768 2:1 692, 702 2:4–8 701 2:8 692, 766, 767 2:9 717 2:9–33 716–717 2:13 692 2:14 748 2:16–23 766, 767, 772 2:19 717 2:24 187, 705, 766, 692 2:24–36 692 2:25 702 2:26 692 2:27 766 2:28–29 702, 704 2:32 692, 723 2:36 187 2:37 766 3:1–3 94 3:1–7 699 3:8 699 3:8–20 805 3:11 159 3:12 715 3:12–18 715, 716 3:12–20 624 3:14–17 715 3:16–17 715 3:17 693, 694, 699 3:18 321, 805, 806 3:19 805 3:20 627 3:21 798 3:27 423, 425, 699 3:28 798 4 307 4–5 902, 1098 4:1 934 4:2 557 4:11 143 4:14 278 4:19 423, 424 4:25–31 639, 640 4:27–28 238 4:29 903

1171

4:31 321 4:34 906 4:38 1099 4:41 919 4:41–43 918 4:43 805 4:44–26:19 280 4:45–26:19 784 4:49 693, 699 5 112, 113, 506, 518, 519, 521, 636, 901 5–26 647 5:1 877 5:1–6:1 136, 143, 488 5:4 519 5:5 143, 519 5:6 321 5:6–21 141 5:8 141 5:9 141 5:9–10 876 5:12 143, 876, 880 5:12–15 874–875, 884 5:13–14 875 5:14 141, 143, 876, 880 5:15 142, 143, 884 5:18 141 5:19 141 5:20 141 5:21 141 5:22 143 5:22–24 519 5:22–31 518, 519, 522 5:23–31 278 5:24 520 5:24–26 520 5:24–27 515, 516, 517, 518, 519–520 5:24–31 516 5:25 520, 1127 5:25–31 278, 280 5:26 520 5:27 279, 515, 806, 1126 5:27–28 521 5:28 877 5:28–29 515, 517 5:28–30 520–521 5:28–31 515 5:29 516, 520–521, 903 5:29–33 903 5:30–31 515, 516, 517, 521

1172

Ancient Sources Index

Deuteronomy (continued) 5:31 279, 877, 951 5:32 802 5:33 903 6 368 6–8 876 6:3 877 6:4–9 368 6:22 551 7 69, 70, 71, 902, 926 7:1 69, 537 7:1–3 541 7:1–4 230 7:1–5 69 7:2 69 7:3 69, 537, 553, 555 7:4 69, 555 7:5 69, 751 7:6 69, 539, 540, 923, 925, 926, 938 7:8 69, 321 7:9–10 69 7:11–12 69 7:13 69 7:13–14 69 7:15 69 7:16 69, 1080 7:17–21 69 7:18–19 69 7:20 789 7:21 69 7:22 70 7:22–26 69 7:23–24 69 7:25–26 69 8–10 448 8:1–5 488 8:4 552 8:5–10 136, 143 8:6 143 8:7 143 8:11 951 8:15 754 8:18 321 9 748 9:1 280 9:4–5 1099 9:22 747 10 902 10:1–5 281

10:6 697, 707 10:15 549 10:22 180, 181 11 876 11:24 802 11:24–25 800, 801 11:31 934 12 277, 279 12–18 368 12–26 875 12:1 321 12:5–6 556 12:6 555 12:8–12 810 12:20 1099 13 67, 368, 673, 674, 676, 677, 906 13:1 270 13:1–5 1124, 1125 13:4 1124 13:5 673 13:6 674 13:7–12 673 13:9 1080 13:10 673, 676 13:10–12 673 13:11 673 13:12 674 13:14 906 13:17 849, 906 14 277, 280 14:2 539, 540 14:3–20 280, 539 14:3–21 280 14:8 1099 14:20 280 14:21 280, 539, 540, 664 15 671 15:1 875, 888, 940 15:1–2 556 15:12 875, 940 15:12–17 110 15:12–18 495, 651, 671 15:15 495 15:19–23 659 15:21 659 16 680, 1097 16:1–17 680 16:3–4 680 16:6–7 917, 1097

Ancient Sources Index

Deuteronomy (continued) 16:7 529, 680, 1098 16:8 873, 875 16:8–9 680 16:13 526 16:15 554 16:18 636, 916 16:18–20 916 17 674, 675, 676, 677 17–18 1124 17:1 659 17:2–7 672, 673, 674, 675, 679 17:5 673 17:5–7 673, 674, 679 17:6 673, 677, 679 17:7 673, 676 17:8–9 916 17:8–13 916 17:9 916 17:9–22 1124 17:12 806 17:14 817 17:14–20 816, 817 17:18 169 17:18–20 803–804 17:19 645, 895 18 518, 521, 756 18:12 1099 18:15 846, 884, 1006, 1124, 1133 18:15–22 877, 1125 18:16–17 516 18:16–18 885 18:18 516, 521, 1006 18:18–21 516 18:18–22 515, 516, 517 18:19 806 19 672, 674 19:1–3 918 19:1–13 672 19:2 919 19:13 1080 19:15 672, 673, 677, 679 19:15–21 672, 677, 679 19:16–21 672, 679 19:19 672, 674, 675, 677, 679 19:20 674 19:20–21 187 19:21 675, 677, 679, 1080 20:1 187

1173

21:11–14 541 21:19 674 22:22 932 22:23–27 932 22:24 674 23:1–9 849 23:2 659 23:2–3 659 23:3 537 23:4–6 559 23:5–6 721 23:6 907 24 932 24:1 931 24:1–4 849, 905, 928, 929, 930, 932, 933, 938, 940 24:2 931 24:4 929, 931, 932 24:9 707 24:16 248 24:17 941 25:11–12 665, 677 25:12 1080 25:13–14 887 25:13–15 887 25:13–16 886 26–34 902 26:5–9 582, 824, 826 26:9 805 26:17 951 26:19 926 27 103 27:1–26 784 27:15–26 944 28 67, 358–359, 368, 369, 647, 816, 817, 902, 903, 908, 948, 949, 950, 953, 954, 955, 956, 957, 958, 979 28–30 955 28:1 957 28:1–68 945 28:1–69 784 28:6 879 28:8 957 28:9 926 28:15–68 1045–1046 28:16–19 944, 957 28:19 879 28:22 1047 28:24–25 957

1174

Ancient Sources Index

Deuteronomy (continued) 28:29 803 28:36–37 238 28:38 535 28:48 903, 957 28:49 903, 904 28:58 895 28:59–68 896 28:64–68 238 28:65 957 28:69 447, 876 28:69–30:20 634 29:1 642, 644 29:7 805 29:8 802 29:9 644 29:12 100 29:21–27 238 30 307, 908 30:1–7 238 30:1–10 639, 978, 1046 30:2 976 30:3 639 30:3–4 307 30:6 965, 976 30:10 976 30:12–13 1126 30:19 1134 30:19–20 1126 30:20 100 31 276, 640, 1125, 1126 31–32 631–650, 903 31–34 642 31:1 631, 634, 640, 642, 645, 649 31:1–6 633 31:1–8 73, 789 31:1–43 632 31:1–32:47 632–634, 637, 638, 640–642, 645, 646 31:2 649 31:2–6 641 31:2–8 798 31:3 71, 73 31:6 71, 73 31:6–8 801 31:7–8 633, 641 31:8 71, 73 31:9 281, 631, 633, 634–635, 636, 641, 642, 643, 644, 647, 649

31:9–13 636, 784 31:10 554, 635, 647, 649 31:10–13 633, 641, 642 31:12 803, 895 31:14 633, 641, 643, 649, 798 31:14–15 633, 635, 638, 640, 641 31:14–21 643 31:14–23 635, 643 31:14–24 789 31:15 76 31:16 1125 31:16–21 633, 635, 638, 641 31:16–22 640 31:17 639 31:18 639 31:19 635, 637, 645, 649 31:19–22 281 31:21 635 31:22 633, 635, 637, 638, 641, 643, 644, 649 31:23 633, 635, 640, 641, 643, 649, 798, 801 31:24 281, 643, 644, 647, 650, 1131 31:24–27 635–636 31:24–29 634, 641, 643, 644 31:24–32:44 635, 638 31:25 636, 650, 789 31:25–27 784 31:26 650 31:26–29 636 31:27 1125 31:28 636, 644, 650, 789 31:28–29 636 31:29 647, 784 31:30 634, 637, 641, 644, 647, 650, 789 31:30–32:44 644 32 595, 647, 769, 903, 1113, 1117 32:1–33 637, 789 32:1–43 193, 634, 641 32:1–44 1109 32:4–6 553 32:8 137–138, 139, 140, 180, 194, 640 32:9–10 193 32:22 903 32:23–24 1083 32:34–43 637 32:35 58 32:37–43 193 32:39 640

Ancient Sources Index

Deuteronomy (continued) 32:42 926 32:43 138–140, 193, 194, 640 32:44 634, 637, 641, 642, 644, 645, 646, 647, 650 32:45 634, 637–638, 642, 645, 650 32:45–47 639, 784 32:46 650, 895 32:46–47 634, 638, 642 32:48–52 1133 32:49 689, 700, 769 32:51 697, 706 33 607 33:2 343 33:8–11 148, 1013 33:9 1013 33:9–10 1013 34 219, 448, 581, 769, 815, 1132, 1136 34:1 700, 769, 1133 34:1–4 1133 34:1–5 700 34:1–12 784 34:4 1112 34:5 700, 799 34:7 475, 1133 34:7–9 1133 34:8–9 813, 815 34:9 798, 806, 1121, 1123–1125, 1126, 1127, 1129, 1130, 1132–1133, 1134 34:10 845, 846, 1006, 1133 34:10–12 640, 813, 815, 1123 34:11 551 34:12 1128 Joshua 1 795–812, 1122 1–5 811 1:1 799, 807 1:1–2 808 1:1–9 822, 823 1:1–18 812 1:2 800, 804, 807, 822, 823 1:2–5 635 1:2–9 799, 800 1:3 800, 802, 806, 822, 823 1:3–4 800, 802 1:3–5 801 1:3–6 801–802, 808 1:3–8 804

1:4 800, 807 1:4–6 801 1:5 800, 806, 807 1:5–6 801 1:6 635, 801, 802 1:7 802, 803, 806, 807 1:7–8 802, 803–804, 808, 809, 846 1:8 802, 803, 807, 808 1:9 804 1:9–11 808 1:10–11 799, 804 1:10–12:24 635 1:11 804 1:12 805 1:12–15 799 1:12–17 808 1:12–18 799, 805 1:13 805, 807, 846 1:14 805, 807 1:15 807 1:16–17 1127 1:16–18 799 1:17 800, 806, 807 1:17–18 806 1:18 806, 808 2 475, 805 2–12 816, 817 2:1 797 2:9 58 3–4 816, 817, 819–820, 820 3–5 810 3–10 822, 823 3:1–5:12 811 3:2 804, 812 3:6 71 3:14 319 4:2 934 4:19 820, 821 5 819, 860 5:1–10 820 5:2–9 820 5:10–12 820, 821 5:11 750 5:12 819 5:13 423, 820 5:13–15 71, 819, 820 6–8 810 6:4 873 6:7 71

1175

1176 Joshua (continued) 6:15 873 6:16 873 6:26 952 7 951, 955 8:34 803 8:35 637, 1127 8:35–40 103 9:22–23 952 10 475, 810 10:10 71 10:28–39 822, 823 10:28–42 822, 823 10:40–42 822–823 10:42 822, 823 11:12 1127 11:15 1127 11:16–23 822, 823 11:23 822, 823 12–13 622 12:3 693, 699 12:6 805 13 307, 774 13–19 822, 823 13–22 785 13:1–7 635 13:9–12 715, 716 13:15–28 773 13:15–31 715 13:15–32 715 13:20 693, 699 14 820 14–21 635 14:1 625 14:1–2 821 14:1–15 819 14:13–14 818–819, 819 15:51 374 17:4 625 18 821 18:1 625, 820, 821, 822 18:7 620 19:51 625, 821 20 917–919, 918, 919 20:1–3 918 20:4–6 918 20:6 918 20:7 919 20:7–8 918

Ancient Sources Index

20:7–9 785, 918 21:1 625 21:43 822 21:43–45 810, 822, 823, 824 22 774 22:1–6 627 22:9–34 308, 629 23 69, 70, 823, 824 23:1–3 823 23:4 70 23:5 70 23:6 802, 803 23:7 70 23:9 823 23:9–13 70 23:10 70 23:11 70, 823 23:13 70 23:14–16 823 23:16 70 24 69, 70, 598, 599, 817, 818, 823–824, 824, 825–826 24:1 824, 826 24:2 231 24:3 535, 823, 825 24:3–14 823, 824 24:5 97, 101, 825 24:5–7 97 24:6 97 24:6–7 97, 823, 825 24:8 70, 823, 825 24:9–10 823, 825 24:11 70, 823, 825 24:12 70, 823, 825 24:17 70 24:18 70 24:19 70 24:24 70 24:25 70, 96, 97, 101, 528, 823, 825 24:26 823, 825 24:28 789 24:29 821 24:29–30 822 24:31 825 24:32 239, 598, 817, 818 Judges 1:26 808 2:1–5 61, 71, 73, 598

Ancient Sources Index

Judges (continued) 2:1–7 69 2:2 70 2:3 70 2:6–22 823, 824 2:7 1126 2:8–9 822 2:12–14 98–99 2:16 98–99 2:20 822, 823, 880, 885 2:20–23 69 3 721 3:4 880, 885 3:9 98–99 3:11 822, 823 3:30 822, 823 4:14 71 5:23 71, 945 5:31 822, 823 7:4 434 8:28 822, 823 10–11 234 10:1 99 10:3–5 99 11 768, 771, 772 11:1–11 719 11:11 234, 704 11:11–31 720 11:13 772 11:16 704, 707, 708, 709 11:16–18 772 11:17 692, 701, 702, 703 11:17–20 692 11:18 701, 705 11:19–22 692 11:19–23 772 11:24 704 11:25–26 719, 720 11:30–31 704 11:32–33 719 11:34–40 720 13 206 13:2 415 13:6 71 13:9 71 14:15 873 16 206 17–21 308 18:30 240

19:17 423 21:18 945 1 Samuel 1 868 1–7 868 1:1 868 1:2 415 2:5 415 2:16 333 2:27–36 790 3 868 4–6 724 6 755 6:13 423 8 816, 817 8:7–9 279 9:27 333 12:8 73 13 790 13:20 549 14:21 944 14:24–46 951 14:49 415 15 790 15:23 158 15:26 158 16 790 20:34 135 22:14 549 24 438 25:24 428 29:9 71 2 Samuel 5:8 91 7:5–11 279 7:11 91, 98 7:15–16 91 8:11 821, 822 11–20 1137 12:18 873 13:34 423 14:17 71 14:20 71 15:25 91 16–18 103 18:24 423 19:28 71

1177

1178

Ancient Sources Index

2 Samuel (continued) 21 790, 951 22 955 22:46 309 24:16 424 24:16–17 71 24:24 939 1 Kings 1:31 427 1:51 333 2 815 2:3 802 2:27 790 3 94–95 3–15 445 3:4 95 3:22 339 5:15–8:66 91 6:14 459 6:38 374 8 810, 951, 954–955, 956, 958–959 8:10 95 8:12–13 576 8:31 959 8:31–32 952 8:33 958, 959 8:34 309 8:35 958, 959 8:37 953 8:46–53 307 8:47–48 958 8:53 576 8:58 880, 885 9:8 1081 11 816, 817 11:7 38 11:33 38 12 759, 815, 816 12:25 858–859 12:25–13:34 859, 869 12:28 363–364 13 790 13:3 322 14:29 204 16:16–28 38 16:31 39 16:34 790 18 992

19 522 19:5–8 71 19:13 987 19:15–16 41 19:17 41 19:35 71 20 39, 40, 576 20:1–6 40 20:29 873 20:34 39, 40 21 522, 576 21:3 935 21:4 935 21:10 670 22 39, 857, 859, 864 22:1 40 22:5 333, 334 2 Kings 1:1 38 1:8 987 3 724, 857, 864 3:1 41 3:4–5 38 3:4–27 722 3:5–6 38 3:6–27 38 3:24–25 722, 723 3:27 722 4:23 875, 879 4:37 428 8 864 8:7–13 41 8:15 41 8:16–24 864 8:18 865 8:20–22 858 8:25 41 8:26 864 8:26–28 41 8:26–29 44 9 522 9–10 859 9:1–6 41 9:14–15 41 9:21–29 42 9:29 374 9:32 425 10:32–33 864

Ancient Sources Index

2 Kings (continued) 10:32–34 42 10:33 620, 805 11 879–880 11–12 445 11:5 875 11:7 875, 880 11:7–9 879 11:9 875, 880 11:10–12 879 12:17–18 42 13:1–9 42 13:10 45 14–25 445 14:1–22 857, 865 14:7 858 14:22–25 865 14:23–29 857, 865 15:8–12 870 15:13–16 870 15:17–24 870 15:23–26 870 15:27–31 870 15:29 240 15:30 870 16:18 875 17 726, 859 17:1–6 240, 870 17:13 880, 885, 1124 17:14 885, 1125 17:23 877, 1124 17:28 727 17:32 726, 727 18:4 748, 751–753, 752, 754, 758–759 18:9–12 240 18:32 240 19:5 547 19:14 330 19:17 42, 44 19:22–32 240 19:35–37 240 21:10 1124 22–23 816, 817 22:13 803 23 364 23:4–20 752 23:8 790 23:13 38 23:16–20 790

1179

23:21–23 95–96 23:36 374 24:2 1124 24:8–17 240, 387 24:18 374 25 815, 816 25:2 376 25:25–26 241 25:27 241 25:27–30 241, 304, 365 Isaiah 2:5–22 65 2:10 65, 66 2:19 65, 66 2:21 65, 66 5 981 5:20–25 983 5:24 894 5:25 956, 957, 958, 959 5:25–29 954, 956, 958, 959, 979 5:25–30 1043 6 759 6:10 967 6:13 536 9–10 981 9:1–6 240 9:7–10:4 954, 956, 957, 958, 959, 979, 983, 1043 9:11 958 9:11–12 959 9:12 967 9:16 958, 959 9:20 958, 959 10:4 958, 959 11:12 307 14:29 754 15 724 15:1–16:1 716 15:4 718 16:8–9 718 19:22 967 24:5 931 24:22 989 29:22 435 30:6 754 31:6 967 37:21–32 240 37:36–38 240

1180 Isaiah (continued) 40 304 40–48 434 40–55 434 40–66 331, 335, 354, 426, 427 40:1–2 982 40:26 425, 426 41:8–9 435 41:9 435 41:10 354 41:14 435 41:16 1080 41:18–19 436 42:1 324 42:15–26 436 43:5 434 43:6 435 43:11 833 43:14 427, 434 43:19–20 436 44:15 426, 434 44:17 426, 434 44:19 434 45:6 434 45:14 426 45:17 434 46:4 434 46:6 426, 434 46:7 434 46:8 434 48:10 434 49 434 49–55 434 49:6 435 49:7 427 49:12 434 49:18 426 49:21 436 49:23 427 51:2 436 51:5 436 51:6 426, 436 51:9 434 51:22 354 52 774 52:2 889 52:4 889 52:6 889 52:12 71

Ancient Sources Index

52:13 889 53:4 434 53:11 434 54:3 935 55:10 427 56–66 312, 434 56:1–7 849 56:7 535 58:8 71 58:12–14 888 59:8 310 59:21 803 60:4 426 60:4–11 881 60:7 535 60:14 427 61:4–7 881–882 61:8–9 535 61:9 535, 882 61:10 882 63:9 71 65:9 427, 535 65:22–23 535 65:23 535 66:11 354 66:18–21 535 66:19 366 66:22 535 66:22–24 535 66:23 426, 874, 887 Jeremiah 1:1–4 900 1:3 377 1:5 899 1:9 884, 910 1:15 878 1:16 895 2–10 927 2:2 923 2:2–3 923–924 2:3 922–927, 932, 937, 938 2:5 893 2:6 216 2:8 837, 891–893, 896, 898, 899, 900, 909, 910, 911 2:20–25 938, 940 3 905 3:1 929, 930, 931, 938, 940

Ancient Sources Index

Jeremiah (continued) 3:1–2 932 3:1–5 849, 922, 928–933, 937 3:2 425, 931 3:4 929 3:4–5 928, 929, 930, 932 3:5 929 3:9 931 3:12 905, 929 3:12–23 967 3:16 906 3:16–17 906 3:17 906 4 904 4:2 900, 904 4:14 426 4:23–25 900–101, 904 4:23–26 910 5 902, 904 5:1 901 5:2 1080 5:4 901 5:14 910 5:15 903, 904 5:24 902 6:8 426, 902 6:16 894 6:19 894 7 724, 896 7:2 878 7:9 899, 901, 932 7:22 901 7:22–23 905, 906 7:23 901, 902, 903 7:24 880 7:26 880, 886 7:28 886 7:34 902 8:8 894, 898, 909 9:2 899 9:3 437 9:5 899, 909 9:11 894 9:11–12 895 9:12 894, 895, 896, 897, 898 9:23 899 11:3 944 11:4 880, 885 11:8 880

11:25 880 13:14 1080 13:15–17 1080 13:20 426 13:27 426 14:7–9 930 14:19 902 14:19–22 930 15:1 907 15:5 426 15:6 1080 15:14 903 16 901 16:2 901 16:5 901 16:10–11 895 16:11 895–896 16:14 901 16:14–15 906 16:15 901 16:19 910 17 877–890, 884 17:1 898 17:5–7 952 17:10 434 17:19–20 878 17:19–27 884–885, 905 17:20 879 17:21 886 17:21–22 881 17:21–23 880 17:23 886 17:24–27 889 18:16 1081 18:18 896, 898 19:1–3 878 19:4 899 19:8 1081 19:15 886 20:14–16 944 21:5 906 22:1 878 22:1–2 878 22:4 880 22:18–30 769 22:24–30 240 22:30 879 23 992 23:7–8 576, 906

1181

1182 Jeremiah (continued) 23:9–32 910 23:11 931 23:13–15 932 23:15 931 23:40 576 24 900 24:1 240 24:7 899 25 576 25:4 880 26 445, 724, 896 26–52 354 26:4 896, 897 27:5–8 936 27:20 240 28:4 240 28:14 903 29:1 330 29:2 240 29:7 907 29:13 903 29:25 330 29:29 330 30 906 30–31 897 30:8 1029 30:16 926, 936 30:18 849, 906 30:21 906 31 900, 978 31:15 901 31:22 907, 911 31:27 535 31:29–30 906 31:31 907 31:31–34 897, 899, 909 31:32 899 31:33 896, 897, 898, 899, 901 31:33–34 899, 909 31:34 899, 901 31:35–37 930 32:7 934 32:16–25 936 32:17–21 547 32:20 551 32:22 805 32:23 897 32:33 886

Ancient Sources Index

32:39 903 33 536, 537 33:4 898 33:10–11 890 33:14–26 536–537, 537 33:15–16 536 33:17–18 536 33:18 890 33:20 435 33:22 536 33:23–26 930 33:26 536 34:8–22 932 34:12 940 34:14 880 34:18 411 35:13 886 36–43 445 36:30 879 37:8 1029 37:16 989 38 990 38:6 989 38:6–7 989 38:9–11 989 38:13 989 38:33 898 39:2 377 39:23 898 41:1 905 41:5 905 41:16–18 241 42–44 906 44 897 44:3 899 44:5 880 44:10 897, 900 44:23 897, 900 46–51 576 46:9 892 46:18 473 48 902 48:1–6 724 48:1–47 716 48:2 718 48:10 944 48:21–24 724, 902 48:22 689, 902 48:30–37 724

Ancient Sources Index

Jeremiah (continued) 48:34 718 48:45–46 902 49:1–2 922, 933–936, 937, 938 49:1–6 935 49:2 936, 938 49:32 1080 49:36 1080 50:15 354 50:16 892 51:10 898 51:23 898 52:1 374 52:5 377 52:30 241 Ezekiel 1–24 1081 1–39 1036 1:1 240 4–24 1023 4:16 1068 4:17 1048, 1058, 1059, 1068 5–9 1080 5:2 1069 5:5–6 1078 5:5–17 1078–1083, 1083 5:6 1082 5:6–7 1079 5:7 1078, 1082 5:7–17 1078 5:8 1053 5:9 1079, 1082–1083 5:10 1068–1069, 1079, 1080, 1081, 1082 5:11 1080 5:12 1069, 1080 5:13 1080 5:14 1053, 1080 5:15 1079 5:16–17 1083 6 1052–1053 6:2 1053 6:2–7 1071 6:3 1022 6:3–6 1022 6:4 1022 6:4–5 1053 6:5 1022, 1071, 1072 6:6 1072

6:14 310 7:2–3 848 7:4 1080 7:8 1080 7:9 1080 7:24 1069 8:12 1058 8:18 1080 9:5 1080 9:9 1058 9:10 1080 11:9 1079 12:14 1069, 1080 13:10 1082 14:5 1079 14:21 953 16:30 354 16:60 1082 17:12 240 17:20 1082 17:21 1069 18 1063 18:3 1048 18:6 1023 18:7 310 18:8 1024 18:9 310 18:11 1023 18:12 310 18:15 1023 19:9 354 20 216, 884, 1054, 1063 20:5 216 20:7–8 1054 20:8 1080 20:9 1054 20:10–13 884 20:13 1082 20:14 1054 20:16 1082 20:17 1080 20:21 1080 20:22 1054 20:40 1028 20:41 1053 22 1063 22:11 1023 22:12 1024 22:16 1053

1183

1184

Ancient Sources Index

Ezekiel (continued) 23:10 1079 24:14 1080 24:21 1047, 1069 24:33 1048 25:11 1079 26:1 377 27:13 366 27:19 366 27:24 354 27:29 892 28:22 1079 28:25 1028, 1053 28:26 1079 30:6 1047, 1069 30:14 1079 30:18 1047, 1069 30:19 1079 30:20 375 31:1 375 33 1046–1052, 1050, 1076 33:1–9 1051 33:2 1047 33:5 1046 33:10 1046, 1048, 1050, 1058, 1059 33:11–20 1050 33:12–20 1052 33:15 1047 33:17 1049 33:20 1049 33:24 364, 437, 810 33:25 1023, 1047 33:25–26 1022–1023 33:26 1023, 1047 33:27 1023 33:27–29 1023 33:28 1047, 1069 33:28–29 1047 34 1027–1029, 1037–1038, 1068, 1069–1070 34:4 1069 34:10 1028 34:11–16 1028 34:23 1028, 1037 34:23–24 1026, 1035, 1037 34:23–30 1038 34:23–31 1027, 1038, 1039 34:24 1028, 1036 34:25 1026–1027, 1029

34:25–28 1070 34:25–29 1069 34:25–30 1021, 1026, 1028, 1030, 1034– 1035, 1036, 1067, 1069, 1070 34:26 1027, 1028 34:26–27 1027 34:27 1021, 1022, 1027, 1028, 1029, 1070 34:28 1027, 1029 34:31 1026 36 965, 1050–1051 36–40 1019, 1037 36:3 1082 36:11 1024 36:16–36 1051 36:20–23 1053 36:22 1084 36:23–38 1019 36:27–35 1084 36:34 1081 37 1019, 1029, 1037–1038 37:14 1037 37:15–19 1037 37:15–23 1037 37:15–28 1037 37:20–23 1037 37:22 1037 37:23 310, 1030 37:24 1028, 1030, 1032, 1035, 1036, 1037 37:24–25 1032, 1036, 1037 37:24–28 1026, 1029–1039 37:25 1028, 1029, 1030, 1032, 1036, 1037 37:25–28 1030 37:26 1030–1035, 1071, 1083 37:26–27 1030, 1032, 1033–1034, 1071, 1076 37:26–28 1071 37:27 1030, 1032, 1033, 1071 38–39 1019, 1063 38:4 892 38:17 1028 39:27 1053 39:28 307 40–48 1032, 1033, 1034, 1060, 1061 40:2 1033 40:4 158 40:49 377 42:13 924 43:7 1034 43:8 1080

Ancient Sources Index

Ezekiel (continued) 44:7 1082 44:9 1082 44:22 542 45:17 874 48:8–22 1032, 1033 Hosea 1–14 856 1:1 854, 856 1:2–5 854 1:2–2:3 856 1:2–3:5 853 1:2–14:9 856 2:4–15 930 2:4–3:5 856 2:4–14:9 856 2:9 930 2:15 930 2:16–25 930 3:1–5 930 3:5 982 4:1–19 856 5:1–7 856 5:1–14:9 856 5:8–12 855 5:8–7:16 856 5:13 855 5:15 926 5:25 982 7:8–10 855 7:11–12 855 8:1–14 856 8:7–14 855 8:9 855 9:1–9 856 9:1–14:1 856 9:3 855 9:10–17 856 10:1–8 856 10:5–8 855 10:6 855 10:9–11:11 857 10:14 855 11:11 855 12 851, 852, 856, 857, 870, 997, 998–1007, 1014 12:1 998, 1002 12:1–2 857, 858, 998, 1001, 1002

1185

12:1–15 856, 857 12:2 855, 1001 12:3 857, 1001 12:3–5 999 12:3–7 857, 998, 1001, 1002–1007, 1011 12:4 437, 474, 507, 999, 1002, 1003, 1004 12:4–5 999, 1003, 1004 12:4–7 1004 12:4–9 857 12:5 474, 507, 858, 998, 999, 1002, 1003, 1004–1005 12:6 474, 998 12:6–7 858 12:7 998, 999–1000, 1001, 1004 12:8 888, 998, 1001, 1002 12:8–9 859 12:8–11 859 12:8–12 1001, 1002 12:8–14 1000 12:9 474 12:10 216, 859 12:10–12 859 12:11 859 12:12 859–860 12:12–13 860 12:13 437, 474, 1005, 1006 12:13–14 1000–1001, 1002, 1004, 1005, 1006, 1007, 1011 12:14 1006–1007 12:15 1002 13:1–14:1 857, 861 13:4 216 14:2–9 855, 857, 861 14:4 855, 861 14:10 857 Joel 4:8 262 Amos 1:1–2:16 960 1:3 967 1:4 926 1:6 967 1:7 926 1:8 967 1:9 967 1:10 926

1186

Ancient Sources Index

Amos (continued) 1:11 967 1:12 926 1:13 967 2:1 967 2:4 894, 967 2:5 892 2:6 967 2:6–7 887 2:7 941 2:8 972 3–6 967 3:1 216 3:1–2 960 3:1–4:13 960 3:1–6:14 960 3:3–6 960 3:3–8 960 3:7–8 960 3:9–12 960 3:13–15 960 4 955, 958, 959 4–5 1046 4:1–3 960 4:4–5 960 4:4–13 960 4:4–5:2 1042 4:5 960 4:6 957–962, 974, 1042 4:6–11 943, 945, 956, 957, 961, 963–964 4:6–12 943–984 4:7 961, 962, 974 4:7–8 969, 1042 4:8 961, 962, 974 4:8–11 958, 959 4:9 961, 962, 974, 1042 4:10 961, 962, 974, 1042 4:11 958, 961, 962, 974, 977, 1042 4:12 943, 960, 963, 972, 974, 981–982, 1042 4:13 943 5:1–2 1043 5:1–17 960 5:1–6:14 960 5:2 887 5:4 967 5:4–5 888 5:5 967 5:6 967

5:11 945, 952 5:14 967 5:18 887 5:20 887 5:21 887 5:23 887 6:8 887, 888 6:13 717, 725 7 992 7–9 857, 887 7:8 1043 7:14 987–988 8:2 848, 849, 888 8:4 887, 888 8:4–7 886–887, 888 8:4–14 887–888 8:5 879, 888 8:6 887, 888 8:7 888 8:9–10 887 8:12 967 8:13–14 887 8:14 724 9:7 216 Jonah 1:3 180 4:2 211 Micah 1–3 1008 2:12 307 4–7 1008 6 997–998, 1007–1011, 1014 6:3 1008, 1009 6:3–5 1007–1011 6:4 216, 1008–1011 6:5 1009, 1010, 1011 6:9–12 886 6:11 888 6:11–16 945 6:14–15 945 7:20 435 Nahum 2:14 1079 Habakkuk 3:5 470

Ancient Sources Index

Haggai 1:3–11 945, 949, 950 1:5–11 945 2:10–14 528 2:15–17 949, 950 2:16–17 945 Zechariah 1–8 362, 920, 985, 995–996 1:3 996 1:3–4 966 1:7 375 7:7 996 7:12 528, 996 9–10 989, 990, 993, 994 9–14 831, 985–996 9:1 986, 992, 993 9:1–8 989 9:1–17 989, 993 9:9–10 989 9:11 985, 988–990, 993–994, 995 9:12 989 9:13 366, 989, 994 9:15 626 10:1–2 994 10:1–3 993 10:4–12 989, 993 10:6 994 10:7 994 11 989 11:1–3 993 11:4–16 993, 995 11:5 988 11:17 993 12–14 986, 989, 994–995 12:1 986, 993 12:1–13:6 993 12:2–13:6 986, 993 13:1 986 13:1–6 986, 994 13:2 986, 994 13:3 986, 994 13:4 987, 988 13:4–5 985, 986–988, 989, 994–995 13:4–6 987 13:5 987–988 13:7–9 986, 988, 993 14:1–21 986, 993 14:16–19 554

Malachi 1:6–14 1011 1:6–2:9 1011 1:14 944 2 998, 1011–1013 2:1–9 1011 2:4–6 1011–1013 3:1 71 3:22 846 Psalms 1:2 803 8 1063 8:5 1063 12:2 432 18 432 18:46 309 29:1 139 34:22 926 36:7 174 39 1063 46:3–4 473 55:5 58 59:6 319 60:36 935 68:2 576 68:26 174 72–73 1063 72:17 900 74:11 833 77:9 432 78 547 78:48 470 78:50 470 78:51 470 89:31 894 97:7 139 98:8 473 103:9 929 104 548 104:10 473 105 365, 547 105:32 902 106 547 106:38 931, 932 107 1063 114:3–6 473 119:34 895 119:44 895

1187

1188 Psalms (continued) 119:53 894 119:55 895 119:136 895 136:1–6 547 139 1063 144:13 333 145:13 434 151 362 Job 1:5 670, 945 1:11 945 2:5 945 2:9 670, 945 2:12 423 3:1–9 945 3:2 418 7:17–18 1063 9:34 58 15:14 1063 25:6 1063 31:5–40 952 42:13–14 415 Proverbs 4:2 895 6:7 636 11:1 887, 888 11:9 932 16:1 279 17:10 965 19:19 667 19:29 1079 20:2 58 20:21 331 20:23 887, 888 28:4 894–895 29:18 895 Ruth 2 211 2:10 211 3 211 Song of Songs 4:8–6:10 136 4:13 333

Ancient Sources Index

Qoheleth 1:10 434 12:5 434 Lamentations 2:9 833 3:53 989 5:7 434 Esther 2:5–6 240 2:8 338 8:3 339 Daniel 2:4 434 9 547, 976 9:5 547 10:20 366 11:2 366 12:6 432 Ezra 1–6 525 3:2 553, 558 3:4 554 3:4–5 554 3:8 558 4:7 353 5–6 529 6:12 526 6:18 526, 553, 558 7 726 7–8 525 7–10 529, 544 7:6 553 7:7 530 7:10 553 7:26 526 9 555, 976 9–10 230, 525–542 9:1 526, 533, 537, 541 9:1–2 533, 537, 558–559 9:1–5 537, 538, 539, 540, 542 9:1–15 531 9:2 526, 533–537, 539, 541, 925 9:3–15 559 9:4 526

Ancient Sources Index

Ezra (continued) 9:6 542 9:6–15 538 9:11 541 9:11–12 537, 538, 555 9:12 553, 555 9:13 542 9:14 537 10 557 10:2–3 559 10:3 526, 553, 558 10:19 532, 542 10:44 531 Nehemiah 1:1–7:5 544 2:1–10 890 5:1–12 888 5:5 888 8 349, 526, 544, 726 8–9 525 8:1 526, 552, 553 8:2 553 8:3 553 8:7 553 8:8 526, 553 8:9 553 8:13 553 8:13–18 554 8:14 526, 558 8:14–15 552 8:15 554 8:18 526, 554, 558 9 98, 100, 545–552, 823, 824, 976, 1114–1115 9:3 553 9:5 546 9:6 546, 547–548, 823, 825 9:7 229, 366, 550 9:7–8 546, 548–550, 823, 825 9:8 549–550 9:9 549, 823, 825 9:9–11 546, 551 9:11 551, 823, 825 9:12 546, 552 9:12–23 546 9:12–37 552 9:13 546, 1098

9:13–20 823, 825 9:14 546, 553 9:15 546, 552 9:16 886 9:16–21 546 9:17 552, 886 9:19 335, 552, 823, 825 9:20 552 9:21 552 9:22 546, 823, 825 9:23 546 9:24 823, 825 9:24–25 546 9:25 823, 825 9:26 553 9:26–28 98–99 9:26–31 546 9:28 332 9:29 553 9:32–37 546, 1115 9:34 432, 553 10 557, 890 10:1–28 557 10:29 553 10:29–30 557 10:30 553, 557 10:31 553, 555, 556, 558 10:31–32 557 10:32 556, 883, 886 10:33 558 10:33–40 557 10:34 558 10:35 148, 149, 181, 526, 553, 558 10:36–38 558 10:37 555, 556 10:38–40 558 11:1–2 544 12 183 12:11 183 12:22 183 12:25–36 889 12:27 890 12:27–43 544 12:31 890 12:38 890 12:40 890 13 557, 890 13:1 553, 558

1189

1190

Ancient Sources Index

Nehemiah (continued) 13:1–2 559 13:1–3 553 13:4–31 544 13:15 880, 881 13:15–16 886 13:15–22 890 13:17 883 13:19 880, 881 13:19–20 883 13:23–30 230 1 Chronicles 1–9 98, 99, 529 2:21–22 99 5:6 240 5:26 240 6:1–15 868 7:1 99 7:18 99 9:17–27 879 12:14 375 13:1 637 16:40 916 17:10 98 21:22 939 22:12 895 22:18 625 23:31 555 24:12 375 25:18 376 27:14 376 2 Chronicles 1:3–6 95

5:11 97 5:11–13 95 5:13 97 6:25 309 7:8–9 917 8 97 12:1 894 14:3 432 15:3–5 98–99 15:8 319 16:12 757 18:4 333 19:4–11 915–916, 916 19:8 916 19:10 916 19:11 916 24:10 1079 24:24 1079 25:4 916 30:16 916 31:1 753 32:1 240 32:9 240 32:17 330 32:21–23 240 35 97 35:1–2 95–96 35:12 1098 35:13 529, 917, 1097–1098 35:17–19 95–96 36:5 374 36:9–10 240 36:11 374 36:16 353 36:20–21 976, 982

Ancient Near Eastern and Epigraphic Texts and Papyri Abibaʿl of Byblos Inscription 22 Ahiqar (CAP 216) 418 Aḥiram Sarcophagus Inscription 18, 20 Amman Citadel Stela 26–27 ʾAqhat 408, 421 Arad Inscriptions 33, 34, 334 Assyrian Inscriptions 366 Atraḫasis 1107 Azarbaʿl Inscription 20

Baʿal Cycle 421 Balua Stela 712–713 Bavian Inscription (Sennacherib) 56, 57, 59, 61, 62, 81 Bentresh Stela 227 Berlin P 11 367 881 Black Obelisk 42–43, 851, 852, 853–854 Broken Statue of Shalmaneser III 43 CAP 216 (Ahiqar) 418

Ancient Sources Index

Chester-Beatty IV Papyrus 757 CTU 1.4 419, 427–428 1.10 419 1.14 415 1.17 418 1.23 418 Deir Alla Inscription 720, 864 Elephantine Letters 330 Elephantine Ostraca CG 152 881 CG 205 881 Eleusis Stela 882 Elibaʿl Inscription 22 el-Kerak Inscription 25–26, 33, 37, 721, 724 Enuma Elish 1107 IV 130 459 V 88 419 VI 1107 VII 1107 Exaltation of Inanna 462 Gezer Calendar 31 Gilgamesh Epic 207, 227–228, 367, 1107 Pennsylvania Tablet IV 137 416–417 Hammurabi. See Laws of Hammurabi Ishbaʿl Jar Inscription 28, 29 KAI 191 (Mesha Stela) 24–25, 34, 37, 38–39, 61, 696, 713–714, 723–724, 725, 864 202 (Zakkur Stela) 43–44, 316 Kefar Veradim Inscribed Bowl 30–31 Kirta 408, 410, 411 KTU 1.100 756 1.107 756 Kuntillet ʿAjrud 34, 413 Kurbail Statue Inscription 42 Kurukh Monolith 39–41 Lachish Letters 330, 334, 413 Laws of Hammurabi 1107 Prologue 53, 54, 55 47:59–73 52 47:59–49:17 50 47:59–51:91 51 47:75–78 52 47:76–80 63

1191

47:93–48:2 52 48:3–58 52 48:34–58 52 48:59–74 56, 63 48:59–94 50 48:59–49:16 53 48:59–49:44 51 48:59–51:91 76–77 49:2–5 56, 63 49:2–17 50 49:16–51:91 53 49:18–22 56, 63 49:18–52 50 49:18–51:91 50 49:45–51:91 51 ¶ 117 48, 49 ¶ 148–149 49 ¶ 196–201 657 ¶ 206–208 657 ¶ 209–210 657 ¶ 251 657 ¶ 282 49 Epilogue 945 Leiden I 350 Papyrus 757 Letter to Aššur (Sargon II) 56, 57, 58, 59, 80–81 Mari Letters 430, 690 Masada Inscription 152, 449 Mesha Stela (KAI 191) 24–25, 34, 37, 38–39, 61, 696, 713–714, 723–724, 725, 864 Middle Assyrian Laws 49, 665 Nash Papyrus 121, 130, 154 Qeiyafa Ostracon 28, 29 Rassam Decagon/Cylinder 56, 57, 58, 59, 60, 61, 62, 83–85 RIMA 2 A.0.99.2 56, 57, 62, 63, 78 2 A.0.101.17 56, 57, 59, 62, 78–79 3 A.0.102.2 56, 57, 58, 59, 61, 79–80 RINAP 3.120 366 4.60 366 4.1 56, 57, 58, 59, 60, 62, 81–82 4.98 56, 57, 58, 59, 62, 83 SAA XI.14 690 Sargon II (Letter to Aššur) 56, 57, 58, 59, 80–81

1192

Ancient Sources Index

Shalmaneser III, Broken Statue of 43 Samaria Ostraca 35–36 Sefire 1A 418, 945 2A 945 Sennacherib (Bavian Inscription) 56, 57, 59, 61, 62, 81 SH 809 690 Shipiṭbaʿl Inscription 22 Siloam Inscription 413 Song of Release 429 Story of Sinuhe 226–227

Sumerian King List 204 Tale of Woe 227 Tel Dan Stela 22–24, 44, 864 Tell Ataruz Inscription 36–37 Tell Fekherye Inscription 945 Tel Reḥov Inscriptions 31–33, 34 Tel Zayit Abecedary 31 Vassal Treaties of Esarhaddon 67, 369, 675–676, 945 Wenamun 226 Yeḥimilk of Byblos Inscription 21–22 Zakkur Stela (KAI 202) 43–44, 316

Deuterocanonical Works Wisdom of Solomon 10:1–11:1 1115 16:5–12 756 16:7 756

14:20 893 15:1 892, 893 38:1–9 757 44:1–49:50 1115

Sirach 4:17 434

1 Esdras 9:36 531

Second Temple Literature Aramaic Levi 16–18 536 22–25 536

16:25–26 536 25:3 536 28:6 495 48:2–3 508

Jubilees 12 1117

Texts from the Judean Desert CD (Damascus Document) 116, 118, 340–341, 1116 1QDanb 153 1QDeut a 154 1QDeut b 145, 154, 631 1QExod 154 1QGen 154 1QH a (Hodayot) 118 1QIsa a 147, 152, 156, 310 1QM (War Scroll) 118, 379–380 1QpaleoLev–Num a 147, 154 1QpHab 432 1QPsb 147

1QS (Community Rule) 116, 118, 147, 432 2QExod a 145, 154 2QExod b 154 2QJer 156 2QNum a 154 2QpaleoLev 153 4QBarkhi Nafshia 153 4QCant a 136 4QDan a 155 4QDan b 155, 156 4QDeut a 153 4QDeut b 144, 154

Ancient Sources Index

4QDeut c 144, 154 4QDeut d 145, 154 4QDeut e 145, 171 4QDeutf 145, 154 4QDeutg 145, 154, 171 4QDeut h 145, 148, 154, 631 4QDeuti 145, 154 4QDeutj 137, 138, 139, 145, 154 4QDeut k1 145, 147, 154, 156 4QDeut k2 145, 147, 154, 156 4QDeutl 154 4QDeut m 154 4QDeut n 136, 141–143, 145, 154, 488 4QDeut o 145, 154 4QDeut p 154 4QDeut q 137, 139, 145, 154, 180, 192, 193–194 4QEschatological Hymn 147 4QExod b 145, 154 4QExod c 145, 154 4QExod d 154 4QExod e 154 4QExod–Levf 144, 145, 153, 189–190, 191, 193, 194, 195 4QFlorilegium 488, 489 4QGenb 145, 154, 171 4QGenc 145, 154 4QGend 154 4QGene 145, 154 4QGen f 145, 147, 154 4QGen g 145, 147, 154 4QGenj 145, 154 4QGen k 154 4QGen–Exod a 145, 154, 180–181 4QIndividual Thanksgiving A 147 4QIsa c 147, 156 4QJera 145, 160, 189 4QJer b 145 4QLam 156 4QLevb 145, 154 4QLevc 145, 154 4QLevd 145, 154, 192 4QLeve 145, 154 4QLevg 154 4QLev–Num a 144, 153, 156 4QLXXDeut 131 4QLXXLeva–b 131 4QLXXNum 131 4QLXXNum105 162

1193

4QMMTb 153 4QNarrative G 147 4QNumb 145, 154, 156, 188, 190 4QpaleoDeut r 145, 154 4QpaleoExod 111 4QpaleoExod m 145, 154, 190, 514 4QpaleoGen m 153 4QpaleoGen–Exod l 145, 154 4QPersonal Prayer 147 4QpIsa b 152 4QpNah 152 4QPs a 156 4QPseudo-Ezekiel 118 4QQoh a 156 4QRP 103, 487 4QSam a 135, 505 4QSam b 145, 152, 189 4QSamc 147, 156 4QSd 153 4QSj 153 4QTb 182 4QTestimonia 147–148, 149, 488, 489 4QtgLev 131 4QTQahat ar 147 4QXIIc 156 4QXIIe 160 4Q158 183–185, 480, 483–489, 491–500, 502–524 1–2 492–500, 507, 508, 509, 511, 521, 522 3 523 4 507, 509, 511, 512, 522 5 522 5–12 511 6 514–515, 516, 517, 521 6–9 514–517 7 516 7–9 515, 516, 517, 518, 521, 522 10–12 522 13 511 14 523 4Q176 1063 4Q252 (4QCommGen A) 152, 380–382 4Q255–264 432 4Q266–273 432 4Q364 (4QRPb) 154, 156, 187, 485, 486, 488, 503, 504 4Q364–367 183–184, 185, 484–485, 502–524

1194

Ancient Sources Index

4Q365 (4QRPc) 136, 149, 154, 156, 181, 182, 183, 185, 186–188, 190, 194–195, 486, 488, 503, 504 4Q365a 185–186, 187, 188 4Q366 (4QRPd) 154, 485, 486, 488 4Q367 (4QRPe) 145, 154, 486, 488 4Q390 117 4Q522 153 5QDeut 137, 144, 153 6QpaleoGen 147, 154 7QpapLXXExod 131 8QGen 154 11QLevb 154 11QM 147 11QpaleoLeva 154 11QPs a 156 11QTa 101, 103, 112, 116, 136, 149, 181–182, 186, 187, 188, 194, 340 11QTb 188 11QtgJob 170

ArugLev 154 ḤevLev 145, 154, 171 ḤevDeut 154 ḤevNum b 145, 154, 171 8ḤevXII gr 137, 162 MasDeut 132, 145, 154, 171 MasEzek 132 MasLeva 132, 145, 154, 171 MasLevb 132, 145, 154, 171 MurDeut 154 MurGen–Exod 145, 154, 171 MurLev 145, 154, 171 MurNum a 145, 154, 171 MurXII 145 Mur 1:2 334 42:2 152 SdeirGen 145, 154, 171 SeDeut 154 SeNumb 145, 171

New Testament Acts 7:2–4 229

Hebrews 11 1116

Rabbinic Works Mishnah m. Berakot 176 m. Kelim 165, 166 m. Megillah 177, 1093 m. Moʿed Qaṭan 167, 168 m. Šabbat 177 m. Sanhedrin 169 m. Soṭah 103, 165 m. Yoma 165 Tosefta t. Baba Meṣiʿa 166, 176 t. Kelim 166 t. Šabbat 170 t. Sanhedrin 168 Jerusalem Talmud y. Megillah 173, 176 y. Šabbat 170, 177–178 y. Sanhedrin 168–169 y. Šeqalim 167, 168, 169 y. Taʿanit 169–170, 171, 648

Babylonian Talmud b. Baba Batra 169 b. Berakot 175, 176 b. ʿErubin 175, 1094 b. Giṭṭin 1094 b. Ketubbot 167, 169, 176 b. Makkot 1049 b. Megillah 152, 167 b. Moʿed Qaṭan 167 b. Nedarim 174, 175 b. Pesaḥim 177 b. Šabbat 170 b. Soṭah 165 Midrashim and Minor Tractates ʾAbot de Rabbi Nathan 175 ʾAggadat Berešit 1092 Genesis Rabbah 176, 203, 1092 Leviticus Rabbah 175 Midrash Bereshit Rabbati 127 Midrash Tanḥuma Lekh lekha 233

Ancient Sources Index

Midrash Tanḥuma (continued) Vayyera 1092 Miqra‍ʾot Gedolot 1094

Sifre Numbers 575 Song of Songs Rabbah 495

Early Christian Writings and Greco-Roman Literature Josephus Against Apion 163, 164 Jewish Antiquities 113, 164, 183, 241, 648, 703 Jewish War 164 The Life 164

Homer Iliad 210–211 Odyssey 206, 225, 228

Eusebius Onomasticon 685, 699, 703

Origen Hexapla 175

Hecataeus of Abdera Aegyptiaca 357

Ovid Fasti 408

Hecataeus of Miletus Periodos Ges 736–737

Pseudo-Xenophon Constitution of the Athenians 882

Herodotus Histories 225, 754

Tatian Diatessaron 103, 212

Hesiod Catalogue of Women 204 Theogony 204

Virgil Aeneid 206

Justinian Digest 569

Medieval Writers Benjamin of Tudela 225 Hazquni, Abraham 1124 Ibn Ezra, Abraham 104, 319, 358, 1124 Levi ben Gershom 1126

Nahmanides (Ramban) 233 Rashi 104–105, 106, 262, 817, 935, 1095–1096, 1102

1195

Author Index Aaron, D. H. 431 Achenbach, R. 601–602, 623, 720, 750, 885 Ackroyd, P. 834 Aharoni, Y. 741 Aitken, K. 421, 422, 946 Albeck, C. 165 Albertz, R. 71, 616, 617, 720 Albrecht, J. E. 255, 256 Alexander, P. S. 731, 741 Allegro, J. 483, 487 Alt, A. 739, 815 Alter, R. 438, 440 Amir, Y. 1100, 1103 Amit, Y. 235–236 Anbar, M. 824 Andersen, F. I. 960, 961, 962, 974 Andersen, H. C. 208 Anderson, B. W. 530 Arico, A. F. 26 Artus, O. 620 Assmann, J. 727 Aster, S. Z. 59, 65, 66, 75 Astruc, J. 246–247, 249, 460, 580, 583, 590 Athas, G. 22 Auld, A. G. 94, 366 Aurelius, E. 749 Bacon, B. W. 693, 695, 707 Baden, J. S. 74, 75, 101, 201–202, 213, 411, 589, 603–604, 614, 616, 623, 747, 819, 1008, 1091, 1096–1097, 1107 Baentsch, B. 620, 621, 1056–1057, 1059, 1064, 1065 Barbiero, G. 276 Bar-On, S. See Gesundheit, S. Barstad, H. M. 972 Barton, J. 460, 844–845 Baumgartner, W. 409 Bautch, R. J. 480–481 Beauchamp, P. 1122

Becker, U. 823 Beckett, S. 450 Becking, B. 529 Bedford, P. R. 951 Beit-Arié, M. 124 Bendavid, A. 382 Benet, W. R. 225 Ben-Gad HaCohen, D. 696 Ben-Porat, Z. 990 Benzinger, I. 780 Berlejung, A. 61, 65 Berlin, A. 462 Berman, J. 369 Berner, C. 456, 471, 517, 519, 565, 603, 611, 612, 747–748, 750 Bernhardt, K. H. 722 Bernstein, M. J. 487 Beyerle, S. 746 Bickert, R. 720 Biran, A. 22 Blank, S. H. 945, 947 Blau, L. 164, 170 Bleek, F. 590, 795 Blenkinsopp, J. 148, 348–349, 536 Block, D. I. 1031, 1079, 1080 Blum, E. 71, 73, 273, 276, 319, 402, 435, 436, 587, 589, 597–599, 608, 609, 781, 799, 818, 956, 999, 1001 Boccaccio, G. 207, 208 Boda, M. 547, 552, 834, 955, 966, 968, 973, 1050 Braiterman, Z. 1095 Braulik, G. 638, 807 Brettler, M. Z. 300 Brichto, H. C. 947 Brockelmann, C. 386 Brooke, A. E. 184 Brooks, P. 1121, 1123 Broshi, M. 182 Brown-Gutoff, S. E. 901 Brueggemann, W. 1101, 1122 Buber, M. 3, 1100

1198

Author Index

Budde, K. 780, 815 Buis, P. 944, 978, 981 Bultmann, R. 1122 Burney, C. F. 704, 789 Burrows, M. 1063, 1064, 1070 Büttner, M. 734 Calvin, J. 1056 Carmichael, C. 51, 662 Carpenter, J. E. 246 Carr, D. M. 87, 93, 94, 104, 113, 352, 400, 407, 428, 431, 492, 497, 501, 525, 528, 529, 603, 613, 838, 915, 917, 938, 939, 940, 1017, 1091, 1098, 1106, 1136–1137 Cascardi, A. J. 1145 Cassuto, U. 233, 243, 247, 563 Charles, R. H. 236 Chavel, S. 671, 677, 681, 790 Childs, B. 1099, 1100–1101, 1102, 1105 Cholewiński, A. 679 Clark, D. R. 773 Clements, R. E. 814 Clines, D. J.A. 148, 411, 596–597, 1121 Coats, G. W. 235 Cogan, M. 39, 81, 853 Coggins, R. J. 972 Cohen, C. 353 Cohen, Y. 429 Cohn, R. L. 212–213 Collins, J. J. 355–356 Conrad, J. 449–450, 451 Conroy, C. 422 Coote, R. B. 960 Cornelius, I. 732 Cornill, C. H. 779, 1059, 1062 Corwin, R. 328 Crane, A. S. 1020, 1036, 1037 Crawford, S. W. 183–184, 185, 187, 480 Crenshaw, J. L. 18, 956, 961, 978 Cross, F. M. 35, 38, 102, 152, 153, 183, 284, 408, 421, 483, 484 Cryer, F. H. 296, 301 Cuddon, J. A. 225 Dällenbach, L. 281–282 Danby, H. 167 Daube, D. 569 Davies, G. I. 17, 698

Davies, P. R. 296, 298, 350–351 Davis Parker, H. D. 19, 26, 32, 33 Deferrari, R. J. 452 deGroot van Houten, C. 1084 Delitzsch, F. 300, 307, 1058 Delnero, P. 87 De Rossi, J. B. 179 Dickey, E. 298–299 Dijkstra, M. 864 Di Lella, A. A. 893 Dilke, O. A.W. 736–737 Dillmann, A. 1066 Dombrowski, B. W.W. 385 Dombrowski, F. A. 385 Donner, H. 597, 742, 743 Dozeman, T. B. 460, 688, 761, 808, 866 Driver, S. R. 314, 328, 379, 626, 1056, 1064, 1066, 1068, 1069, 1070 Duhm, B. 1001 du Mesnil, A. 452 Duvall, J. S. 893 Edenburg, C. 13, 362 Ehrensvärd, M. 302, 428, 432 Eichhorn, J. G. 249, 460, 589, 590 Eichler, B. L. 658 Eichrodt, W. 1116 Eissfeldt, O. 359, 594–595 Elliger, K. 314, 956, 1067 Erisman, A. R. 691 697 744 Eshel, E. 134, 141 Even-Shoshan, A. 892 Ewald, G. H.A. 405, 406, 407, 409, 567, 833, 1121 Fabry, H.-J. 975 Fantalkin, A. 530 Finkelstein, I. 15–16, 17–18, 351 Finnegan, R. 421, 438 Fischer, G. 934, 935, 939, 962, 965, 982, 1128 Fischer, H. 739 Fishbane, M. 107, 108, 109, 110, 111, 115, 270, 272, 276, 281, 492, 665, 837, 918, 919, 920, 924, 925, 926, 927, 930, 939, 941, 1098 Flinders Petrie, W. M. 699 Flury-Schölch, A. 900 Fohrer, G. 595, 1017, 1062, 1066, 1067

Author Index

Fokkelman, J. P. 209, 210 Foley, J. 422–423 Forster, E. M. 202–203, 207–208, 211, 221, 222 Fox, M. V. 895 Frankel, Z. 191 Freedman, D. N. 38, 151, 189, 960, 961, 962, 974 Frevel, C. 1133 Friedman, R. E. 696, 697 Fritz, V. 819 Frolov, S. 1132 Fronzarolio, P. 459 Fuller, R. E. 134 Galling, K. 435 Gass, E. 719 Gass, H. M. 299 Geiger, G. 159, 592 Gentry, P. 125 George, A. R. 207 George, L. 842, 843, 844 Gerstenberger, E. 949, 972, 973, 974, 978 Gertner, M. 1004 Gertz, J. C. 402, 463, 465, 475, 599, 600, 612, 818 Gese, H. 303 Gesenius, H. F.W. 190, 295, 328, 329, 333 Gesundheit, S. 273, 274, 874, 1098 Gevirtz, S. 386, 947 Giesebrecht, F. 379, 383 Gillman, A. 1094 Ginsberg, H. L. 287, 1067 Ginsburg, C. D. 123, 128 Ginzberg, L. 208 Giuntoli, F. 574 Goldstein, R. 308 Gordis, R. 164, 168 Gordon, C. 301 Goren, Y. 227 Gottlieb, L. 295 Graf, K. H. 249, 359, 391, 583, 592, 619, 842, 1015, 1059, 1075, 1087 Gräfenhan, A. 591 Graffy, A. 1048, 1049 Graham, W. A. 107 Graupner, A. 286 Graupner, M. 963, 975 Gray, G. B. 623

1199

Grayson, A. K. 39, 42 Green, W. H. 243, 283, 284 Greenberg, M. 285, 470, 612, 1022, 1048, 1059, 1077, 1080–1081 Greenfield, J. C. 312 Greenstein, E. L. 472, 1094 Gressmann, H. 563–564, 609 Grimm, J. 208 Grimm, W. 208 Grund, A. 874, 876 Grünwaldt, K. 1059, 1069–1070 Gunkel, H. 208–209, 234, 247, 318, 406, 443, 460, 564, 582, 593–594, 609 Hadjiev, T. S. 963, 976, 977, 978 Haldeman, S. S. 453 Hale, K. 450 Halliday, M. A.K. 254 Halpern, B. 37 Hanhart, R. 137 Haran, M. 378, 694, 706, 783, 790, 793, 1062, 1091 Harford-Battersby, G. 246 Harris, R. 104, 1092 Hartley, J. E. 971 Hasan, R. 254 Havrelock, R. 744 Hayes, C. 534 Hays, J. D. 893 Hehn, J. 467 Hendel, R. 408, 474 Hengstenberg, E. W. 845 Herr, L. G. 770, 773 Hertzberg, H.-W. 595 Heschel, A. J. 1099, 1103 Higham, T. 776 Hillers, D. R. 948 Hirth, F. 703 Hobson, R. 87 Hoffman, Y. 372 Hoftijzer, J. 322 Holladay, W. L. 537, 930, 931, 933, 936 Hölscher, G. 815 Holzinger, H. 458 Hornkohl, A. 295 Horowitz, W. 736 Houston, W. J. 1122, 1138 Huffines, M. L. 453 Hughes, J. 597

1200

Author Index

Hupfeld, H. 249, 580, 583 Hurowitz, V. A. 56, 865 Hurvitz, A. 295, 296–300, 307, 308, 309, 310, 311, 313, 328–329, 331, 339, 345, 348, 350, 352, 355, 371–372, 373, 378, 379, 382, 385, 389, 1032 Hutton, R. R. 680 Hyatt, J. P. 930 Ilgen, K. D. 583 Isaac, E. 734, 738 Iser, W. 765 Jackson, B. S. 51, 662 Jacob, B. 565, 590, 612 Jamieson-Drake, D. 17 Japhet, S. 98, 182, 481, 529, 532, 541 Jefferson, A. 1123 Jenks, A. W. 696 Jeremias, J. 956, 958, 961 Ji, C.-H. 36 Johnson, W. M. 542 Joosten, J. 89, 296, 313, 318, 323–324, 355, 427, 471, 708 Kalisch, M. M. 1057 Kallai, Z. 232, 743–744 Kara, J. 468 Kasche, G. H. 737–738 Kaufman, I. T. 35, 36 Kaufman, S. A. 347, 501 Kaufmann, Y. 348, 360, 837, 1077, 1084 Kautzsch, E. 311 Kearney, P. J. 1131 Keel, O. 758–759 Kegel, M. 845 Kellogg, R. 202 Kennicott, B. 133, 179 Kessler, J. 831, 834, 1046 Kilian, R. 1066 Kim, D.-H. 295, 306, 348 Kim, K.-R. 191 Kirkpatrick, P. G. 422 Kittel, R. 780 Kiuchi, N. 970 Klein, A. 1071 Klein, R. W. 99 Kletter, R. 770 Klostermann, A. 1016, 1059, 1068, 1069

Knauf, E. A. 296, 362, 367, 716, 726, 802, 809, 819, 820, 821, 822 Knohl, I. 307, 602–603, 933 Knoppers, G. N. 958 Köckert, M. 885 Kong, L. 734 Kopilovitz, A. 1075, 1076, 1083 Körting, C. 885, 886 Kratz, R. G. 114, 116, 117, 118, 480, 492, 538, 591, 602, 1136 Krause, J. J. 802, 804 Kropat, A. 328 Krüger, T. 874 Kuenen, A. 249, 359, 583, 585, 591, 593, 595, 619, 814, 833, 1016, 1057, 1058, 1060, 1064, 1065, 1087 Kugel, J. 1092, 1143 Kutscher, E. Y. 164, 296, 310, 328 LaBianca, Ø. S. 773 Labov, W. 348 Lagrange, M.-J. 571 Lai, H.-C. 534, 536 Lämmert, E. 409 Lange, A. 131, 134, 136 Lemaire, A. 18 Lemche, N. 296, 300 Leuchter, M. 537, 680 Levenson, J. D. 234, 1098, 1103 Levin, C. 342, 396, 599, 879, 1060, 1061, 1117 Levinas, E. 1094 Levine, B. A. 361, 372, 373, 623, 969, 978, 1063, 1067 Levinson, B. M. 108, 119, 244, 269, 271, 273, 369, 496, 525, 603, 665, 925, 954, 1139 Levita, E. 127–128 Levitt Kohn, R. 940, 941, 1016, 1050, 1070 Levy, T. E. 776 Lieberman, S. 164 Lilly, I. 1020–1021 Lindblom, J. 834 Lipiński, E. 935 Lipschits, O. 241, 540 Liverani, M. 37 Loewenstamm, S. E. 384, 459, 624 Lohfink, N. 632, 636, 643, 645

Author Index

Long, B. O. 929, 933 Lord, A. B. 421, 422–423 Luckenbill, D. D. 62, 81 Lundbom, J. R. 936 Lust, J. 1020, 1036, 1037 Luzzatto, S. D. 224 Lyons, M. A. 1016, 1017, 1018, 1019, 1028, 1047, 1048, 1081, 1083 Machholz, C. 102 Mackie, T. P. 1018, 1021, 1022, 1024 Maier, C. 877, 890, 891, 893, 894, 898 Malamat, A. 240 Mankowski, P. V. 383 Mann, T. 243, 1122 Marböck, J. 893 Margalit, B. 288 Markl, D. 1128 Marquis, L. M. 621, 622, 623, 624, 626, 628 Mason, R. A. 834, 835 Mathews, K. 189 Mayer, W. 80 Mays, J. L. 974 Mazar, A. 32 McCarthy, D. J. 862 McKane, W. 930, 935, 936 McKay, J. 853 McKenzie, S. L. 529 Meyers, C. L. 988, 990 Meyers, E. M. 988, 990 Michaelis, J. D. 738 Milgrom, J. 277, 345, 348, 355, 538, 670, 924, 925, 927, 962, 967, 968, 970, 971, 972, 973, 976, 978, 1028, 1033, 1061 Millard, A. R. 18 Miller, C. L. See Miller-Naudé, C. Miller, J. M. 765, 768, 771 Miller-Naudé, C. 295 Milstein, S. 87 Milton, J. 450, 451 Mizrahi, N. 296, 299 Moers, G. 226–227 Moore, G. F. 704 Morgan, R. 844–845 Morrow, W. 48 Mowinckel, S. 1140 Müller, R. 87, 100, 1067, 1070, 1071

1201

Naudé, J. A. 296 Naveh, J. 22, 312 Na‍ʾaman, N. 40, 725–726, 751 Neef, H.-D. 71, 1004 Nelson, R. D. 822, 918, 948 Neusner, J. 166, 1140, 1141 Newman, D. 729 Nicolet, C. 740 Niditch, S. 428 Nihan, C. 277, 532, 602–603, 652, 654, 655, 658, 659, 665, 667, 670, 671, 672–673, 677, 679, 681, 682, 810, 970, 971, 972, 973, 975, 979, 980, 1028, 1061, 1063, 1071, 1072, 1073, 1075, 1076 Nöldeke, T. 1133 North, F. 933 Noth, M. 102, 314, 365, 579, 594, 608, 611, 614, 626, 629, 686, 689, 695, 698, 705, 706, 721, 739, 767, 780, 781, 792, 798, 815, 826, 927, 1133, 1137 Nurmela, R. 834 O’Banion, J. D. 1143 O’Brien, E. J. 255–256 Ólafsson, S. 296 Olson, D. 1131–1132 Osumi, Y. 66 Oswald, W. 519 Otto, E. 460, 601, 602, 603, 640, 659, 717, 810, 819, 876, 906, 1003, 1004, 1063, 1065, 1133 Pakkala, J. 87, 91, 100, 101 Pardee, D. 422 Pardes, I. 226 Park, C. C. 732 Parker, S. B. 408 Parry, M. 421 Paton, L. B. 1056, 1059, 1060, 1063, 1064, 1068 Paul, S. M. 312, 350, 434, 960, 962, 974, 978, 981 Pedersen, J. 300 Peels, H. G.L. 971 Perlitt, L. 1133 Perrault, C. 208 Petersen, D. L. 949, 951 Pfeiffer, H. 999, 1001

1202

Author Index

Piccirillo, M. 742 Polak, F. H. 296, 329, 419 Polzin, R. 313, 314, 338, 345, 349 Popper, J. 359, 591, 592–593 Pritchard, J. B. 226, 864 Pury, A. de 366, 474, 1130 Qimron, E. 329 Rad, G. von 224, 360, 363, 393, 582–583, 780, 826, 834, 861, 862, 1115, 1121 Redford, D. 365 Reinmuth, T. 882 Reisner, G. A. 35, 36 Rendsburg, G. A. 296, 301, 313, 315–316, 347, 386 Rendtorff, R. 283, 315, 589, 596, 608, 780, 781 Rezetko, R. 295, 296, 308, 345, 348, 349, 371, 428, 432 Ringgren, H. 964 Rofé, A. 315, 316, 319, 338, 620, 789, 916, 917, 918, 919 Rogland, M. 276 Rollston, C. A. 17, 19, 34 Römer, T. 408, 436, 437, 538, 602, 747, 748, 810 Rom-Shiloni, D. 238, 929, 960, 980, 1029 Rooker, M. F. 296, 297 Rosenzweig, F. 1099–1100, 1102, 1103–1105, 1108 Roskop, A. E. See Erisman, A. R. Routledge, B. 771, 772, 773 Rudolph, W. 583, 584 Ruprecht, E. 596, 598 Rüterswörden, U. 754 Ryssel, C. V. 379 Sagi, A. 1095 Saltman, A. 128 Sanford, A. J. 258–259 Sass, B. 17–18, 755 Schart, A. 887 Schechter, S. 340–341, 1106 Schmid, K. 218, 363, 402, 475, 814, 832, 901, 1133 Schmid, W. 452 Schmidt, L. 628, 715, 716, 718, 721 Schmitt, H. C. 814

Schmitt, R. 755 Schniedewind, W. 33–34, 297 Scholes, R. 202 Schottroff, W. 946 Schunck, K.-D. 883 Schwartz, B. J. 73, 260, 604, 619, 669, 925, 1047, 1080, 1083, 1091, 1096, 1105, 1107, 1138, 1139 Schweid, E. 1097 Schwienhorst-Schönberger, L. 63, 67, 68 Seebass, H. 395, 602, 626, 627, 628, 715, 746, 766 Seeligmann, I. L. 915–917 Seetzen, U. 739 Segal, M. 184–185, 485–486, 488, 489 Seow, C.-L. 296 Seybold, K. 932 Siegel, J. P. 127 Simon, R. 246, 249, 460 Simpson, C. A. 780 Sivan, D. 354 Ska, J. L. 281, 360, 407, 409, 410, 444, 460, 1122, 1132 Skehan, P. W. 193–194, 483, 893 Skinner, J. 224 Smend, R. 395, 584, 595–596, 780, 845, 848, 1058, 1065 Smith, M. S. 472 Smith, W. C. 1101 Soggin, J. A. 961 Soja, E. 740–741 Sommer, B. D. 119, 323, 496, 921, 940, 1114 Sonnet, J.-P. 1088 Southwood, K. 534 Speiser, E. A. 286 Sperber, D. 763 Sperling, S. D. 346 Spieckermann, H. 885, 886 Spinoza, B. de 249, 358, 543, 795, 814 Stackert, J. 114, 119, 671, 845, 969, 970, 973, 980, 1083 Stade, B. 1008 Staerk, W. 780 Stead, M. R. 834, 990 Steck, O. H. 107 Stegemann, H. 146, 186–187 Sternberg, M. 440 Steymans, H.-U. 368

Author Index

Stipp, H.-J. 91 Stromberg, J. 1021 Strugnell, J. 184, 484, 487, 517 Stump, R. W. 734 Sturt, P. 259 Sumner, W. A. 705 Sun, H. 1072 Swanson, D. D. 182 Sweeney, M. A. 866, 869, 988, 989 Swift, J. 225 Tadmor, H. 39 Tal, O. 530 Talmon, S. 111 Talshir, D. 329, 346, 348 Talstra, E. 955 Teeter, D. A. 137 Tertel, H. J. 91 Thiel, W. 877 Thiselton, A. 947 Thomas, R. 422, 423 Thompson, T. L. 15, 296, 862 Tigay, J. 12, 13, 501 Tigchelaar, E. J.C. 187 Tollington, J. 834 Tooman, W. 1035 Toorn, K. van der 135, 281, 352 Tov, E. 88, 103, 130, 131, 134, 151, 152, 153, 167, 183–184, 185, 187, 188, 484, 485, 486, 487, 503–504, 575, 576, 803, 1025 Tuan, Y. F. 734 Tuell, S. 1062 Turner, V. 472 Ullendorf, E. 362 Ulrich, E. 111, 132, 150, 185, 193–194, 486 VanderKam, J. C. 185 Van Seters, J. 232, 318, 352, 407, 408, 422, 435, 567–568, 583, 589, 626, 701, 717, 767, 768, 780, 796, 824, 862, 1117 Vatke, W. 842, 843, 844, 845, 1087 Veenhof, K. R. 946 Veijola, T. 363, 580, 885, 886 Vervenne, M. 179, 284–285 Vielhauer, R. 1001

1203

Vismann, C. 1143 Vollmer, J. 1001 Volz, P. 584, 933 Vorländer, H. 364 Vroom, J. 654 Vyhmeister, W. K. 776 Watts, J. W. 1088 Wazana, N. 744 Weimar, P. 395, 465 Weinfeld, M. 877, 954, 1129 Weiss, M. 440, 960 Wellhausen, J. 1, 7, 94, 249, 314, 348, 355, 360, 383, 467, 557, 563, 567, 568, 571, 576, 577, 582, 583, 590–592, 593, 596, 604, 614, 698, 761, 780, 814, 815, 820, 821, 824, 833, 841, 842, 843–844, 861–862, 1015, 1016, 1017, 1057, 1058, 1060, 1061, 1062, 1065, 1066, 1072, 1075, 1087, 1105, 1133, 1136 Wendland, E. R. 993 Wenham, G. 224, 230, 238, 439 Westermann, C. 204, 205, 224, 230, 231, 235, 237, 408, 573 Wette, W. M. L. de 358–359, 368, 449, 1087, 1111 Wevers, J. W. 137, 192, 631 White, S. A. See Crawford, S. W. Whitt, W. D. 1003, 1004 Whybray, R. N. 407, 563, 590 Wiegand, H. E. 272 Williamson, H. G.M. 552 Witte, M. 590, 601, 1088–1089 Wöhrle, J. 834, 887 Wolff, H. W. 360, 956, 957, 958, 961, 966, 967, 972, 975, 1006 Wright, J. K. 733 Wright, R. M. 296 Würthwein, E. 722 Wüst, M. 716 Wyatt, N. 288 Yeivin, I. 178 Yoo, P. Y. 1132 Young, I. 295, 296, 300, 301, 332, 345, 348, 349, 371, 387, 427, 428, 432 Younker, R. 773 Yun-Sin, S. 434

1204

Author Index

Zacuto, A. 128 Zahn, M. M. 87, 90, 185, 480, 487, 488, 489, 514, 521, 522 Zakovitch, Y. 117, 229, 440, 939, 1052, 1133 Zenger, E. 814

Zevit, Z. 295 Zimmerli, W. 314, 591, 1017, 1018, 1024, 1032, 1037, 1052, 1058, 1067 Zimmern, H. 383 Zsolnay, I. 57