Two Sides of a Coin: Juxtaposing Views on Interpreting the Book of the Twelve / the Twelve Prophetic Books 9781463218973

A conversation between James D. Nogalski and Ehud Ben Zvi on the question of The Twelve, its implications for the histor

200 84 661KB

English Pages 102 Year 2009

Report DMCA / Copyright


Polecaj historie

Two Sides of a Coin: Juxtaposing Views on Interpreting the Book of the Twelve / the Twelve Prophetic Books

Citation preview

Two Sides of a Coin: Juxtaposing Views on Interpreting the Book of the Twelve / the Twelve Prophetic Books

Analecta Gorgiana

201 Series Editor George Anton Kiraz

Analecta Gorgiana is a collection of long essays and short monographs which are consistently cited by modern scholars but previously difficult to find because of their original appearance in obscure publications. Carefully selected by a team of scholars based on their relevance to modern scholarship, these essays can now be fully utilized by scholars and proudly owned by libraries.

Two Sides of a Coin: Juxtaposing Views on Interpreting the Book of the Twelve / the Twelve Prophetic Books

By Ehud Ben Zvi James D. Nogalski Introduction by Thomas Römer

 2009

Gorgias Press LLC, 180 Centennial Ave., Piscataway, NJ, 08854, USA Copyright © 2009 by Gorgias Press LLC All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. 2009


ISBN 978-1-60724-303-8

ISSN 1935-6854

The papers in this volume are also forthcoming in French in Th. Römer, Jean-Daniel Macchi, Christophe Nihan (ed), La formation des livres prophétiques (Le Monde de la Bible, Geneva, Labor et Fides, 2010).

Printed in the United States of America

TABLE OF CONTENTS Table of Contents.....................................................................................v Introduction: The Book of the Twelve—Fact and Fiction? by Thomas Römer ................................................................................1 One Book and Twelve Books: The Nature of the Redactional Work and the Implications of Cultic Source Material in the Book of the Twelve by James D. Nogalski ...............................11 1. Recent Redactional Investigations on the Book of the Twelve ....................................................................................12 Progress Toward a Consensus ....................................................12 Doubts about the Task.................................................................16 The Nature of the Redactional Work in the Book of the Twelve ....................................................................................22 2. Form and Function of the Redactional Work ......................30 The Role of a Writing’s Sitz im Buch Can Affect Its Form .....30 The Functions of the Levites in Chronicles, Ezra, and Nehemiah...............................................................................40 Is the Twelve Hypothesis Likely from an Ancient Readers’ Perspective? by Ehud Ben Zvi ....................................................47 1. Introduction ...............................................................................47 2. Why Readings are Crucial and How to Approximate the Way in which the Present Prophetic Books were Read by their Primary Rereaderships in Yehud.........................54 3. Main Arguments Pointing at a Historical Reading of Fifteen Prophetic Books in Yehud: Data and Implications...........................................................................64 3.1 One Scroll and Readings........................................................64 3.2 Main Textually Inscribed Markers Signaling the Intended Readers How to Read—Part I— Superscriptions......................................................................72



TWO SIDES OF A COIN 3.3 Main Textually Inscribed Markers Signaling the Intended Readers How to Read—Part II Endings and Individualized Voices...........................................................77 3.4 Additional Markers—Genre .................................................79 3.5 The Central, Connective Weakness of the TH from an ancient Readers’ Perspective, or Why the Twelve Prophetic Books Were Not Read as a the “Book of the Twelve” ...........................................................................80 Excursus..........................................................................................84 4. Back to Methodology: Practical, Concrete Differences Between TH Approaches and the One Advanced Here In Terms of Questions They Raise and How they Shape and Interpret Evidence ...................................85 4.1 Practical Differences Concerning Recourse to Memory Studies ....................................................................................86 4.2 Practical Differences Concerning Openings and Endings ..................................................................................86 4.3 Practical Differences—Sequential vs. Branched Modes of Reading and the matter of General Web-like Constructions of the Repertoire and Knowledge of Ancient Literati .....................................................................90 4.4 Practical Differences—The Question of “Main Themes”.................................................................................94

INTRODUCTION: THE BOOK OF THE TWELVE—FACT AND FICTION? THOMAS RÖMER COLLÈGE DE FRANCE—UNIVERSITY OF LAUSANNE (SWITZERLAND) At the end of the 1950s the standard French Introduction to the Hebrew Bible introduced the reader to biblical prophecy as follows: “Who are the classical prophets? … Men of a message (dabar), men of the spirit (ruaʚ), these inspired people are ahead of their time; they represent the religion of the future … The prophetic phenomenon is at the very heart of the Old Testament. One may distinguish between the Speaking and the Writing Prophets. Nevertheless the appearance of the latter, during the eighth century, is accidental. The book has no other function than to extend their preaching.”1 Fifty years later, such a presentation of the biblical prophets and the prophetic books is no longer possible. The major shift in current research on the prophetic phenomenon in the Hebrew Bible took place in the 1970s. Before that time, the scholarly debate about the prophets was mainly interested in reconstructing their ipsissima verba, the Prophet’s authentic words. Commentaries on the prophetic books were mainly concerned to distinguish between authentic and non-authentic oracles, the former alone being considered important for the understanding of the prophetic message (and also for the faith of the ancient as well as André GELIN, “Les livres prophétiques postérieurs,” in A. ROBERT and A. FEUILLET (ed.), Introduction à La Bible. Tome I Tournai: Desclée & Cie, 1959, 467–582, 471–472. Translation is mine. 1




modern readers). The prophetic book was nothing else than the record of the prophet’s preaching, to which some “disciples” and redactors had unfortunately added invented passages. The new method of redaction criticism radically changed the perspective.2 The personality of the prophet is no longer the center of interest. The important task is to explain the reasons for the existence of prophetic books in the second part of the Hebrew Bible. In the most radical redaction critical approaches the prophetic person even disappears (or if he remains he may have pronounced an insignificant amount of oracles which were totally altered by later redactors).3 The existence of the prophetic book does not result from the activity of a prophetic individual but is the work of scribes and redactors. Or to put it as does Konrad Schmid: “There are no ‘authentic’ prophets in the whole Old Testament. The first writing down of oral logia and the choice of how to combine different smaller literary units are already acts of interpretation.”4 In the last twenty years the redaction critical approach has been combined with investigations on the compositional techniques with which the scribes and/or redactors fostered the unity 2 It is often argued that the origin of this approach can be found in the work of Willy MARXSEN, Der Evangelist Markus. Studien zur Redaktionsgeschichte des Evangeliums (FRLANT 49; Göttingen: Vandenhoeck & Ruprecht, 1956); ET: Mark, the Evangelist: Studies on the Redaction History of the Gospel (Ashville - New York: Abingdon 1969). See Norman PERRIN, What is Redaction Criticism? (London: SPCK, 1970), who points out that Marxsen “is responsible for the name Redaktionsgeschichte” (p. 33) even if he had some forerunners. 3 Compare Robert P. CARROLL, Jeremiah (OTL; London: SCM Press, 1986) and Susanne Rudnig-Zelt, Hoseastudien. Redaktionskritische Untersuchungen zur Genese des Hoseabuches (FRLANT 213; Göttingen: Vandenhoeck & Ruprecht, 2006). Rudning-Zelt is much more confident in the possibilities of Literarkritik, but both agree about the impossibility to connect the “historical” prophet and the book. 4 Konrad SCHMID, “La formation des Prophètes Postérieurs (Histoire de la rédaction)” (ed. Thomas RÖMER, Jean-Daniel MACCHI and Christophe NIHAN, Introduction à l'Ancien Testament (MdB 49); Geneva: Labor et Fides, 2004), 318–328, 320. See further Ehud BEN ZVI, “The Prophetic Book: A Key Form of Prophetic Literature,” in The Changing Face of Form Criticim for the Twenty-First Century (ed. Marvin A. SWEENEY and Ehud BEN ZVI; Grand Rapids, Michigan - Cambridge, UK: 2003), 276–297.



of the prophetic books. In 1983 Rolf Rendtorff argued that the book of Isaiah couldn’t be understood as a simple juxtaposition of a Proto-, Deutero-, and Trito-Isaiah. There are thoroughgoing themes and catchwords (the idea of consolation: 12,1; 40,1; 66,13; kabod yhwh: 6,3; 40,5; 66,18, the designation of Yhwh as the “holy one of Israel,” etc), which show that the same redactor or group of scribes edited all 66 chapters of the book.5 This interest in compositional techniques is also a new trend in Psalms research. It appears that these should not be considered as strictly independent units, which have been gathered by juxtaposition. The group that edited the psalms apparently had in mind to organize them according to thematic and theological devices.6 Does this mean that these scribes wanted their audience to hear or their readers to read the psalms as a “book” from the beginning to the end? The same question is, even more heavily, debated as to the socalled “Book of the Twelve,” the “Minor Prophets.” There is no doubt that these twelve prophets were considered in the last three or two centuries B.C.E. to constitute one book, approximately as long as the book of Isaiah (see 4 Ezra 14,44; Contra Apionem 1:38– 42). And there is no doubt that the Dodekapropheton presents an arrangement that suggests the reader a walk through history from the time when Israel (the North) existed through the Assyrian and Babylonian invasions and deportations to the rebuilding of the Jerusalemite Temple in the Persian period. One may nevertheless ask the question whether this chronological organization was at the very beginning of the “Twelve” or whether it is a much later attempt to organize originally independent documents in a meaningful way. Should one speak of a “Book of the Twelve” or better of

5 Rolf RENDTORFF, Das Alte Testament. Eine Einführung (NeukirchenVluyn : Neukirchener, 1983), 210–212; ET: The Old Testament. An Introduction (London: SCM Press, 1985). See also “Zur Komposition des Buches Jesaja,” VT 34 (1984), 295–320. 6 Frank-Lothar HOSSFELD and Erich ZENGER, Die Psalmen. Psalm 1– 50 (NEB 29; Würzburg: Echter Verlag, 1993); Psalmen 51–100 (HThKAT; Freiburg - Basel - Wien: Herder, 2000). For a skeptical evaluation of this idea: Erhard S. GERSTENBERGER, “Theologies in the Book of Psalms,” in The Book of Psalms. Composition and Reception (ed. Peter W. FLINT and Patrick D. MILLER; VT.S 99; Leiden - Boston: Brill, 2005), 603–626.



an “anthology” or a compilation of different prophetic scrolls brought together for economical or archival reasons? The two contributions in this volume by James Nogalski and Ehud Ben Zvi reflect the current state of the discussion. These authors present opposite views about the formation and composition of the Twelve. Both articles originated in presentations for a doctoral program for students of the Hebrew Bible, organized by the French speaking universities in Switzerland (Fribourg, Geneva, Lausanne, and Neuchâtel) and held in Geneva on December 6th 2008. In the name of the organizers of this meeting I would like to express my warmest thanks to Professors Ben Zvi and Nogalski for the long trip they undertook and especially for their excellent papers and their openness to discussion. Even if this discussion did not resolve the question, it certainly contributed to clarification and to a better understanding of both positions. It is therefore a good initiative to publish both articles together in order to make them available to a broader audience.7 I do not want to take any position or choose my camp in writing this brief introduction. Instead, I would like to highlight some of the questions and arguments trying to locate those in the broader context of scholarly research on the Hebrew Bible. Let us begin with some agreements. Both authors agree that at the end there is a “book of the Twelve” and that it is possible to read the twelve “chapters” of this book as being part of a comprehensive prophetic “message” (even if there is disagreement about the degree of coherence). Nogalski and Ben Zvi also share the opinion that the specific character of each of the twelve prophets is to be acknowledged in one way or another. Neither Nogalski nor Ben Zvi argue for the idea that all the components of the Book of the Twelve were written at the same time by the same author. However, they strongly disagree on the meaning of the Twelve. Is the Book of the Twelve an idea that already originated in the Persian period (Nogalski) or did the idea emerge much later and without real impact for the understanding of the twelve Minor Prophets (Ben Zvi)? French versions of both papers will be published together with other contributions to the doctoral program which dealt with the origins and formation of the prophetic books. 7



It is not astonishing that Nogalski and Ben Zvi emphasize different arguments and facts. Nogalski puts much weight on the chronological continuity in the Book of the Twelve, which can hardly be disputed. There are, however, some differences between the sequence of the Twelve in the Massoretic Text and the LXX.8 Nogalski asserts that the Massoretic tradition preserves the original arrangement, but this view is disputed.9 The sequence of LXX perhaps reflects the tripartite eschatological scheme that also structures the books of Isaiah; Jeremiah LXX and Ezekiel.10 In this case we would have for the Twelve a situation that compares to the book(s) of Jeremiah: There are apparently two competing ideas about the sequence and the way in which the Book of the Twelve should be read. For Ben Zvi, the difference between MT and LXX indicates that in the Persian time there was no idea of a unified book, since there were different possibilities to arrange the (ten or already twelve?) scrolls together. But would this argument also apply to the book of Jeremiah? Apparently even Jeremiah was conceived in different ways as one book, in the Massoretic as well as in the LXX arrangement. Whereas Nogalski insists on the presence of catchwords at the “seams” of each prophetic unit, Ben Zvi points to the fact that the Book of the Twelve has no general title; most of the twelve books, however, display individual titles, which are a signal that the reader

8 In LXX, Amos comes second and Joel fourth, and Jonah precedes Nahum (the whole LXX sequence is: Hos, Am, Mi, Joel, Ob, Jon, Nah, Hab, Zeph, Hag, Zech, Mal). 9 Marvin A. SWEENEY, “Sequence and Interpretation in the Book of the Twelve,” in Reading and Hearing the Book of the Twelve (ed. James D. NOGALSKI and Marvin A. SWEENEY; Society of Biblical Literature. Symposium Series, 15; Atlanta: Society of Biblical Literature, 2000), 49–64. 10 Konrad SCHMID, “Die Literatur des Alten Testaments. II. Hintere Propheten (Nebiim),” in Grundinformation Altes Testament (ed. Jan Christian GERTZ; Uni-Taschenbücher 2745; Göttingen: Vandenhoeck & Ruprecht, 2007), 303–401, 362–63. For another explanation see Jean-Daniel MACCHI, “Les douze petits prophètes,” in Introduction à l'Ancien Testament (ed. Thomas RÖMER, Jean-Daniel MACCHI and Christophe NIHAN; Le Monde de la Bible, 49; Genève: Labor et Fides, 2004), 379–382.



should understand them separately.11 The presence of catchwords and overlapping themes can easily be detected (some examples: Jo 4,16 echoes Am 1,2; Am 9,2–4 prepares Ob 4; Zeph 3,19 parallels Hag 1,2 etc)12 but were the ancient hearers or readers aware of them? They did not have concordances or computers at their disposal to quickly check out parallels or recurring words. It is also true, as recalled by Nogalski that Exod 34,6–7 and the “day of Yhwh”13 play an important role in the Book of the Twelve. Nevertheless, as Ben Zvi rightly underlines, none of these themes is limited to the Twelve. The same holds true for recurring words; most of them also appear in the three Major Prophets or in other texts of the Hebrew Bible. The differences between Nogalski and Ben Zvi also reveal different methodological approaches. Nogalski clearly favors the redaction critical approach combining it with insights from synchronic methods (intertextuality, compositional techniques), whereas Ben Zvi investigates the text above all as a historian interested in the implied ancient readers’ perspective in the Persian (or early Hellenistic) period. Nogalski is sympathetic to reconstructions of earlier stages of the Book of the Twelve, especially the “Book of the Four” (Hos, Am, Mi, and Zeph, as advocated by Albertz and Wöhrle14) and a scroll containing Haggai and Zech 1–8 (Schart).15 Ehud BEN ZVI, “The Prophetic Book: A Key Form of Prophetic Literature,” in The Changing Face of Form Criticism for the Twenty-First Century (ed. Marvin A. SWEENEY and Ehud BEN ZVI; Grand Rapids/Cambridge: William B. Eerdmans Publishing Company, 2003), 276–297. 12 James D. NOGALSKI, “Intertextuality in the Twelve,” in Forming Prophetic Literature. Essays on Isaiah and the Twelve in Honor of John D. W. Watts (ed. James W. WATTS and Paul R. HOUSE; Journal for the Study of the Old Testament. Supplement Series 235; Sheffield: Sheffield Academic Press, 1996), 102–124. 13 James D. NOGALSKI, “The Day(s) of YHWH in the Book of the Twelve,” in Thematic Threads in the Book of the Twelve (ed. Paul L. REDDITT and Aaron SCHART; BZAW 325; Berlin / New York: Walter de Gruyter, 2003), 192–213. 14 Rainer ALBERTZ, “Exile as Purification. Reconstructing the ‘Book of the Four,’” in Thematic Threads of the Book of Twelve (ed. Paul L. REDDIT and Aaron SCHART; BZAW 325; Berlin - New York: de Gruyter, 2003), 232–25; Jakob WÖHRLE, Die frühen Sammlungen des Zwölfprophetenbuches. Entstehung und Komposition (BZAW 360; Berlin / New York: Walter de 11



Ben Zvi without denying the fact that the prophetic books result from a long process of rewritings and redactional activity is very skeptical about the possibility of reconstructing the different stages of their formation. He claims that the redactors swiped away all traces of editing and additions because they wanted the readers to read one book of Hosea, Amos etc. Who were the authors of the Book of Twelve? In his paper Nogalski makes the interesting suggestion that the composition of the Book of the Twelve was made by a group of Levites during the Persian period, who edited the Book of Twelve by inserting hymnal passages having in mind a public (and cultic?) reading of the Book. Ben Zvi ascribes each scroll of the Minor Prophets to the “literati” of the Persian period, who were also the authors of Isaiah, Jeremiah, and Ezekiel. Here things may become somewhat paradoxical. If one follows Ben Zvi’s idea that the same group of literati were at the origin of all prophetic books and were able to create any “deuteronomistic,” “priestly” or whatever style, one could conclude, contrary to what he argues, that this strengthens the idea of the unity of the book of the Twelve. On the other hand, Nogalski, emphasizing the original independency of most of the books of the Twelve and postulating late redactional interventions, recognizes in a way the specificity of most of the Twelve’s components. So maybe “the truth is out there” to pick up the slogan of a popular science fiction television series. The articles by Nogalski and Ben Zvi printed in this volume raise important questions, which are to be addressed in any field of historical critical research on the Hebrew Bible. Without pretending to be exhaustive I would emphasize the following: How do “redactors” or “literati” work? Ben Zvi insists on the fact that biblical texts nowhere inform their readers about the redactional processes they underwent: “The ongoing process of redacGruyter, 2006). Jörg JEREMIAS, Der Prophet Hosea (ATD 24/1; Göttingen: Vandenhoeck & Ruprecht, 1983) and Schart (see next footnote) argued that only Hosea and Amos were put together in a first step. 15 Aaron SCHART, Die Entstehung des Zwölfprophetenbuches: Neubearbeitungen von Amos im Rahmen schriftenübergreifender Redaktionsprozesse (BZAW 206; Berlin - New York: de Gruyter, 1998). See also Jakob WÖHRLE, “The Formation and Intention of the Haggai-Zechariah Corpus,” Journal of Hebrew Scriptures 6/10 (2006):



tion was not bent on promoting, or archiving and analyzing itself; instead its function was to shape a series of texts in which the last, if successful, was meant to supersede and erase the memory of the previous one.”16 There are however indications that some texts at least do inform the reader about their growth. The best example is Jer 36. In this narrative the reader learns that there were at least two different scrolls containing oracles of Jeremiah, and that the second one had been enlarged at the time it had been written on the basis of the first scroll. There are also literary devices such as juxtaposition (see for instance Gen 16,9–11) or the Wiederaufnahme (repetitive resumption, see for instance the repetition of Am 7,9 in 7,17 in order to insert 7,10–17), which informs an attentive audience about the growth of a text. How to analyze the presence or lack of titles and conclusions? It is a fact that the Book of the Twelve has no general title whereas most of the Twelve display a superscription. But if one takes a look at the Pentateuch there is no title at the beginning either. Bereshît bara’ certainly is not a title for the whole Torah, but for Gen 1 or 1–3. The case of the Pentateuch is interesting for the debate about the Twelve, because the same questions arise. As for the Twelve, the Pentateuch or even the Hexateuch including Joshua is constructed as a chronological sequence: from the beginning of the world until the death of Moses or the conquest of the land and the death of Joshua. The only book of the Torah that bears a real title is the book of Deuteronomy.17 There are three identifiable concluding formulas in Lev 26,46; 27,34 and Numb 36,13, which interrupt the sequence between Leviticus and Numbers. How to take these observations into account? In one stage of its formation the Torah was clearly meant to be read as one “book,” but it is also obvious that the different narrative and legal materials that are now incorporated into the Penta- or Hexateuch were originally transmitted independently. The construction of a chronological framework is perhaps linked with the promulgation of a (proto-) Pentateuch in the middle of the Persian era. So when the Torah claimed to be authoritative it was presented as one “book,” but without hiding See BEN ZVI (below), p. 59. This observation could give credit to the original distinction between Deuteronomy and the “Tetrateuch.” 16 17



the diversity of the compiled material. Interestingly, the same observations may apply to the “Enneateuch” because the books of Genesis to Kings also follow a chronological progression from the beginning of the world to the destruction of Jerusalem, or from expulsion from the paradise to the expulsion from the land.18 Again, it is difficult to explain the aim of this arrangement. Were all these books meant to belong and to be read together? Was the Enneateuch an attempt to create the Israelite historia despite the (later) canonical separation between Torah and Nebiim? The so-called Primary History was certainly never understood as a single book, but this did not exclude the possibility of a chronological reading of the story. Could the same also apply to the Twelve? The question of the end of the Dodekapropheton. There is an ongoing debate about the question whether the book of Malachi was always an independent unit or whether it was part of Zechariah. Another debated issue is the status of the book of Jonah inside the Twelve. Recent investigations argue for a late date and a late integration of this untypical book into the Twelve.19 Was it perhaps conceived to stand at the very end of the collection? One Qumran manuscript attests the position of Jonah after Malachi20 but unfortunately there is no evidence for the other books in this manuscript so the question must be left open. Coming back to the very end of Malachi, one should ask whether the last verses Mal 3,22–24 were conceived to conclude only the Book of the Twelve.21 It seems to me that the final passage of Malachi was added in order to conclude the whole Nebiim, which means that these verses were written at earliest in the Bernard GOSSE, “L' inclusion de l'ensemble Genèse - II Rois, entre la perte du jardin d'Eden et celle de Jérusalem,” ZAW 114 (2002): 189– 211. 19 Ehud BEN ZVI, The Signs of Jonah: Reading and Rereading in Ancient Yehud (Journal for the Study of the Old Testament Supplement Series 367; Sheffield: Sheffield Academic Press, 2003); SCHMID, “Propheten,” 382– 382. 20 Odil H. STECK, “Zur Abfolge Maleachi – Jona in 4Q76 (4QXIIa),” ZAW 108 (1996), 249–253. 21 Hans-Peter MATHYS, “Anmerkungen zu Mal 3,22–24,” in Vom Anfang und vom Ende. Fünf alttestamentliche Studien (Beiträge zur Erforschung des Alten Testaments und des Antiken Judentums 47, Frankfurt am Main: Peter Lang, 2000) 30–40. 18



second century B.C.E. The expression “law of Moses” (3,22), does not occur elsewhere in the Latter Prophets but in the Former Prophets forms an inclusion with Josh 1,7–8. Both texts emphasize the hermeneutical principle that the Prophets can only be read and understood in reference to the Law. The mention of the return of Elijah (3,23) establishes a link with the Former Prophets of whom Elijah is the main character and enables or confirms the possibility of an eschatological reading of the Twelve. Did the Book of the Twelve originally end with Mal 3,21, a passage that contains thematic similarities with the last verse of the book of Isaiah and that also mentions the imminence of Yhwh’s day? Was this an invitation for the ancient readers of the Twelve to understand the whole book in this eschatological, proto-apocalyptical perspective? Or is this reading limited to Malachi? The answer will be different whether one follows the arguments of Nogalski or those put forward by Ben Zvi. However, both options are not necessarily exclusive, as shown by the history of reception of the twelve Minor Prophets.

ONE BOOK AND TWELVE BOOKS: THE NATURE OF THE REDACTIONAL WORK AND THE IMPLICATIONS OF CULTIC SOURCE MATERIAL IN THE BOOK OF THE TWELVE1 JAMES D. NOGALSKI BAYLOR UNIVERSITY A spate of redactional studies since 1993 has made progress in developing a comprehensive reconstruction of the redactional history of the Book of the Twelve. The first section of this paper will explore these recent discussions in three sections: areas where progress toward a consensus has been made, followed by a response to some of the objections raised about the task, and an assessment of the nature of the redactional work which isolates four areas of continuing debate. The second portion of the essay will focus on a fresh proposal which, it is hoped, will stimulate discussion that will move the discussion of redactional work on the Book of the Twelve forward. Namely, it will be argued that the role of cultic source blocks redactionally incorporated into the Book of the Twelve has not been adequately appreciated. The extent of this cultically oriented material should invite a reevaluation of the function of cultic material in the literary context and the implications of the use of this material for understanding the tradents.

1 I wish to thank Paul Redditt, Aaron Schart, and Roy E. Garton for commenting on a draft of this paper, thereby strengthening its presentation, though they are not responsible for any confusion which may remain.




1. RECENT REDACTIONAL INVESTIGATIONS ON THE BOOK OF THE TWELVE Progress Toward a Consensus It is now widely accepted that the Book of the Twelve should be treated as a redactional unit. Progress toward a consensus has occurred on three fronts: the deliberative nature of the order of the writings, the priority of the MT sequence over the LXX, and the extended transmission of two pre-existing corpora. First, there is wide agreement that the ordering of the twelve writings is hardly inadvertent. Chronology plays too clear a role in the placement of the writings to be accidental. Dated superscriptions create a chronological frame that unfolds across the Twelve. These dated writings move from those writings attributed to eighth century prophetic figures (Hosea, Amos, Micah) to a seventh century figure (Zephaniah). This group is followed by writings with a late sixth century focus on the rebuilding of the temple (Haggai, Zechariah). Four additional writings tie directly into this chronological movement. Malachi assumes a Persian period setting after the temple has been reconstructed, which secures its position after Haggai and Zechariah. Nahum and Habakkuk deal with a theological portrayal of the seventh century, requiring that they come after Micah, the last eighth century prophetic figure. Jonah, on the other hand, presents a narrative about a prophet who lived in the days of Jeroboam II, and this requires that Jonah be placed prior to Micah since the three kings mentioned in Mic 1,1 all postdate the time of Jeroboam II. Thus, ten of the Twelve owe their location to a chronological framework. In addition to chronology, themes, catchwords, and citations play a role in the location of many of the writings in the Twelve. Two of the clearest examples of these contextual links appear in Joel and Obadiah (ironically, the two writings which do not exhibit the chronological orientation of the other ten). The last few verses of Joel contain citations of Amos. Joel 4,16 draws on Amos 1,2 (the opening verse of Amos after the superscription), and Joel 4,18 parallels Amos 9,13 (the third to the last verse in the book). Synchronically speaking, then, Joel effectively encompasses the beginning and end of Amos. Similarly, the essential theme of Obadiah is already anticipated in Amos 9,12.



Several of the writings which presume the chronological framework also make sense thematically in their setting. By way of example, the fall of Assyria in Nahum and the rise of Babylon in Habakkuk will be explored later in this paper. Malachi deals with issues of the postexilic community after the temple has been rebuilt wherein they must once again confront problems related to improper sacrifice. Some have argued (or assumed) the thematic overlap was the result of choices in positioning more or less completed writings, but most have seen some degree of editorial work evident in the process of arranging these twelve writings.2 A second, and related, area of relative consensus concerns the priority of the MT sequence of writings when compared to the LXX. Conversations over the last 15 years have tended to conclude that the MT represents the oldest order, though questions about the rationale for the LXX’s changing the order have not been entirely resolved.3 Occasional arguments for LXX priority have proven incapable of convincing the overwhelming majority that the LXX order takes priority in the case of the Twelve. More importantly for future study, a sustained treatment of the LXX has not been undertaken to explore the implications of the changes in sequence by treating the LXX as an entity in its own right, though Marvin Sweeney has suggested that a significant shift in emphasis on the nations occurs for the reader when the sequence is

2 This

issue will be discussed in more detail below. The order of the MT (Hosea, Joel, Amos, Obadiah, Jonah, Micah) and LXX (Hosea, Amos, Micah, Joel, Obadiah, Jonah) differ among the first six writings while the last six are identical in both traditions. One other sequence may be exhibited in one Qumran manuscript which seems to present Jonah after Malachi, but this order is idiosyncratic to this point and there is no conclusive proof that other writings from the Twelve were included on the same scroll. For a more complete discussion of these issues, see Russell FULLER, “The Formation of the Book of the Twelve: The Evidence from the Judean Desert,” in Forming Prophetic Literature; Essays on Isaiah and the Twelve in Honour of John D.W. Watts (James W. WATTS and Paul R. HOUSE, ed.; JSOTSup 235), Sheffield, Sheffield Academic Press, 1996, 86–101; Barry Alan JONES, The Formation of the Book of the Twelve: A Study in Text and Canon (SBLDS 149), Atlanta, Society of Biblical Literature, 1995, 6–7, 237–239; and Odil Hannes STECK, “Zur Abfolge Maleachi – Jona in 4Q76 (4QXIIa),” ZAW 108, 1996, 249–253. 3



changed.4 Others have been satisfied to note that the change of sequence can be explained simply as a decision to group the three eighth century prophets (Hosea–Amos–Micah) together which then leaves Joel–Obadiah–Jonah in the same sequence as the MT.5 A third area of wide-spread agreement appears in the suggestion that two pre-existing collections created the foundation for what comes to be the Book of the Twelve. One group comprises most of Hosea–Amos–Micah–Zephaniah, while the other contains Haggai–Zechariah 1–8. Each of these collections were “published” and edited together prior to their association with the remaining writings. The first of these collections has been called by various names, but it includes Hosea, Amos, Micah, and Zephaniah. Nogalski called it the Deuteronomistic Corpus because of certain similarities these writings share with the theological perspective of the Deuteronomistic History.6 Schart concurred with the idea of a common redactional history shared by these four writings, but he qualified Nogalski’s observations in at least two significant ways. First, he delineated the common history more carefully (arguing that Hosea and Amos shared a history prior to the incorporation of Micah and [only later by including] Zephaniah). Second, Schart concludes that “Deuteronomistic” is not the best term, in part beMarvin A. SWEENEY, “Sequence and Interpretation in the Book of the Twelve,” in Reading and Hearing the Book of the Twelve (James D. NOGALSKI and Marvin A. SWEENEY, ed.; Symposium 15), Atlanta: Society of Biblical Literature, 2000, 49–64. 5 See already Dale Allan SCHNEIDER, The Unity of the Twelve, Ann Arbor: University Microfilms International, 1979, 224–225; James D. NOGALSKI, Literary Precursors to the Book of the Twelve (BZAW 217), Berlin, De Gruyter, 1993, 2–3. Aaron Schart implies the connections between Hosea, Amos, and Micah go even further, especially given the similar dates in the superscriptions, as well as other similarities of content between these three writings. See Aaron SCHART, “Reconstructing the Redaction History of the Twelve Prophets: Problems and Models,” in Reading and Hearing the Book of the Twelve, 37–38. See also Aaron SCHART, Die Entstehung des Zwölfprophetenbuch: Neuarbeitungen von Amos im Rahmen schriftübergreifender Redaktionsprozesse (BZAW 260), Berlin, De Gruyter, 1998, 218–220. 6 See the summaries in NOGALSKI, Literary Precursors, 276–280; and James D. NOGALSKI, Redactional Processes in the Book of the Twelve (BZAW 218), Berlin, De Gruyter, 1993, 274–275. 4



cause so many of the phrases typically considered to be distinctively Deuteronomistic elsewhere are lacking in this group of four. Schart, therefore, suggests the term D-corpus as a means of distinguishing this corpus from “Deuteronomic/Deuteronomistic” materials that use Deuteronomic phraseology more consistently.7 Others, such as Albertz, prefer a more neutral term, and simply speak of the Book of the Four Prophets (Vierprophetenbuch).8 Most recently, the reason for the debate has been clarified considerably with the work of Wöhrle (who also calls this corpus the Book of the Four Prophets). He shows that the “Deuteronomistic” flavor points to a phenomenon whereby these writings periodically, but specifically, allude to the accounts of Hezekiah and Josiah in 2 Kings 17; 20; 22–25.9 As a result of these conversations, the idea that these four writings were shaped together over time now rests on fairly stable ground. The second pre-existing corpus arises from the common editing of Haggai and Zechariah 1–8. Since this suggestion had already been proposed in Old Testament scholarship prior to recent investigations of the Twelve as a corpus, it is not surprising that this idea has received strong endorsement among Book of the Twelve scholars.10 A few remain skeptical, but the majority of scholars now sees editorial connections, at least, between Haggai and Zechariah


Die Entstehung, 156, 218–233. See Rainer ALBERTZ, “Exile as Purification: Reconstructing the ‘Book of the Four,’” in Thematic Threads in the Book of the Twelve (Paul L. REDDITT and Aaron SCHART, ed.; BZAW 325), Berlin, De Gruyter, 2003, 232–251. For a discussion of the background issues see Rainer ALBERTZ, “In Search of the Deuteronomists: A First Solution to a Historical Riddle,” 1–17 in The Future of the Deuteronomistic History (Thomas RÖMER, ed.; Bibliotheca Ephemeridum Theologicarum Lovaniensium 147), Leuven: Peeters, 2000. 9 See the charts and summary in Jakob WÖHRLE, Die frühen Sammlungen der Zwölfprophetenbuches: Entstehung und Komposition (BZAW 360), Berlin, De Gruyter, 2006, 269–270, 275–282. 10 See especially the seminal works of W. A. M. BEUKEN, Haggai - Sacharja 1–8. Studien zur Überlieferungsgeschichte der frühnachexilischen Prophetie (Studia Semitica Nederlandica 10), Assen, Van Gorcum, 1967; and Rex MASON, “The Purpose of the ‘Editorial Framework’ of Haggai,” VT 27, 1977, 413–421. 8



1–8.11 The background of these two writings shows concern for explaining the prophetic role in the rebuilding of the temple and in the leadership of society in the late sixth century. To summarize, progress has been made toward a consensus in three areas: 1) recognizing that the arrangement of the Twelve writings in the corpus is both deliberate and reflects significant editorial work; 2) the LXX order derives from changes made to the MT sequence; 3) two pre-existing multivolume corpora, when combined, not only account for the chronological framework across the growing Book of the Twelve, they also provide the substance of many of the connecting motifs. These three areas have been treated from various perspectives, and significant points of agreement have been isolated. However, before moving to discuss the nature of the redactional material, it should be noted that scholars have also voiced skepticism or caution concerning common redactional work among the writings of the Book of the Twelve, or at least about modern scholarship’s ability to reconstruct this activity. Doubts about the Task Alongside progress toward a consensus, doubts concerning the process of finding “Buchübergreifende” redactions have been expressed in works by Ben Zvi, Beck, and Petersen. Ehud Ben Zvi voices two reasons for taking a cautious approach to the issue of common redaction in/among the writings of the Book of the Twelve: the lack of a single title, and the lack of control for parallels.12 See recent commentary discussions in David L. PETERSEN, Haggai and Zechariah 1–8 (OTL), London, SCM, 1984, 37–39; and Eric M. MEYERS and Carol L. MEYERS, Haggai, Zechariah 1–8 (AB 25B), Garden City, Doubleday, 1987, xliv-xviii. A few have argued for the inclusion of Malachi and/or Zechariah 9–14 with this corpus, but most reject these arguments. The first to make this argument in a sustained fashion was Ronald PIERCE, “Literary Connectors and a Haggai-Zechariah-Malachi Corpus,” JETS 27, 1984, 277– 89. 12 Ehud BEN ZVI, “Twelve Prophetic Books or ‘The Twelve’: A Few Preliminary Considerations,” in Forming Prophetic Literature; Essays on Isaiah and the Twelve in Honour of John D.W. Watts (James W. WATTS and Paul R. HOUSE, ed.; JSOTSup 235), Sheffield, Sheffield Academic Press, 1996, 125–156. See also Martin BECK, Der ‘Tag YHWHs’ im Dodekapropheton. 11



First, Ben Zvi claims the Book of the Twelve contains no title indicating it presents itself as a single work.13 In so doing, Ben Zvi puts his finger on an important characteristic of the Twelve, but his concerns do little to address the issue of common redaction. Ancient traditions concerning the Twelve as a single corpus were not an invention of 20th century critical scholarship. These traditions appear to be based on intertextual interplay between these Twelve writings. Further, evaluation of other Old Testament works shows that titles are only one way that editorial connections were made. The Torah has no titles as such, but few would doubt the interconnectedness of the edited works. Chronicles leads to Ezra by repeating two verses, thereby joining the end of Chronicles (2 Chr 36,22– 23) with Ezra (1,1–2). Consider evidence from the Leningrad Codex. The book of Psalms has no unifying title for the corpus in the Leningrad Codex. It does have superscriptions that set off many of the individual Psalms, though most of the superscriptions do not appear on a line by themselves.14 Despite these superscriptions at the beginning of individual psalms, the Psalter is widely recognized to have undergone editorial work on groups of psalms which are not named as editorial units in the corpus as we have it.15 The nature, history, Studien im Spannungsfeld von Traditions- und Redaktionsgeschichte (BZAW 356), Berlin, Walter de Gruyter, 2005, 17. 13 BEN ZVI, “Twelve Prophetic Books,” 137. 14 Individual psalms are separated from the preceding psalm with one blank line in the codex. The superscriptions are usually followed by a gap of 3–10 letters, like the divisions between any other verse. Some psalms (e.g., Ps 15; 50) do have the superscription on a line by themselves, but this is not typical. 15 Beginning with the influential work of Gerald Henry WILSON, The Editing of the Hebrew Psalter (SBLDS 76), Chico, California: Scholars Press, 1985, work on the editing of the psalter has been ongoing. See a good summary of the redactional growth of the Psalter in Klaus SEYBOLD, Die Psalmen: Eine Einführung (Kohlhammer Urbantaschenbücher 382), Stuttgart, Kohlhammer, 1986, 36–54. See also the recent discussion in Joel S. BURNETT, “A Plea for David and Zion: The Elohistic Psalter as Psalm Collection for the Temple’s Restoration,” in Diachronic and Synchronic: Reading the Psalms in Real Time (Joel BURNETT et al, ed.; Library of Hebrew Bible/Old Testament Studies 488), London, T&T Clark, 2007, 95–113; for an example of reading an extended section of the Psalter, see also W. H.



and function of this editorial work is still a matter of discussion, but its existence is largely assumed in recent studies of the Psalter. Moreover, none of the prophetic books in the Leningrad Codex have their own title; they merely start after a gap of a few lines.16 There are, however, Masoretic notes at the end of the Twelve, and at the mid-point of the corpus which do treat the Book of the Twelve as a unit in its own right.17 Thus, Ben Zvi’s argument about the lack of a unifying title for the Twelve carries little weight on the issue of common redaction, but evaluating manuscripts does provide evidence that favors treating the Twelve as a corpus, not just Twelve disconnected writings. The existence or absence of a title is not indicative of common editorial work. Ben Zvi offers a second word of caution about reconstructing common redactional work among the writings of the Twelve when he argues that many of the observations regarding the interplay between two texts lack controls to establish them as parallels and links.18 Ben Zvi raises a valid concern, but only to a point. More attention has been paid to this issue in literature subsequent to his essay.19 Still, one has to recognize that very few editorial connec-

BELLINGER, “Reading from the Beginning (Again): The Shape of Book I of the Psalter,” in Diachronic and Synchronic, 114–126. 16 It should be noted that the number of lines separating the writings of the Book of the Twelve from one another is smaller than the number of lines separating the book of Isaiah, Jeremiah, and Ezekiel from their neighboring books. This pattern appears to conform to the rules described in Baba Batra 13b. Thus, even the separation of the writings of the Book of the Twelve illustrate they were treated differently. 17 Consider the Masoretic note in the margin beside Mic 3,12. It reads: “One half of the scroll by verses.” The notes at the end of the Twelve (right after notes concerning Malachi) reference the total number of verses for the Book of the Twelve (1,050) and cite Mic 3,12 as the midpoint of the corpus. Simple arithmetic shows that the numbers used in these verse counts are accurate. 18 BEN ZVI, “Twelve Prophetic Books,” 135–137, 139–142. 19 See, for example, evaluations of the catchword connections proposed by Nogalski and evaluated by Schart and others in later works. Not every argument has proven convincing, but these evaluations do not negate the use of catchwords as a technique. For example, see SCHART, Entstehung, 97 (positively) or 230 (negatively). See also several of the essays in



tions in the prophetic canon could meet all of the criteria laid out by Ben Zvi. It is the nature of the connections that they demand a high degree of awareness of the content on the part of the reader and editor. Moreover, he ignores the fact that these connecting devices are used within the context of redactional studies in other “scrolls” or “writings.” These methods are not limited to the Book of the Twelve. Using his criteria, for example, one must eliminate eo ipso any connections created by common words because they are common, but some thematic lines of connection are created by combinations of relatively common words.20 In the end, it should be the interpretive contributions to the task and the convincing power of the proposals (including the evaluation of whether recurring words appear by accident) which determine the usefulness of a proposed link. The interest generated by redactional studies of the last 15–20 years and the intersection of results makes the task promising. Ben Zvi also calls for a more careful process of defining words like audience and tradents. He is correct that more attention needs to be paid to this area. The study of prophetic tradents still stands at the front end of a very complicated task. The task of defining tradents more carefully will be addressed in more detail below by exploring the role of cultic texts and cultic concerns in the growth of the Book of the Twelve, especially in those six writings where more debate exists.21 Ben Zvi’s cautions are taken further by Martin Beck, who presents a sustained, skeptical counter-argument against the idea of an extended redactional history affecting the Book of the Twelve as a corpus.22 He contends that the Book of the Twelve is really an anthology highlighting the day of YHWH, but that the writings were Erich ZENGER, ed., “Wort JHWHS, das geschah . . .”: Studien zum Zwölfprophetenbuch (Herders Biblische Studien 35), Freiberg, Herder, 2002. 20 In the Book of the Twelve, the most prominent example is the recurring motif of the fertility of the land which is often signaled by combinations of common words connoting the presence or absence of agricultural bounty: wine, vine, grain, and fig-tree. See James D. NOGALSKI, “Recurring Themes in the Book of the Twelve: Creating Points of Contact for a Theological Reading,” Interpretation 61, 2007, 128–130. 21 See part two of this essay (“Form and Function of the Redactional Work,” beginning on 14). 22 BECK, Der ‘Tag YHWHs’ im Dodekapropheton, 2005.



only connected in two stages at a very late date (the Hellenistic period), first as a book of ten. Later, Jonah and Malachi were added to make the Book of the Twelve. Thus, the writings were independently transmitted until quite late according to Beck. Catchwords are dismissed as too unconvincing, or unprovable. Quotes are explained away as shared tradition history or citations so limited in scope as to have no literary value when trying to understand the corpus. He denies these texts have an expanded literary horizon. He is very similar to Ben Zvi in this regard. He frequently challenges the conclusions of those who have argued for connective redactional implantations by systematically discounting the observations of Nogalski, Schart, Bosshard-Nepustil, etc. In the end, he rejects so many of the arguments as failing to meet his list of criteria that one almost wonders why he speaks of the Book of the Twelve at all. Like Ben Zvi, Beck makes some valid points about the need to argue carefully and account for opposing views consistently. However, Beck falls into many of the same traps (perhaps unavoidably) of those he critiques when he begins making his own case for the importance of the day of YHWH. For example, he maintains that the day of YHWH language in both Joel and Zephaniah could be more easily explained as Joel drawing upon Zephaniah than the implanting of Joel language into Zephaniah. However, he concludes that the best explanation is to deny there are any strong connections between the two writings since Exodus traditions and Zion theology so prominent in Joel are lacking in Zephaniah.23 Both of these claims are problematic. His presentation assumes connections to Exodus traditions are present in Joel, but documenting these links to the Exodus traditions utilizes some of the very same methods he denies are sustainable in editing the Book of the Twelve.24 Beck is not clear why he thinks Zion tradi23 BECK,

Der ‘Tag YHWHs’ im Dodekapropheton, 117. Links between Joel and Exodus are most thoroughly and creatively demonstrated by Siegfried BERGLER, Joel als Schriftinterpret (Beiträge zur Erforschung des Alten Testaments und des antiken Judentums 16), Frankfurt, Peter Lang, 1988. Bergler, however, makes extensive use of allusion via combinations of isolated words to make the case. Beck ignores the fact that the same kinds of techniques used by Bergler appear in discussion of redaction in the Book of the Twelve. 24



tion is lacking in Zephaniah. Given the fact that much of Zeph 3,8–20 explicitly addresses Lady Zion, one can only assume he has some predetermined idea of what constitutes Zion tradition which could account for his assertion. Further, his own evidence for dating the combination of independent writings in the Greek period in order to form the Twelve is not as convincingly presented as his literary critical arguments. In addition to Ben Zvi, Beck raises the issue of the random growth of writings as something that could account for (at least some of) the editorial additions. He notes that additions containing similar themes do not automatically require common redactional work. Such cautions are well heard by those working in this area. However, part of this critique, too, is limited by one’s predisposition. The nature of the connections is such that, if one determines in advance to approach the issue skeptically, one will likely not be convinced that such connections are present. In many cases, Beck argues against connections which others have seen, but argues for connections between texts which would strike others as questionable. In the end, such a skeptical approach does not adequately account for the nature of the ancient traditions that the Twelve was counted as a single book.25 A third scholar expressing caution about the task is David Petersen. Petersen raises questions regarding how one’s terminology connotes something about the presumed model of compilation and its significance for interpretation.26 Petersen makes an important contribution regarding terminology. To call the corpus “The Book of the Twelve” conveys something to those who approach this question for the first time. Undoubtedly, modern readers bring presuppositions regarding what constitutes a book. The Book of the Twelve is certainly not a modern novel; nor is it merely a catalog of prophetic sayings arranged with a single, consistent organizing principle. Distinctions must be made between what modern readers associate with the word “book” and the unifying functions found within the Book of the Twelve. Petersen recognizes a complex history of the Twelve, and he thinks that the best way to ap25 See

NOGALSKI, Literary Precursors, 2–3. L. PETERSEN, “A Book of the Twelve?” in Reading and Hearing the Book of the Twelve, 3–10. 26 David



proach the unifying elements is from the perspective of theme. He believes that “the day of YHWH” is an appropriate theme for speaking about the dominant literary features of the Twelve because of the frequency of the phrase and because the time span covered by the Twelve is larger than any other prophetic corpus. Petersen is less clear about how one speaks of theme without dealing with how the theme “unfolds” or with why some texts exhibit this theme while others do not. Other terms have also been used in attempting to avoid the problem of the presumptions associated with “book,” but these terms also connote some semblance of commonality as well as a lack of continuity (be it literary, chronological, or theological). Some of these terms include collection, corpus, compendium, or “thematized anthology” (Petersen’s own term). Terminology requires careful delineation by anyone working in this area. Petersen is correct. It is important to keep the character of the Book of the Twelve in mind when speaking about the whole. The terms “anthology” and “compendium” can be used effectively to accent certain elements of the corpus in order to describe parts of its character, but these terms do not go far enough to convey the sense of intentionality that is implicit in the chronological arrangement, or evidenced between some of the editorial links which show a cognizant exploration of an issue or theme across multiple writings. This issue will be discussed in more detail below by reflecting on the nature of redaction. The Nature of the Redactional Work in the Book of the Twelve Questions have also been asked by sympathetic proponents of the Book of the Twelve as a redactional entity about how to characterize the redactional work transcending the individual writings in the Twelve, leading to new models for the growth of the corpus and a need to evaluate the nature of the redactional processes themselves. In contrast to the skepticism of Beck, Wöhrle seeks a more comprehensive model that can account for the development of all the writings in the Book of the Twelve. Wöhrle is correct when he argues that the convincing power of such a redactional model would be increased dramatically if it could account for the growth of each



writing.27 However, such a comprehensive model has not yet been put forward. Wöhrle adds constructively to the conversation, but his own 467 page analysis, in the end, only offers a tentative theory for some of the twelve writings. He does not have room in this volume to analyze Hosea, Obadiah, Nahum, Habakkuk, Malachi, or Zechariah 9–14 (though he does give some hints as to where he thinks these works may fit). He only studies Joel, Amos, Micah, Zephaniah, Haggai, and Zechariah 1–8 in detail. Wöhrle raises at least two issues with his model that require comment: Joel’s development and the role of Hosea. Wöhrle contends that much of the previous scholarly work has either assumed the essential unity of Joel (Nogalski,28 Schart29) or approached Joel as a writing which has grown over time (Bosshard-Nepustil30) without laying the groundwork carefully. Wöhrle sides with the latter group, though he thinks both groups have approached the growth of the book through a faulty starting point. Nevertheless, Wöhrle concludes that Joel’s involvement with other writings of the Book of the Twelve requires that one account for the growth of Joel and the growing multivolume corpus simultaneously.31 This assertion raises an important question affecting several of the writings, not only Joel. Specifically, how does one account for the shape of the writings? Wöhrle points to very few substantive tensions not already noted in previous scholarship. The differences in the models hinge upon when and how the independent units were combined. Specifically, how one answers two questions dictates how one explains Joel’s compositional history: first, did the bulk of 27 WÖHRLE, Die frühen Sammlungen, 24–27, 466–467. See also his extensive review of recent redactional studies, 12–24. WÖHRLE has recently released his second volume that extends his model into the remaining sections of the Twelve, though still does not deal with Hosea. Time does not allow incorporation of the second volume, but neither does it change the points raised in this discussion to any great degree. See WÖHRLE, Der Abschluss des Zwölfprophetenbuches: Buchübergreifende Redaktkionsprozesse in den späten Sammlungen (BZAW 389; Berlin: De Gruyter, 2008). 28 WÖHRLE, Die frühen Sammlungen, 12–14. 29 WÖHRLE, Die frühen Sammlungen, 16–18. 30 WÖHRLE, Die frühen Sammlungen, 15–17. 31 WÖHRLE, Die frühen Sammlungen, 436–453.



Joel 1– 2 exist independently, or did an early version of these chapters already serve as the bridge to other writings in what would come to be the Book of the Twelve? Second, how does one explain the relationship of Joel 1–2 to Joel 3–4? Wöhrle’s model for Joel suggests six distinct layers of redactional activity, though the first two are more extensive than the others.32 According to his analysis, Wöhrle finds numerous connections in the foundational layer to the subsequent writings in the exilic version of the Book of Four (Amos, Micah, Zephaniah), but not to the preceding writing, Hosea.33 The lack of connections from Joel to Hosea, according to Wöhrle, leads to another important issue in his model, one where Wöhrle stands alone: the role of Hosea. Wöhrle argues that the incorporation of this early corpus of Joel literally replaced Hosea as the opening book of the multivolume corpus.34 For at least two reasons, this suggestion appears to be even less convincing than his diachronic hypothesis. First, the physical changes this theory would require of the scroll would be hard to envision. Of course, ancient editors could and did omit material as even a cursory comparison of Chronicles and Kings would confirm. However, such a model of the redaction of the Book of the Twelve seems highly implausible. It would require that the redactors take a collection which begins with Hosea in the first position, then eliminate Hosea in favor of Joel, only to replace Hosea back in the first position at a later point. One would have to account for someone maintaining Hosea as an independent scroll for an indeterminate period before it was reintroduced into the multivolume corpus. One would also have to explain why someone deemed Hosea inadequate for inclusion yet a subsequent version returned Hosea to its place of prominence as the first writing of the corpus. A second reason for rejecting the inclusion/exclusion/reinclusion hypothesis for Hosea would be the recognition of Hosea’s continuing role in the function of the growing corpus. The mention of several illustrations will have to suffice in this venue. To begin, Wöhrle himself suggests that the superscription of Joel 32 See

his summary in WÖHRLE, Die frühen Sammlungen, 428–435. Die frühen Sammlungen, 436–449. 34 WÖHRLE, Die frühen Sammlungen, 450–453. 33 WÖHRLE,



imitates the “word pattern” of the book of the four.35 However, the superscription of Joel is closer to Hosea than to Amos. It would seem odd to imitate the superscription (word of YHWH versus words of Amos), but to drop the writing. Additionally, the communal call to repentance comprising the end of Hosea and the beginning of Joel is a powerful parallel that makes better sense when both are present than it would if Joel was the only exemplar. Relatedly, much of the material within Hosea 4– 13 contains accusations against the people of YHWH (primarily Israel, but with periodic [often redactional] applications to Judah as well), whereas Joel contains no accusatory material to explain the guilt of the people. Finally, several intratextual links exist between Hosea and Joel, and other writings, which Wöhrle either ignores or too quickly dismisses. In addition to the presumption of guilt in Joel (see Schart,36 and Nogalski37) and the parallel calls to repentance in Hosea 14 and Joel 1–2, connections have been noted between Joel and Hosea 2 (Nogalski38 and Braaten39), and the involvement of the quote of Exod 34,6–7 in the Book of the Twelve with the themes of the names of the children of Hosea 1 (Van Leeuwen40). It seems unwise to ignore these connections in a model of the growth of the Book of the Twelve. Nevertheless, Wöhrle’s discussion does highlight an important distinctive concerning the way that Joel functions in the Book of the Twelve hermeneutically and theologically. Joel dramatically shifts the focus of the multivolume corpus at the time it is incorporated. Specifically, whereas Hosea and Amos periodically apply the message of those writings to Judah, that application remains a 35 WÖHRLE,

Die frühen Sammlungen, 38–39. Die Entstehung, 266–267. 37 NOGALSKI, Redactional Processes, 15–18. 38 NOGALSKI, “Recurring Themes in the Book of the Twelve,” 128– 36 SCHART,


Laurie J. BRAATEN, “God Sows: Hosea’s Land Theme in the Book of the Twelve,” in Thematic Threads, 108–111. 40 Raymond C. VAN LEEUWEN, “Scribal Wisdom and Theodicy in the Book of the Twelve,” In Search of Wisdom: Essays in Memory of John G. Gammie (Leo G. PERDUE, Bernard SCOTT, and William WISEMAN, ed.), Louisville, Westminster John Knox, 1993, 31–49 (especially 34–36). 39



rather sparse portion of the message—until Joel is incorporated. At that point, reference to the priests, Zion, and the temple highlight Judah and Jerusalem. To be sure, Joel assumes the guilt of Israel has been applied to Judah, but Joel explicates the implications for Judah in a way that Hosea and Amos do not. Moreover, the paradigm of “history” that unfolds in Joel has a much broader scope, especially with respect to the future of Zion, than does Hosea and Amos.41 Wöhrle’s treatment underscores the extent to which the model one develops, affects how one deals with the texts. There is a difference in quantity and character of the editorial connections in the editorial work on the various writings in the Book of the Twelve. The present form of some writings appears to have been compiled using pre-existing material with awareness of their context in the Book of the Twelve. This awareness affects the sources chosen, their combination, and the links connecting them. Other writings appear to be presumed (like Hosea, contra Wöhrle), yet they have had little in the way of editorial additions in later stages of the growth of the Twelve. Generally, the six dated writings have had redactional glosses added, short insertions into existing literary texts which function as invitations to compare one text with another within the Book of the Twelve.42 They have not, however, been radically reshaped when other writings were added to the multivolume corpus. More importantly, the six writings not included in the two pre-existing corpora are still the focus of debates regarding the extent of material incorporated into individual writings for the Book of the Twelve. The role played by each writing’s location in the Book of the Twelve needs more attention as a contributing factor to its final form. Nogalski argues Joel and Obadiah essentially owe their shape to their location in the Book of the Twelve. For Joel, Nogalski, following Bergler, suggests that three compositional (source) blocks were brought together by a redactor who shaped and supplemented the materials in light of the existing literary context of Hosea on 41 James D. NOGALSKI, “Joel as Literary Anchor in the Book of the Twelve,” in Reading and Hearing the Book of the Twelve, 91–109. 42 NOGALSKI, Literary Precursors, 276–277.



one side and Amos on the other.43 These sources included a composite call to repentance (Joel 1–2*), a description of the enemy attack on the day of YHWH (2,1–10), and an eschatological call to judgment on the nations (Joel 4*). He intimates that Joel 3 was added with a different focus.44 Schart agrees that Joel owes its form to its location in the Book of the Twelve. In his seminal work, Schart only suggests a few redactional supplements (4,4–8 and 1,2– 3) for Joel, yet he demonstrates connections to both Hosea and Amos (as well as Zephaniah and Obadiah).45 By contrast, Bosshard-Nepustil46 and Beck47 see Joel as a composition which grows diachronically with the Book of the Twelve. For Obadiah, Nogalski also assumes three source blocks (a composite text in 1–9; 10–14+15b; and 15a, 16–21*) have been adapted structurally with particular attention to Amos 9.48 Others see the tensions in Obadiah as signs of more sources, with Weimar finding six49 and Wehrle50 finding as many as seven redactional layers in this short 21 verse writing. Nahum and Habakkuk probably existed prior to their incorporation into the Book of the Twelve, but were significantly expanded with pre-existing hymnic material (along with other redactional comments) when they were incorporated into the Book of the Twelve. Broad agreement exists that Nahum 1 and Habakkuk 3 were added to their respective writing to adapt these two writings for the multivolume collection. In this case, major “redactional”


Redactional Processes, 1–6. Redactional Processes, 27–28 (footnote 74). 45 SCHART, Die Entstehung, 261–282 (see especially, 278). 46 Bosshard-Nepustil argues for four stages. See Erich BOSSHARDNEPUSTIL, Rezeptionen von Jesaia 1–39 im Zwölfprophetenbuch: Untersuchungen zur literarischen Verbindugn von prophetenbüchern in babylonischer und persischer Zeit (OBO 154), Freiburg, Switzerland, Universitätsverlag, 1997, 277–283. 47 BECK, Der ‘Tag YHWHs’ im Dodekapropheton, 142–151, 178–182. 48 See NOGALSKI, Redactional Processes, 89–92, 74–78. 49 Peter WEIMAR, “Obadja. Eine redaktionskritische Analyse,” BN 27, 1985, 35–99. 50 Josef WEHRLE, Prophetie und Textanalyse der Komposition Obadja 1–21. Interpretiert auf der Basis textlinguistischer und semiotischer Konzeptionen (Arbeiten zu Text und Sprache im Alten Testament 28; St. Ottilien: EOS, 1987). 44 NOGALSKI,



material was added to these writings, but it was not all composed originally for the Book of the Twelve. Jonah creates difficulty. On the one hand, its genre characteristics make it stand out from other writings in the Book of the Twelve, and its theological agenda (the willingness of YHWH to show compassion to the nations) are in many respects antithetical to those of Joel and many of the writings of the Book of the Twelve. On the other hand, recent studies have also demonstrated thematic or verbal links to other parts of the Twelve: thematic ties to Zech 8,20–23 and Mal 1,10–14; intratextual relationships with a series of texts playing off of Exod 34,6–7; the anti-prophetic stance in Mic 4,2–4 and Zech 13,1–6; and the citation of Jonah 2 in conjunction with Mic 7,19.51 To what extent do these connections affect our understanding of Jonah’s role in the Book of the Twelve? Malachi remains perhaps the least well defined, both in terms of the point of incorporation relative to Zechariah 9–14 and in terms of its function in the Book of the Twelve. A strong consensus exists that its opening and concluding sections relate to other themes in the Book of the Twelve, but no consensus has been reached regarding the form of the book prior to its inclusion in the Book of the Twelve,52 or regarding the relative point when Malachi was added to the Book of the Twelve (before, with, or after Zechariah 9–14).53 For example, see Martin ROTH, Israel und die Völker: Eine Untersuchung zu den Büchern Joel, Jonah, Micha und Nahum (FRLANT 210), Göttingen, Vandenhoeck und Ruprecht, 2005, 167–171. 52 Some of the same problems relative to the use of source material in the composition process in Joel come into play with Malachi as well. Thus, Bosshard and Kratz argue for a diachronic and developmental model consisting of several layers of editorial activity, while Schart sees the writing as a more unified compilation. See Erich BOSSHARD and Reinhard Gregor KRATZ, “Maleachi im Zwölfprophetenbuch,” Biblische Notizen 52, 1990, 27– 46; and SCHART, Entstehung, 293–295. 53 See, for example, SCHART, Entstehung, 291–303 and NOGALSKI, Redactional Processes, 210–212, who treat Malachi as a nearly completed composition. Schart, however, thinks it is added after Zechariah 9–14, while Nogalski believes it comes into the corpus prior to Zechariah 9–14. Bosshard sees the inclusion of the core of Malachi prior to Zechariah 9– 14 (BOSSHARD-NEPUSTIL, Rezeptionen von Jesaia 1–39, 426–428). See also Odil Hannes STECK, Der Abschluß der Prophetie im Alten Testament. Ein 51



One critique that has been leveled against the task of determining the outlines of the formation process is that the practitioners of this task have approached the question with a narrow focus. This charge of a narrow focus has been or could be leveled at most of those who have written on the formation of the Book of the Twelve: De Vries,54 Beck,55 Nogalski,56 Schart,57 BosshardNepustil,58 Schwesig,59 Wöhrle.60 Nogalski has been critiqued for focusing too heavily upon the “seams” of the writings (i.e., the opening and concluding verses) in order to evaluate the presence of catchwords as a redactional linking device. Schart has been critiqued for working from Amos outward, and trying to associate too much of Amos’ redactional formation with redactional work on other writings in the Twelve. Bosshard-Nepustil has been critiqued for relying too heavily upon parallels between the Twelve and Isaiah. De Vries, writing on the nebi’im as a group, focuses his technical investigation exclusively on the formulaic elements used to signal statements about the future. Beck and Schwesig focus upon the Day of YHWH, as they reconVersuch zur Frage der Vorgeschichte des Kanons (Biblisch Theologische Studien 17), Neukirchener Verlag, 1991, 33–34, 42–55. Steck sees the early layers of Malachi (similar to those described by Bosshard and Kratz [see previous note]) as a Fortschreibung to Zechariah 8, prior to the inclusion of Zechariah 9–14, while he believes other sections of Malachi were added with major sections of Zechariah 9–14. 54 Simon J. DE VRIES, From Old Revelation to New: A Tradition-Historical and Redaction-Critical Study of Temporal Traditions in Prophetic Prediction, Grand Rapids, Eerdmans, 1995. For a summary of his conclusions relative to the Book of the Twelve, see also Simon DE VRIES, “Futurism in the Pre-exilic Minor Prophets Compared with that of the Postexilic Minor Prophets,” in Thematic Threads in the Book of the Twelve (Paul L. REDDITT and Aaron SCHART, ed.; BZAW 325), Berlin, De Gruyter, 2003, 252–272. 55 Martin BECK, Der ‘Tag YHWHs’ im Dodekapropheton. 56 NOGALSKI, Literary Precursors; see also Redactional Processes in the Book of the Twelve. 57 SCHART, Die Entstehung. 58 BOSSHARD-NEPUSTIL, Rezeptionen von Jesaia 1–39 im Zwölfprophetenbuch. 59 Paul-Gerhard SCHWESIG, Die Rolle der Tag-JHWHs-Dictungen im Dodekapropheton (BZAW 366), Berlin, De Gruyter, 2006. 60 WÖHRLE, Die frühen Sammlungen des Zwölfprophetenbuches.



struct it, with Beck basing his study almost exclusively on the presence of the phrase, while Schwesig focuses upon those passages he terms day of YHWH poems. Wöhrle focuses on what he considers the “early” collections of the Twelve. What gets lost in these critiques is the degree of overlap which has resulted in these investigations, providing a significant number of checks and balances, as well as a confusing array of combinations. To be sure there is more in common between some of these presentations than others, but in many ways, some clarity has begun to emerge as a result of the multiple starting points. This brief overview thus shows that much work remains to be done, especially for the six undated writings. At least four prominent points of disagreement have not been resolved by the various redactional studies. Without additional investigations and proposals, these four points in particular represent areas where disagreements appear to have stalled progress toward reconstructing the development of the corpus and describing the group(s) responsible for the development. First, most studies agree that Joel plays a pivotal role in the Book of the Twelve, but there is genuine disagreement regarding its unity and the point of its incorporation into the Book of the Twelve. Is the character of Joel that of a composite writing created in relatively short order or a writing that reaches its current shape over an extended period? Second, there is no clear agreement as to whether Malachi was joined to Zechariah 1–8 before Zechariah 9–14 was added, or whether Malachi was added to the collection after Zechariah 9–14 was added to Zechariah 1–8. Third, the nature of the Day of YHWH in the Book of the Twelve remains a subject of debate. Finally, the role that a writing’s Sitz im Buch plays in its formation is too often left unexplored. It is this last issue to which I would like to devote the remainder of this essay.

2. FORM AND FUNCTION OF THE REDACTIONAL WORK The Role of a Writing’s Sitz im Buch Can Affect Its Form What role does a writing’s Sitz im Buch play in its formation? More attention needs to be given regarding how the inclusion of source blocks in one writing for a particular function in the Book of the Twelve can create literary tensions with another portion of the Twelve. The process of determining the growth of the Book of the Twelve becomes far more complicated when source material re-



flecting one historical setting is used in the Book of the Twelve for a literary purpose different from that originally intended. Consider the case of Nahum and Habakkuk. It seems clear that somehow Nahum and Habakkuk are intended to function in the Book of the Twelve as prophetic reflections upon YHWH’s role in the downfall of Assyria and the rise of Babylon. This function makes sense of the placement of Nahum and Habakkuk between Micah and Zephaniah, and several scholars have argued they are interconnected editorially.61 In this sense, these two writings seem to presuppose the chronological frame created by the interconnected superscriptions of Hosea-AmosMicah-Zephaniah. However, as some have noted, these two writings also create certain tensions within the content of the writings in sequence.62 For example, Nineveh/Assyria appears to be swept aside at the end of Nahum while Habakkuk anticipates Babylon’s imminent rise. However, the Assyria oracle in Zeph 2,13–14 returns to anticipating Assyria’s downfall. Moreover, if the analyses are correct, the biggest portions of Nahum and Habakkuk that were added to the existing form of those writings are not the portions which deal explicitly with Assyria in Nahum or with Babylon in Habakkuk. Rather, the largest editorial accretions in Nahum and Habakkuk are the theophanic hymns in Nah 1,2–8 and Habakkuk 3,1–19, neither of which mention the respective enemy by name. The theophanic hymn of Nah 1,2–8 mentions enemies (1,2) and 61 Childs already makes this connection. See Brevard S. CHILDS, Introduction to the Old Testament as Scripture, Philadelphia, Fortress, 1979, 454. See also the very different presentations of SCHART, Entstehung, 246–251 and BOSSHARD-NEPUSTIL, Rezeptionen von Jesaia 1–39 im Zwölfprophetenbuch, 393–397, who nevertheless see Nahum 1 and Habakkuk 3 in close association with one another. 62 This use of source material also affects how one understands the unnamed city in Zeph 3,1–8. Originally, Zeph 3,1–8 almost certainly reflects an oracle delivered against Jerusalem, but some want to interpret its function in Zephaniah as a continuation of the anti-Assyrian oracle in 2,13–15 (such as WÖHRLE, Die frühen Sammlungen, 219–220, 226–227, who also notes the Syriac version does the same thing by referring in 3,1 to the “city of Jonah” rather than just “the city”). Others argue that the original addressee, Jerusalem, makes better sense as an ironic ending to the collection of oracles against foreign nations in Zephaniah (see NOGALSKI, Literary Precursors, 175–178).



adversaries (1,8), not Assyria. Habakkuk 3 mentions the “head of the wicked house” (3,13) and “the people who attack us” (3,16), but it does not mention Babylon or the Chaldeans (see Hab 1,6). In both cases, however, good reasons exist for assuming that those shaping the writings intended these enemy references to be interpreted as Assyria in Nahum and as Babylon in Habakkuk. It must, however, be noted that the redactors did not feel compelled to add the specific reference to the hymns when they were incorporated into the respective writings. This anomaly has implications for interpreting both the theological agenda of the editors of this phase and for extracting data regarding the identity of the tradents. First, consider the implications of the use of hymns (and other cultic material) as a major source for expanding the growing corpus. There is wide agreement that this cultic material is not typical for the role of a prophet in ancient society. And yet, such cultic genres are hardly unique in the Book of the Twelve. The theophanic hymns in Nahum 1* and Habakkuk 3* play an integral role in the expansion of those writings. The thanksgiving hymn of Jonah 2 offers another case in point. Whether or not one agrees that the hymn was added for the incorporation of Jonah into the Book of the Twelve, the thanksgiving hymn in general and this one in particular have undeniable cultic connections. Seen in this light, the communal call to repentance at the core of Joel 1–2* also looms large, as do the cultic concerns underlying confrontation with the priests and people in Malachi. Additionally, the concerns of several of the night visions in Zechariah 1–6 reflect as much or more concern for the priestly role of Joshua as the focus at the end of the Haggai does for the political role of Zerubbabel. I am certainly not suggesting that this cultic connection comes from a single hand, or is all cut from a single cloth. I do suspect, however, it has not been given adequate attention in discussions of the prophetic tradents. Somehow, concern for the cult—its proper execution, its personnel, and its modes of expression—needs to be taken seriously in developing a model to explain the growth of the corpus. Second, one must consider carefully the difficulty of delineating a consistent and coherent theological agenda on the part of the editors who are shaping these writings when so much of the material they incorporate draws from sources they themselves did not compose. This dichotomy likely plays a major role in understanding



why so much disagreement continues around the six debated writings, all of which have significant relationships to cultic materials. This use of cultic source blocks both says something about and complicates the task of understanding the intentions of the editors/compilers of the Book of the Twelve. This revelation and complication are two sides of the same coin. On one side of the coin, the evidence suggests editors had access to cultic texts, access which results in the use of a wide array of forms associated with the cult and which lasts some considerable time. This access could perhaps be understood as a tradition out of which the various author/editors operate, or (and in my opinion more likely) it could be understood as evidence that the editorial use of this material was possible because the compilers had access to physical sources with which they could work. What, then, would be the nature of these sources and how do they relate to the final form of the writing? This question leads to the other side of the coin. How does one deduce editorial intention when a prophetic writing incorporates pre-existing source material with a cultic background? Some observations are in order concerning the ways in which tensions are created and ignored in this process. Something in the source block draws the editor’s eye, but tensions with the broader context are not necessarily eliminated. Nah 1,2–8 and the incorporation of Nahum into the Book of the Twelve offers an illustration. The focus of this theophany (prior to its attachment to Nahum) accentuates the role of YHWH in punishing the wicked. One can safely assume this motif manifests an important theological affirmation for the editors. The fact that the same topic plays a prominent role in Habakkuk 3 further solidifies this assumption. However, given the broad chronological outline of the Book of the Twelve, one cannot deduce from this theology that the editors were interested in replacing the focus on Assyria with a focus on the nations (contra Roth, for example).63 Rather, it works the other way around. In order to affirm that YHWH is the one responsible for the downfall of Assyria, the theophanic hymn accentuates YHWH’s role in the events that follow. For this reason, the material focusing upon Assyria’s downfall (the bulk of Nah 2–3) receives a theological emphasis in 63 ROTH,

Israel und die Völker, 289–290.



the form of an introductory theophany focusing upon the power of YHWH. Without the theophanic hymn (Nah 1,2–8), and its editorial transition (Nah 1,9–2,1*), the pre-existing collection of AntiAssyria oracles would be far less theologically oriented since YHWH would be almost completely absent from the early corpus.64 Nevertheless, when incorporating Nahum into the collection, the relatively brief Assyria oracle in Zeph 2,13–15 was not eliminated from Zephaniah, and this decision (or oversight) created a literary tension within the Book of the Twelve that did not exist earlier. The decision to include Nahum and Habakkuk recognized an important gap in the existing book of four. Specifically, the focus upon Israel and Judah in Hosea-Amos-Micah-Zephaniah does not deal with the issue of foreign occupation or the change from one foreign overlord to another. By contrast, the inclusion of Nahum-Habakkuk takes on this issue more in theological than historical or literary terms. By presenting Assyria’s demise as an act of YHWH’s justice and not just the downfall of a hated enemy, the expanded form of Nahum intones YHWH’s power. YHWH is far more than a territorial deity focused exclusively upon the fate of the small kingdom of Judah. Moreover, Nahum portrays YHWH not only as a God powerful enough to bring Assyrian hegemony to a close, but as a God who holds foreign powers to a standard of conduct which imposes limits upon their ability to terrorize other nations.65 In this respect, the point of combining the theophany hymn with the celebration of Assyria’s downfall comes very close to reiterating (or extending) major themes in Isaiah 10. In Isa 10,5–6, YHWH announces his intention to utilize Assyria as the means by which YHWH will punish Judah and Israel. YHWH famously refers to Assyria as “the rod of my anger” (10,5) as the tool he will use to punish Judah. Immediately thereafter, 64 Concerning the editing of Nahum for the Book of the Twelve presumed here, see James D. NOGALSKI, “The Redactional Shaping of Nahum 1 for the Book of the Twelve,” 193–202 in Among the Prophets: Language, Image and Structure in the Prophetic Writings (Philip R. DAVIES and David J. A. CLINES, ed.; JSOTSupp 144), Sheffield, JSOT Press, 1993. See also SCHART, Entstehung, 242–244, 246–251. 65 See NOGALSKI, Redactional Processes, 105–106, 149–150 181; SCHART, Entstehung, 248–249.



YHWH demonstrates YHWH’s own awareness that Assyria does not know its actions are controlled by YHWH: Ah, Assyria, the rod of my anger—the club in their hands is my fury! 6 Against a godless nation I send him, and against the people of my wrath I command him, to take spoil and seize plunder, and to tread them down like the mire of the streets. 7 But this is not what he intends, nor does he have this in mind; but it is in his heart to destroy, and to cut off nations not a few. 8 For he says: 5

“Are not my commanders all kings? 9 Is not Calno like Carchemish? Is not Hamath like Arpad? Is not Samaria like Damascus? 10 As my hand has reached to the kingdoms of the idols whose images were greater than those of Jerusalem and Samaria, 11 shall I not do to Jerusalem and her idols what I have done to Samaria and her images?” When the Lord has finished all his work on Mount Zion and on Jerusalem, he will punish the arrogant boasting of the king of Assyria and his haughty pride. (Isa 10,5–12, NRSV) 12

Immediately after announcing the use of Assyria as a tool to punish Judah, this passage makes two rhetorical points in 10,6–12 which both bear on the role of Assyria and Babylon in the Book of the Twelve. First, not only is Assyria unaware of YHWH’s plans, but YHWH knows Assyria’s king has his own intentions, to expand Assyria’s power (10,7). Second, once YHWH’s use of Assyria to punish Judah is complete, YHWH will punish the king of Assyria (10,12). In this respect, Isaiah 10 “anticipates” Assyrian hegemony as a product of Assyrian arrogance and greed which YHWH allows for a time. A third point connecting Isaiah 10 and Nahum appears in the speech of the Assyrian king’s litany of rhetorical questions designed to accentuate Assyrian power over Judah (Isa 10,8–11). These rhetorical questions stand out as an ironic parallel in light of the confrontation of Assyria with a similar rhetorical question introducing Nah 3,8–11, in the mouth of YHWH, designed to accentuate the absurdity of Assyria’s claim of strength in comparison to the intentions of YHWH. YHWH asks Lady Nineveh whether she is better than Thebes—a river city deemed impenetrable because it used the river as a significant portion of its defenses. The rhetorical



question in Nah 3,8 functions to condemn Assyrian weakness rather than create an impression of Assyrian power. This ironic twist is accentuated further by the fact that Assyria was the power which defeated Thebes, but the context of Nah 3,8–11 assumes YHWH, not Assyrian might, was the ultimate source of Thebes’ downfall. The point here is not whether Nahum draws upon Isaiah, though one could make that case. The point is the way in which two very similar theological statements are made. Isaiah 10,5–12 makes its points through a speech of YHWH, and YHWH’s quote of the Assyrian king within that speech. By contrast, a virtually identical “story line” is created by Nahum’s Sitz im Buch and Nahum’s compositional form by utilizing source blocks rather than a redactor’s composition. Regarding the Sitz im Buch, the last two chapters of Micah (whose literary setting concludes in the time of Hezekiah if one takes Mic 1,1 as a clue to the reader), presents three lines of thought rhetorically speaking: 1) following extensive accusations against Judah, YHWH condemns Judah for following the paths of Omri and Ahab (6,16–7,6); 2) Lady Zion anticipates punishment from “an enemy” who gloats over her, but who will herself soon be destroyed (7,8–10); and 3) Jerusalem will eventually be delivered, but punishment of the enemy will only come following Jerusalem’s punishment (7,7.9.13). Thus, Micah 6–7 sets up a scenario quite similar to Isaiah 9–10 by anticipating Assyrian occupation as punishment from YHWH. However, the Book of the Twelve’s version of this story line does not present its view as foreshadowing this process from the end of the eighth century (as does Isa 10,5–12). Instead of merely eliciting a perspective from a time of Hezekiah (via Isaiah and Micah), the impression from reading the Book of the Twelve moves the reader along chronologically when Nahum focuses upon the end of the process—nearer Assyria’s punishment. The reader of Nahum assumes a later time period than the one found in Micah because the vivid focus on Assyria’s destruction in Nahum forces a shift in the reader’s perspective. The addition of Nahum and Habakkuk creates a more elongated unfolding of these elements in the Book of the Twelve. This elongation adds texture to the chronological frame created by the superscriptions of Hosea-Amos-Micah-Zephaniah. By continuing beyond the promise of the enemy’s destruction (Mic 7,8ff), the



reader of Nahum in the Twelve senses YHWH’s affirmation that Assyria will not go unpunished. Hence, as with Isa 10,5ff, YHWH plans to punish Assyria when he is finished using her. YHWH’s punishment of the wicked is affirmed via the incorporation of a theophanic hymn (Nah 1,2–8*) and the editorial transition (1,9–2,1 [Eng. 1,9–1,15]) onto the pre-existing pronouncement of Assyria’s destruction so much a part of the core material in Nahum 2–3. This affirmation of punishment of the wicked also becomes an experience of an impending promise for Judah/Lady Zion (Nah 2,1 [Eng. 1,15]; note also the similarity to Isa 52,7). This pronouncement does not, however, end the story line in the Book of the Twelve because, at the beginning of Habakkuk (1,2–4), the description of the problems of the people is remarkably similar to the accusations of Mic 7,1–6: violence, bloodshed, and the perversion of justice abound in Judean society. Habakkuk 1,2–4 and the foundational material in 1,12–17 present these accusations largely in the form of an individual complaint song (another cultic form). Because of Judah’s problems, Hab 1,5ff announces YHWH’s use of yet another enemy, Babylon, who will punish Judah. This description of the Chaldeans contains elements which present Babylon’s strength in markedly similar (but stronger) terms to the description of Assyria in Nahum.66 In the case of Habakkuk, the redactional expansions to the pre-existing material identify Babylon as the enemy (1,6) and equate the actions of the oppressor with the country YHWH will send.67 Thus, a second enemy nation is sent by YHWH (1,5–6) because, despite Assyria’s downfall, Judah has not turned toward YHWH.68 Nevertheless, the incorporation of the theophany and prayer in Habakkuk 3 serves as a promise from the prophet that after YHWH punishes Judah and Jerusalem (using Babylon), YHWH will destroy Babylon (the “peo66 See

NOGALSKI, Redactional Processes, 146–150. Based upon the redactional observations in NOGALSKI, Redactional Processes, 140–146, which delineates the wisdom-oriented material as source material and the Babylonian commentary as the redactional comment. 68 Note also the similarity of Hab 1,6 (“Behold, I am about to raise the Chaldeans . . . ”) to introduce the arrival of Babylon against Judah in the Book of the Twelve and Amos 6,14 (“Behold, I am about to raise a nation . . . ”) to introduce punishment for Israel by the Assyrians. 67



ple who attack us” in 3,16) and restoration will then be possible (3,16–19). This connection backwards in Habakkuk is strengthened by the affirmation in Hab 3,18 which parallels the prophet’s response in Mic 7,7: “I will wait for/exult in the God of my salvation.” When comparing Isaiah 10 with Micah, Nahum, and Habakkuk, the commonalities are clear. In Isaiah (10,1–2) and the Book of the Twelve (Mic 7,1–6; Hab 1,2–4), the society of Judah is described as one of bloodshed and the perversion of justice. In both contexts, YHWH reveals his decision to punish Judah with a foreign power. In both contexts, the foreign power has no knowledge of YHWH’s purpose. It has its own agenda and thinks it is operating from its own strength. In each case, however, the reader learns this strength is an illusion when compared to the power of YHWH. In the Book of the Twelve, this process unfolds not once, but twice. The prophet acknowledges a dual-stage punishment (first Judah, then the foreign power) which leads to the prospect of restoration (waiting for the God of salvation). In Nahum, Judah’s punishment is implicitly ongoing, while Nineveh’s demise is portrayed as imminent. In Habakkuk, Judah’s punishment is imminent but its sin is ongoing while the punishment of Babylon must wait until Judah’s punishment is over. Thus, the transitions from Micah to Nahum to Habakkuk exhibit considerable coherence which derives from a redactional agenda to portray a prophetic message using source blocks. This segment of texts presents a theological reflection upon the seventh century decline of Assyria and the rise of Babylon as the work of YHWH, even while confronting YHWH’s own people. Unlike the presentation of Assyria in Isaiah 10, this twofold process, which accounts for Assyrian and Babylonian control of Judah in the Book of the Twelve, largely relies on the use of preexisting material—arranged, amplified, and connected—to present this story line. Yet, the character of the pre-existing material has stronger connections to what has traditionally been perceived as cultic forms rather than prophetic forms: theophanic hymns in Nahum 1 and Habakkuk 3, and the underlying complaint of Habakkuk 1. The connections to the Book of the Twelve isolated in this process serve two functions: to heighten correlations to the immediate literary context and to the broader corpus. These cases have



largely been made in earlier discussions and, though there have been some questions raised, the assertions have also found considerable support. Catchwords between Micah 7 and Nahum 1 strengthen a sense of connectivity between these two texts, but they were not written by the same hand. The Habakkuk theophany’s use of Joel imagery echoes the sense of an impending judgment which will make the land infertile before YHWH delivers his people. The way that Nahum 1 connects with Joel is equally important. The citation of Exod 34,6–7 in Nahum 1 presents the flip-side of Joel’s citation of Joel 2,13. Joel 2,13 cites Exod 34,6, focusing upon YHWH’s ,, patience, and desire for repentance, while Nah 1,2b–3a (inserted into the acrostic poem) evokes Exod 34,7: YHWH’s promise to punish the wicked. In so doing, this citation functionally parallels the allusions to Exod 34,7 in Joel 4,21 (Eng. 3,21) noted by several scholars.69 Further, this separate invocation of Exod 34,6 and later 34,7 in Joel 2,13 and 4,21, corresponds to the allusions to Exod 34,6 in Mic 7,18–19 and to Exod 34,7 in Nah 1,2b-3a. Thematically, Joel’s use of Exod 34,6–7 deals both with YHWH’s patient, long suffering grace toward those who repent and the eventual punishment of the guilty. The transition from Micah to Nahum does the same when it incorporates both Exod 34,6 and 34,7. How then does one explain this phenomenon of the use of pre-existing cultic texts in redactional work shaping prophetic writings to reflect upon history? This is not an easy question, but there do seem to be two avenues from which to begin a conversation on this topic. First, one can ask what group(s) would have access to these texts associated with the cult? Second, one should also ask, whose interests are served in compiling this reflection upon history that exhibits theological admonitions, cultic forms, the framing of history, and a developing eschatological (and proto-apocalyptic) perspective? The remaining section of this essay will explore one possibility: identification of the Levites as a group associated with the collection of the Twelve in its latter stages.

69 See

VAN LEEUWEN, “Scribal Wisdom,” 41.



The Functions of the Levites in Chronicles, Ezra, and Nehemiah The books of Chronicles, Ezra, and Nehemiah represent Persian period narrative literature. As such, these books offer important insights into traditions of the past filtered through the lens of Persian Period reality, along with some level of historical revisionism. These writings do not represent a seamless narrative from the hand of a single author, but at some level they have been edited together by persons sharing a substantially similar perspective. At the very least, they are intended to be read in conjunction with one another, as evidenced by the fact that the last two verses of Chronicles (2 Chr 36,22–23) are the same as the opening verses of Ezra (1,1–3a), and that Ezra functions as a major character in both Ezra and Nehemiah.70 The role of the Levites, and especially their relationship to the Zadokite and the Aaronide priests, has been a subject of some discussion for understanding the Persian period world of Judah.71 This question is complex because of conflicting data, the sparsity of sources, divergent numbers, and the difficulty of sifting through the various genealogical presentations. However, for the purposes under discussion in this essay, the exact historical relationships of the passages to one another need not be resolved in their entirety in order to appreciate the breadth of cultic functions associated with the Levites. The variety of functions assigned to the Levites in these texts suggests this group (however the network of families was constituted at any given time) may well have been a pivotal factor in the production of texts for use in the temple.

70 Ezra and Nehemiah, of course, are much more integrally connected to one another than to Chronicles. For a summary of the similarities, see the discussion of the evidence (ancient and modern) for connecting Ezra and Nehemiah more closely, see Hugh G. M. WILLIAMSON, Ezra, Nehemiah (WBC 16), Dallas, Word, 1985, xxi-xxiii. 71 For example, see Antonius H.J. GUNNEWEG, Leviten und Priester: Hauptlinien der Traditionsbildung und Geschichte des israelitisch-jüdischen Kultpersonals, Göttingen, Vandenhoeck and Ruprecht, 1965; Menahem Haran, Temples and Temple-service in Ancient Israel: An Inquiry into the Character of Cult Phenomena and the Historical Setting of the Priestly School, Oxford, Clarendon, 1978, 99–111.



A survey of the functions attributed to the Levites across these books suggests they are the only group mentioned in these postexilic narrative texts who had the resources, access, and intellectual wherewithal to compile the writings of the Twelve which demonstrate so many connections to cultic source blocks. These functions include specific tasks in four broad areas that impinge upon their ability to compile literary collections: 1) access to the temple itself, 2) training in musical composition and performance, 3) access to the means of production for scrolls, and 4) training in scribal duties (both for writing and for teaching). 1) Physical proximity to the temple is beyond dispute for the Levites. The genealogical material in 1 Chr 9,18.26–27 mentions that Levites guarded the temple gates, closed them at night and opened them again in the morning. According to Chronicles, an aging David assigned Levites several roles in the temple after it is no longer necessary for the Levites to carry the ark of the covenant since it will have a new place inside the temple (1 Chr 23,26–32).72 These tasks included “care of the courts and chambers, the cleansing of all that is holy, and any work for the service of the house of God” (1 Chr 23,28, NRSV). In Chronicles, a number of texts also indicate the Levites guarded the temple gates themselves, while other texts mention the Levites and gatekeepers separately, but in close proximity to one another. Levites are named as gatekeepers in 1 Chr 26,17, which lists the number of Levites stationed around the temple gates each day: six in the east; four in the north; four in the south; four (2+2) at the storehouse; and four in the west (two + two at the colonnade). Levites also serve as guards of the treasuries where tithes and cultic contributions are stored (1 Chr 26,20). Relatedly, it is the Levites who are depicted as collecting temple taxes, offerings, and other contributions: 2 Chr 24,5–6 (for temple repair during the These narratives and others combine sources from different time periods, but in their final form represent the agenda of organizing temple personnel in the late postexilic period under the authority of ancient traditions. It is not always easy to determine which texts reflect the later realities and which an ideal order for the Chronicler. See WILLIAMSON, Ezra, Nehemiah, 18–19. 72



reforms of Joash); 35,7–9 (during the reforms of Josiah); Ezra 2,68–70; 8,29–30; 8,33 (note that the gifts are also counted by Levites). According to Nehemiah 13, the second temple had a storage room which was the responsibility of the Levites themselves. While Nehemiah was away, the room was given by the chief priest to Tobiah, his relative and Nehemiah’s nemesis. This move cost the Levites their ability to support themselves from the temple treasury according to 13,5.10. As a result, the Levites (along with the singers) had been forced to return “to their fields.” The functions attributed to the Levites in the programmatic list of duties assigned by David included temple cleansing and repair: 1 Chr 23,28; 2 Chr 29,4–19 (Hezekiah); Ezra 1,5 (rebuilding temple); 3,8–9 (oversight of temple workers). Relatedly, purification of the Levites was an important expectation and shows up in various ways. In Ezra 9,1, the people, priests, and Levites need purification because they have married foreign women, so in 10,5.15 the priests and Levites divorce their foreign wives. In Neh 9,38, various leaders including Levites, sign Nehemiah’s covenant of purity (10,9 lists the names of the heads of Levites who sign). Other texts focus on the general purity of the Levites (Neh 10,28; 12,30; 13,22). 2) Chronicles, Ezra, and Nehemiah collectively suggest a close (but admittedly complex) association of Levites with temple music, both in terms of performance and presumably also its composition. Levites are frequently associated with singing and playing instruments during cultic celebrations (1 Chr 15,16–24; 2 Chr 5,12; 7,6; 8,14–15; 29,25–26.30; 30,21–27; 34,12; 35,5.14–18; Ezra 3,10; Neh 12,8; 12,24–25; 12,27). Invocation, praise, and thanksgiving become expressions for leading temple singing (1 Chr 6,31.48; 9,33– 34; 16,4; 2 Chr 20,19; Neh 9,4–5). The guild of Asaph (1 Chr 15,16–24; 2 Chr 20,14; Ezra 2,40; 3,10; Neh 7,43–45) and the Korahites (2 Chr 20,19) appear with, or are counted among, the Levites. These groups are credited with significant collections of the Psalter.73 Numerous texts mention the Levites with temple 73 Two small collections of Asaph and Korahite psalms appear together in Pss 42–49 (Korahites) + 50 (Asaph) and 73–83 (Asaph) + 84–88 (Korahites) in books two and three of the Psalter.



singers (1 Chr 9,33; 15,16.27; 2 Chr 35,15; Ezra 2,70; 7,7.24; Neh 7,1.73; 10,28; 11,22; 12,47; 13,5.10). Many of these references point to divisions of the two secondary clerical functions, but some also associate this activity with the Levites themselves. 3) The probability that Levites have access to the means of production for scrolls can be deduced from several texts. Pasture lands are given to the Levites at the command of David, an act which implies they own lands used for cattle throughout the country (1 Chr 6,6474; 13,2). The Levites run into trouble with Jeroboam I when they refuse to help with sacrifice at the new northern kingdom sites and thus lose these fields because of their commitment to worshiping YHWH only (2 Chr 11,13–14). Clearly, this episode is recounted to bolster their claim to lands in the postexilic community. When they lose their place at the temple in Nehemiah, the Levites have “fields” to which they can return (Neh 13,10). The skinning of animals offered as sacrifice is a task attributed to the Levites during the monarchy (2 Chr 29,34; 35,11) and/or Persian period (assuming that part of the rationale for the Chronicler is to root the cultic practices of his day in the authority of David). Over time, this task may have been assigned to other temple servants, but presumably under the oversight of the Levites. At any rate, these skins would provide the Levites with ongoing control of the hides essential for producing scrolls. Thematically, several texts also intimate that the Levites had a vested interest in the fertility of the land because their livelihood depended upon it (1 Chr 9,31; 23,24.29; 2 Chr 31,4–5; 32,28; Ezra 7,13–18; Nah 5,1–11; 10,31–39). Some even use the same stock formulaic groupings which appear in fertility texts in the Book of the Twelve (vine/wine, grain, and oil). 4) Levites are also portrayed as scribes and teachers. Scribal groups and activities associated with Levites would reflect the tasks 74 The entire chapter points to a composite use of sources, including some clear relationships to Joshua 21, which also associates these pasture lands with the Levites (21, For a thorough discussion of the composite nature of the chapter and the interplay with Joshua, see Sara JAPHET, I & II Chronicles: A Commentary (OTL), Louisville: Westminster John Knox, 1993, 145–162 (especially 145–149). These lands around the city are not conceptualized merely for planting, but for tending cattle as well (see Josh 14,4).



they needed to fulfill. Scribes would have had to count and record the gifts brought to the temple. Scribes would be needed to write, or at least record in written form, the songs which the Levitical groups performed. These could have been recorded and stored in rooms in the temple to which the Levites had access. One finds references to certain Levites as scribes (1 Chr 24,6; 2 Chr 34,13), and exiled Levites play a central (though at times reluctant) role in the book of Ezra, whose chief character is the scribe of scribes (see the account of the recruitment of Levites in Ezra 8,15–20). The role of Levites as teachers plays prominently in Chronicles and Nehemiah. This teaching involves understanding the Torah and application to the history of YHWH’s people (2 Chr 17,8; 35,3; Neh 8,7–11). This task involved the study of Torah and other texts, which one may safely presume were available to them at the temple. The association of Levites with the study of Torah, and with the collection (and production) of psalm texts, at least significantly raises the possibility that the Levites also had access to prophetic scrolls.75 This prophetic connection is increased when it is recognized that Haggai and Zechariah traditions play a significant (albeit confusing) role in Ezra’s narrative of the early restoration period.76 One text in particular, Neh 9,32, stands out thematically in reference to the teaching of the Levites because of its correlation with the interests of the Book of the Twelve in general and the six debated writings in particular. In the context of Ezra’s teaching, with the help of Levites (see Neh 9,4–5), Ezra recounts Israel’s history with God, summarizing the time from Moses through the monarchy. In the end, Ezra accents God’s compassion in delaying the punishment of his people, a punishment that begins to unfold from the incursions of the Assyrians in the eighth century:

75 While problematically referring to the text as a “Levitic sermon,” Beuken’s observations concerning the similarity of Zech 1,3–6 and 2 Chr 30,6–9 are also instructive at this point because of the reflection on prophets and history in Zech 1,3–6. See BEUKEN, Haggai - Sacharja 1–8, 110–115. 76 Ezra 5,1; 6,14. See discussion of the narrative framework in WILLIAMSON, Ezra, Nehemiah, 73–74.



Now therefore, our God—the great and mighty and awesome God, keeping covenant and steadfast love—do not treat lightly all the hardship that has come upon us, upon our kings, our officials, our priests, our prophets, our ancestors, and all your people, since the time of the kings of Assyria until today (Neh 9,32, NRSV).

This verse, in its literary context, introduces a final challenge to the people of Nehemiah’s day by interpreting history from the eighth century to the late Persian period. In one sense, this verse could summarize portions of the book of Kings, but it also describes the main theological emphases of the Book of the Twelve: punishment of YHWH’s people begins with the time of the kings of Assyria (Hosea, Amos, Micah). God’s fidelity and compassion delay punishment (see the earlier discussion of the quotes of Exod 34,6–7 in the Book of the Twelve), but once the punishment begins (Nahum and Habakkuk), it continues into the postexilic period (Malachi). To be sure, the form of the story is presented differently in the Book of the Twelve, but the essential outlines assumed in Neh 9,32–37 and the collected writings of the Book of the Twelve are very similar in this respect. Several texts imply that the duties and fate of the Levites shifted over time, but not always in ways which improved their status.77 In this regard, early postexilic optimism regarding the temple may have periodically given way to a sense of marginalization in relating to the power structure of the temple. One of the recurring suggestions for the move from prophetic eschatology to apocalypticism has long involved the suspicion of this deep sense of alienation.78 Further, a group associated with Levites would certainly have had interest in the proper role of the cult in society. Themes of cultic purity were associated with Levites in the narrative texts. They would have been well schooled in the ancient traditions reSee, for example, the summary description of delineated views within P material in GUNNEWEG, Leviten und Priestern, 185–188. 78 See, for example, Paul L. REDDITT, “The Book of Joel and Peripheral Prophecy,” CBQ 48, 1986, 225–240. Redditt sees the marginalized group behind Joel as “separatist, exclusivistic, and nonmessianic” (238), but as prophetic rather than priestly in its orientation. 77



garding the fertility of the land which could have accounted for the ongoing interest in the state of the agricultural products used in cultic feasts. Relatedly, the “sons of Levi” appear to be the group in need of reforming and restoring who are mentioned in Malachi (2,4.8; 3,3; see also Zech 12,13).79 In sum, these observations concerning the functions performed by Levites in postexilic narrative literature offer a tantalizing web of connections between those functions and the resources (physical and intellectual) needed to produce the corpus of the latter stages of the Book of the Twelve. To be sure, this picture needs further development, refinement, and probably correction in places. However, no other group mentioned in these biblical texts comes close to having the variety of skills, assets, training, and networks necessary to pull together a prophetic collection combining the kinds of source material present in the latter stages of the Book of the Twelve. The cultic connection of the six debated writings merits more attention than it has heretofore received, both in terms of its literary function in the Book of the Twelve and the implications for the production of the corpus. The Levite-teachers portrayed in Ezra/Nehemiah represent a group that had interests in interpreting history from the eighth century. It seems plausible to postulate that the cultic circles in which the Levites moved inherited the Book of the Four/Deuteronomistic Corpus (which was probably transmitted in Palestine) after the rebuilding of the temple, and that this group had already begun to be influenced by those groups associated with those repatriated by the Persians who controlled the temple. The fate of this group changed over time, sometimes having a more privileged status, while at other times they were more marginalized within the power structure. This changing status could also help account for the more pessimistic attitudes for society as a whole in the latter portions of the Book of the Twelve.

Redditt argues that a non-Zadokite Levite is responsible for composing Malachi, but extending this suggestion to the group responsible for compiling the Book of the Twelve in its final form needs careful consideration. See Paul L. REDDITT, Haggai, Zechariah, Malachi (NCB), Grand Rapids, Eerdmans, 1995, 151–152. 79

IS THE TWELVE HYPOTHESIS LIKELY FROM AN ANCIENT READERS’ PERSPECTIVE?1 EHUD BEN ZVI UNIVERSITY OF ALBERTA 1. INTRODUCTION The “Twelve Hypothesis” (hereafter TH), as all grand hypotheses, does not refer to a single proposal, but to a particular range of ideas, proposals, methodological approaches and, of course, assumptions.2 In general terms, one may characterize or approximate the realm occupied by the TH through a series of related, and at times, converging statements such as: (a) during the Persian/Hellenistic period there was a “Book of the Twelve” (hereafter, the BT) not twelve prophetic books simply written (perhaps as 1 I dedicate this essay to my counterparts in this debate over the Twelve Hypothesis. Although we disagree on many matters, I greatly appreciate their work, which has contributed much to the field, and to the sharpening of my own thinking. 2 One of the best overviews of the research space I refer to as the TH, along with interactions with those who raise concerns about it, is in P. L. REDDIT, “The Formation of the Book of the Twelve: A Review of Research,” in: Thematic Threads in the Book of the Twelve (P. L. REDDITT and A. SCHART, ed.), Berlin/New York, Walter de Gruyter (BZAW 325) 2003, 1–26. Of course, the survey does not include works published in the most recent years, such as M. BECK, “Das Dodekapropheton als Anthologie,” ZAW 118, 2006, 558–581; the entire discussion in Interpretation 61/2, 2007, 115–97; and J. WÖHRLE, Die Fruhen Sammlungen des Zwolfprophetenbuches: Entstehung und Komposition, Berlin/New York, Walter de Gruyter (BZAW 360) 2006.




early as that time) in a single scroll; (b) books such as Amos, Micah, Hosea, Obadiah, and Jonah or any in the Twelve drew their meanings from the whole, namely the BT and were (or are to be) understood as “parts” of that “whole;” (c) the BT was read as a (more or less) self-contained unit and sequentially; (d) the BT stood on par with the books of Isaiah, Ezekiel, or Jeremiah, not as a loose collections of clearly separate prophetic books that reflected the general discourse of their time,3 and therefore, there were four prophetic books, not fifteen, in the repertoire of ancient Israel in the late Persian/early Hellenistic period; to put it in a different way, links between Micah and Jonah, for instance, must essentially and by definition be of a different quality (i.e., close inner book bonds that point at inner-book textual coherence) from those between Micah and Isaiah or Jeremiah (which are by definition external and therefore, may point only at general intertextuality among independent works); (e) the BT evolved through shared compositional/redactional processes that we (i.e., today’s scholars) are able to reconstruct, and which included at an earlier stage some form of a Book of the Four—the alternative is that precursors of separate prophetic books developed into the present books through compositional/redactional processes that cannot be reconstructed with any certainty, but that took place within a general intellectual discourse out of which emerged all the prophetic and most other (later biblical) books, and (f) the BT shows a coherent, literary structure that characterizes it as a book and which is a property of the text itself. A careful reading of the last statement shows that the TH occupies a somewhat loose space shared by what we tend to call, probably in a misleading manner, historical and literary approaches. Even as we all acknowledge that the borders between these two approaches are often porous, there is still much that separates be-

By “discourse” I refer to a vast realm that includes the ways of thinking, webs of images, texts, “common” knowledge and linguistic registers that shaped (a) which issues or set of issues were likely to come up in a community, and (b) the ways in which the community went about thinking about these issues when they arose and the range of possible responses and interactions within these responses. 3



tween, for instance, Nogalski’s and House’s approaches to the TH.4 Being a historian of the intellectual discourse of ancient Israel, my contribution here will focus on historical concerns and reconstructions and the relevant areas within the TH that relate to these concerns.5 To illustrate, the former is much more interested in matters of production, redaction, and historical contingency than the latter, but both still participate in this shared space. Cf., for instance, J. D. NOGALSKI, Literary Precursors to the Book of the Twelve, Berlin/New York, Walter de Gruyter (BZAW 217) 1993; and idem, Redactional Processes in the Book of the Twelve, Berlin/New York, Walter de Gruyter (BZAW 218) 1993) with P. R. HOUSE, The Unity of the Twelve, Sheffield, Sheffield Academic Press (JSOTSupS 77) 1990, and their respective methodological worlds. 5 There is no doubt that, from a literary perspective, it is possible to read the twelve prophetic books as one single book, which is exactly what those who work within the TH framework do, or for that matter, to read it within a linear, sequential mode of reading, which is what many among them do. The question that interests me as an historian is whether the literati in ancient Yehud, among which these books emerged in (more or less) their present form did so. For the argument supporting my claim to locate the development of the concept of prophetic book and most of the prophetic books in the Persian period and among Jerusalem-centered literati see, for instance, my “The Concept of Prophetic Books and Its Historical Setting,” in: The Production of Prophecy: Constructing Prophecy and Prophets in Yehud (D. V. EDELMAN and E. BEN ZVI, ed.) London, Equinox (BibleWorld) 2009, and my previous works on the books of Hosea, Jonah, Micah, Obadiah, and Zephaniah. I would like to stress that although I locate the development of the concept of prophetic book and most of the prophetic books in the Persian period and, accordingly, I do not deny the possibility that the text of some of these books (e.g., Zechariah, Ezekiel) continued to evolve after the collapse of the rule of Darius III, and did not reach the present masoretic form until the Hellenistic period (cf. the present debate about some texts within Genesis, even as most scholars agree that the Pentateuch emerged in a Persian context). In any event, the matter does not have a substantial bearing on the main arguments advanced in this contribution. Similarly, the question of whether some texts which are part and parcel of the prophetic books might have had some (textual or oral) prehistory in diasporic settings is not relevant to the present discussion. After all, the concept and genre of the prophetic book and the individual books we have are the product of Jerusalem-centered literati in Yehud and what4



Despite its relative recent surge in scholarly circles,6 the TH belongs to the realm of the general (or grand) hypotheses in our field.7 The most successful of these grand hypotheses shape a historical (general) narrative that explains the development of an important corpus of literature and, more importantly, is responsive to the longings of many in the guild at the time in which a particular grand hypothesis develops or reaches its peak in terms of acceptance. To put it simply, hypotheses that answer the questions that many scholars wish to have answered and that do so in a way consistent with what numerous scholars would like to imagine (e.g., clear lines of continuation between monarchic and post-monarchic Israel, basic historicity, basic ability to reconstruct textual histories and learn more about the past tend to surge in terms of acceptance among scholars).8 These grand hypotheses directly impact studies in their relevant area because of both their explanatory and generative power. ever sources these literati may have used were fully appropriated and resignified by them. 6 The popularity of the TH in scholarly circles surged since the 1990s. Of course, forms of this hypothesis existed since the 19th century (H. Ewald). For surveys of the history of scholarship in regards to either the unity or the formation of the Book of the Twelve, see B. A. JONES, The Formation of the Book of the Twelve: A Study in Text and Canon, Atlanta, Scholars Press (SBLDS 149), 1995), 13–40. See also P. L. REDDIT, “The Formation,” 2–3 and M. A. SWEENEY, The Twelve Prophets, Collegeville, Minn., Liturgical Press (Berit Olam), 2000, vol. 1, xix-xxvii. 7 Along with, for instance, the DH hypothesis, the J-E-D-P hypothesis. 8 Consistency with widely accepted ideological tendencies/evaluations within the scholarly guild, or a subgroup of it at any particular time contributes as well. For instance, the JEDP hypothesis at its inception and for a long time went hand in hand with certain evaluations of the “religious” history of Israel; Y. Kaufmann’s counter position reflected a counter evaluation. The emphasis on the (unconditional) promise to David in many DH studies, the focus on the actual words and message of the (much lionized) prophets, or the stress on their social concerns all are consistent with larger metanarratives shared by scholars and the societies in which they operated. The clear tendency within the TH—and much of the field in the last two decades—to construe the last textual stage as a culmination of centuries long, continuous theological thinking is certainly in accordance with present day trends.



They seize the imagination of many because they provide a comprehensive, yet somewhat flexible, explanatory frame that allows for integration of numerous observations, but above all because they create a research framework that generates new observations. They provide, as it were, a new interpretative lens with which to approach the material, and therefore, find new data. Of course, grand hypotheses not only lead to new research because they provide a new lens, but as they try to cover a large amount of, at least on the surface, disparate material they cannot but encounter numerous points of tension between the texts and the hypothesis.9 Thus to sustain the hypothesis, some troublesome texts have to be re-evaluated, explained or explained away— whether synchronically or diachronically—and, at times, proposals for fine-tuning aspects of the hypotheses are advanced. All these considerations lead to the proliferation of new research within the space occupied by the grand hypothesis. There is no doubt, in my mind, that the TH in its historically oriented version/s is one of these grand hypotheses. Certainly, it has generated much research aimed at dealing with objections, providing a new lens to look at the texts. It shaped a grand narrative that binds together monarchic, exilic, and Persian period Israel and their intellectual and literary worlds through a thread of continuous re-writing of an ongoing, shared text, and a chain of ongoing theological thinking in which generations build upon the work of past generations, across and despite the chasm created by historical disasters. Moreover, through carefully argued redactional-critical arguments, it provides the series of temporally sequential (reconstructed, but claimed to be historical) proto-texts that are necessary to support (or at least to provide a sense of verisimilitude) to the narrative. Those who follow this narrative cannot but be excited as they can feel and track theological steps and text in a transgenera-

This holds true for the DH Hypothesis, the Documentary Hypothesis, or the TH. As expected and as it should be, a significant amount of scholarship in the last ten years by colleagues supporting the TH was meant to provide responses to questions, concerns, and textual observations raised by scholars who did not accept it as it “erupted” into scholarly discourse. 9



tional chain linking Israel despite the vicissitudes of history, while at the same time reflecting on them. Although, one may say that most of the redactional critical approaches to prophetic literature attempt to provide such narratives, the TH shapes a far more comprehensive, interlinked, and sophisticated narrative than those that can be developed out of redactional critical studies of the individual twelve books and which were a common place in historical studies of the Hebrew Bible/Old Testament till recent times. Finally, as in the case of all of grand narratives, there is a sense among many of those who participate in the TH that the validity of their hypothesis is to a large extent a settled matter; that what remains to be done is not to prove it, but to further use and refine it.10 I understand why the TH is attractive to many. I am very glad that it was advanced and fully aware that it has provided, and will likely continue to provide an important impetus for new research. But as a historian of the intellectual discourse/s of the Persian period my focus is on the rereaderships of the relevant texts within Yehud. In my opinion, a historical reconstruction of the most likely rereadings within the Yehudite literati does not support the TH.

This position that the matter is basically settled is at times directly expressed, but often is indirectly expressed. If one entitles a book “Reading and Hearing the Book of the Twelve” or “Thematic Threads in the Book of the Twelve,” one communicates that the assumed starting point is the existence of such a book. The number of works that do so are a legion, partially because of the popularity of, and the excitement created by the TH. Of course, there is always an acknowledgment that not every one in the field will be, or can be, brought to accept the hypothesis. I have plenty of personal anecdotes that testify to that point. I want to stress at this point that unlike some other debates in our field, the one surrounding the TH has been conducted without any personal attacks or hyperbolic claims. I began to think on these matters when recently I was pointedly asked why I think that this is the case in the debate about the TH. To be sure, I am convinced that the personal qualities of many of the scholars involved and their friendly and respectfully interpersonal relations account for much, but I tend to think that there is more. This essay, however, is not about sociology of knowledge and the matter should be left for another occasion. 10



Moreover, much of what is valuable in terms of observations raised by the TH can be better explained, without recourse to the TH. To a large extent, the matter is one of methodology. I understand why scholars for whom the preferred approach is redactional-critical may end up supporting the TH.11 My point, however, is that if the starting-point for historical research is the question of whether it is more likely that the ancient rereaders of these texts in Yehud read them as (a) one (sequential) unit, the Book of the Twelve, which existed on par with the books of Isaiah, Jeremiah, and Ezekiel, or (b) as twelve separate books which along with the books of Isaiah, Jeremiah, and Ezekiel shaped a repertoire of fifteen prophetic books that existed within a larger repertoire of authoritative texts in the community, then the answer is (b). Of course, different methods create different sets of (relevant) data, and different data cannot but lead to divergent conclusions. It is my contention, however, that from a historical perspective aimed at reconstructing the likely discourse of Yehud, an approach that focuses on “(re)reading” is to be preferred over one grounded on considerations about successive redactions and individual authors/redactors/editors.12 This preference is not, and cannot, be a matter of turning the burden of proof around or overcoming perceived (history of research) biases in favor of a particular approach,13 but must be based on historical and historical methodology considerations. I began to dwell on these matters about ten 11 I want to stress that I do not dispute the possibility that scribes reproducing what they considered to be twelve separate prophetic books in one scroll may have been influenced by a text they had just written/copied into that scroll and that such influence might have led them (unintentionally or even intentionally) to slightly reshape here and there another text/verse in that scroll. This is, however, a far cry from the TH and in any case a process like that would have no substantial bearing in the argument I am advancing here. 12 This consideration is not dependent on the particular approach to reconstructing the likely readings of these texts in Yehud that I advance below. 13 Compare and contrast with A. SCHART, “Reconstructing the Redaction History of the Twelve Prophets: Problems and Methods,” in: Reading and Hearing the Book of the Twelve (J. D. NOGALSKI and M. A. SWEENEY, ed.), Atlanta, Society of Biblical Literature (SymS 15), 2000, 34–48 (36).



years ago;14 I hope that this contribution will advance the debate, and at the very least, will cause pause for thought before accepting the TH as an “agreed fact.” Given the centrality of methodology in this debate, I will first outline the basic approach I am following and the reasons for preferring it. Then, and in that light, I will turn to the arguments that point at a reading of fifteen (instead of four) prophetic books in Yehud and examine some of their implications.

2. WHY READINGS ARE CRUCIAL AND HOW TO APPROXIMATE THE WAY IN WHICH THE PRESENT PROPHETIC BOOKS WERE READ BY THEIR PRIMARY REREADERSHIPS IN YEHUD Although we are used to stating that books were authoritative in Yehud, it is not the books themselves that were authoritative. Rather it is the particular reading of a book that existed in the community that was considered authoritative. The written text served to elicit that reading in the community and as a material, symbolic presence of a reading.15 It is not the written text per se that reflects and shapes the ideological world of the community and serves to socialize it around particular norms and shared memories, but the reading/s of the text that are accepted within that community. Similarly, it is not whatever an author may have had or not in mind at a particular time that shapes and reflects the discourse of the reading community, but the intention that these readers assign to the voice speaking to them, as it were, through the book. This is a voice that they themselves construe on the basis of the text of the book and their world of knowledge, including the particular cultural and linguistic code that “competent readers” in the commu14 E. BEN ZVI, “Twelve Prophetic Books or ‘The Twelve.’ A few Preliminary Considerations,” in: Forming Prophetic Literature: Essays on Isaiah and the Twelve in Honor of John D. W. Watts (J. W. WATTS and P. R. HOUSE, ed.), Sheffield, JSOT Press (JSOTSup 235) 1996, 125̻156. 15 See my “One Size Does not Fit All. Notes on How Did the Chronicler Deal with the Authoritative Literature of Yehud,” paper presented at the 2008 SBL meeting held in Boston, and to be published in a full essay version in a collected essays volume.



nity are required to use to read these texts.16 In other words, it is their implied, not any historical or actual author that carried an authoritative voice within the community and, as such, both reflected and shaped its discourse. These considerations, by themselves, lead to a preference of historical reader oriented approaches over author/redactor approaches,17 for the purpose of reconstructing ancient communal discourses. But as they do so, they raise the very practical question of how can historians approximate a reconstruction of the ways in which ancient works were read by a particular historical group, and particularly when such a group did not leave explicit, stand-alone commentaries on the biblical books, as is the case with the primary communities within which and for which the prophetic books were composed and reached their present form (or forms)?18 16 I dealt with some of these issues in my “Would Ancient Readers of the Books of Hosea or Micah be “Competent” to Read the Book of Jeremiah?,” forthcoming in a collected essays volume. 17 Additional considerations favoring a reader oriented approach are advanced below. 18 Historians have a solid starting point when they have access to clear, explicit commentaries. For instance, Ibn Ezra’s, Calvin’s, or Luther’s respective commentaries on biblical books are good sources for reconstructing the readings of their respective communities, of those strongly influenced by them, and for reconstructing the character and message of the implied authors that each of these communities created. The pesharim from Qumran point to a mode of reading that was accepted by a particular group/groups. Josephus’ rewriting of the biblical narratives represents another way in which readings are manifested, though in this case, there are considerable questions of the nature and identity of the community of readers. Moreover, Josephus’ rereading of biblical accounts was not considered authoritative. External commentaries on the present compositional form of the prophetic books could develop only at times later than these forms. Textual variants that have no substantial bearing on the overall meaning of the text can and did exist among texts that represent the same “present compositional form,” as even a comparison between close texts such as MurXII and the MT of the twelve prophetic books show. It is worth noting that books may have more than one “present form,” as is the case with the MT and LXX versions of the book of Jeremiah. In fact, pluriformity was a common feature of authoritative texts at least in late Second Temple period, but probably earlier as well, and has important direct im-



In these cases, comparative approaches (i.e., checking attested readings in somewhat related communities of readers) may be helpful,19 but in the main there is no alternative but to rely on internal markers for approximating the actual range of readings of the primary reading community/ies of readers. The most important internal marker for this purpose is the concept of “intended reader,” or even better, “intended readership community.”20 Books presuppose a community of readers able to decode them, and a request for their active participation in the process of decoding. The prophetic books presuppose, for instance, a community of readers with a world of knowledge that includes, at the very least, an ability to read biblical Hebrew, including matters of semantic and pragmatic markers, syntax, literary style and genre awareness, macrostructural markers and a sense of what is familiar and what is being de-familiarized, among many others. Understanding the meaning of even seemingly simple literary motifs requires a well-internalized cultural poetics.21 In addition, readers in this community are supposed to identify with Israel, know of its deity, be familiar with spatial maps that include sacred centers (e.g., Jerusalem) as well as boundaries, be aware of social maps that involve social institutions, constructions of gender, as well as customs and (constructed) memories and soplications for the study of the modes of reading of these books. See below. 19 See below. 20 The basic approach advanced here concerning the intended community of readers of the book may be compared with the one used in D. KRAEMER, “The Intended Reader as a Key to Interpreting the Bavli,” Prooftexts 13 (1993), 125–40 for the study of the historical discourse of the Babylonian Talmud, and with that advanced in E. CONRAD, “Forming the Twelve and Forming Canon,” in: Thematic Threads in the Book of the Twelve (P. L. REDDITT and A. SCHART, ed.), 90–103. These approaches are, of course, related to general theories about reading advanced by U. Eco and S. Fish, among others. My point here concerns only the contribution that these approaches may have for historical reconstruction of ancient discourses. 21 See, for instance, D. SEEMAN, “ ‘Where Is Sarah Your Wife?’ Cultural Poetics of Gender and Nationhood in the Hebrew Bible,” HTR 91, 1998, 103–25; idem, “The Watcher at the Window: Cultural Poetics of a Biblical Motif,” Prooftexts 24, 2004, 1–50.



cially shared hopes and fears that shape a sense of self-identity. Of course, readers in this community must be aware of, and read, more books than the one on whose reading of it we happen to focus. In fact, these readers must be aware and read a certain repertoire of books. Moreover, they must have a ‘proper’ education, which cannot but involve the acquisition and valorization of the cultural capital, through which the readers are socialized. In general terms, one may say that since language does not work in a vacuum, but within a particular cultural context, the reception (and therefore, the production) of texts such as the prophetic books presupposes, at the very least, an intended readership community with sufficient cultural-linguistic knowledge and (actual and constructed) resources to be able to decode the text. The crucial point is that the intended readership community must resemble in some way the actual primary community in which the prophetic books emerged in their present form (or forms), since the latter were communicatively successful, and as such they were read, reread, thought about, copied, and thus transmitted from generation to generation from the Persian period on.22 Two more considerations are particularly relevant to the present endeavor. First, a successful prophetic book is one that becomes a permanent part of the authoritative repertoire of the literati in Yehud for generations and as such is read again and again, time after time, and in different contexts. Such a book must not only allow but encourage reading and rereading. Books that are meant to be read and reread continuously must presuppose an intended rereadership, that is a community of readers that approaches the text not only in a way informed by their general literary repertoire, but also within a mode of reading that is not really linear, since every time they read a unit within the book, their awareness of all other units in the book inform their reading (and vice versa).23 In addition, rereading is not only open to an inner-book Had there been an insurmountable gap between the intended and primary readerships of the prophetic books, they would have failed as communicative tools and, thus would have been rejected and not turned into authoritative books. 23 This point will be come important for evaluating some of the arguments associated with the TH. 22



continuous process of reconfiguration of existing information through multiple webs of meanings, but also to (a) recognition and reading activation of subtle and multiple meanings and structures embedded in the text, by the sheer process of multiple rereadings, time after time, and (b) the development of close intertextual links with other works being read and reread constantly by the same group. Second, since understanding usually “means to meet the interactive/communicative expectations of a communicator,”24 the intended communities of rereaders construct their communicators— that is, their implied authors of the books they are supposed to read—in a way consistent with their expectations and readings of the books. Given that these intended communities of rereaders cannot be miles apart from the historical Persian period communities of readers that read, reread, and meditated on these authoritative books, the image of the authors that they construed could not have been miles apart from that of the implied author of the books suggested by an analysis of the intended rereaderships. As mentioned above, it is this author, namely the one construed by the primary communities of rereaders who carried an authoritative voice. It is this (imagined) author who exerted an enduring influence in the discourse of the literati of ancient Yehud, rather than any historically accurate—in our sense of the terms—authors or editors across centuries and generations who may have contributed to the process that led to the present text, but significantly, whose memory was erased through the process. This point is particularly worth stressing for two reasons. First, the intended communities of rereaders were asked to read continuously their books through what today we may call a “synchronic” mode of reading, and not in a way strongly informed by the redactional history of these texts, with all its layers and authors/editors. A number of observations lead to this conclusion. For instance, these texts nowhere inform their readers that as they read their books they must take into serious consideration that cer24

Cf. G. RUSCH, “Comprehension vs. Understanding of Literature,” in: The Systemic and Empirical Approach to Literature and Culture as Theory and Application (S. TÖTÖSY DE ZEPETNEK and I. SYWENKY ed.), Siegen, Siegen University, 1997, 107–19 (p. 115).



tain portions of the text should be taken as “original,” others as a “first redactional/editorial layer,” still others as a “second redactional/editorial level” and so on. Moreover, these texts fail to provide the information that would be necessary for the rereaders to clearly delimit these sections and without which—if one were to follow this hypothesis—proper readings of the text would be impossible. Furthermore, the ongoing process of redaction was not bent on promoting, or archiving and analyzing itself; instead its function was to shape a series of texts in which the last, if successful, was meant to supersede and erase the memory of the previous one.25 This being so, what matters for understanding the ways in which books were read is the preponderance of clear textually inscribed markers and their synchronic inner-book links (which provide both textual coherence and individual texture to each book), along, to be sure, with the issues that they evoked within a particular world of knowledge and general literary repertoire (i.e., general intertextuality within a shared discourse). This approach will, by necessity, create data in a way very different from that of traditional redactional critical approaches. This is so because particular textual wordings would be analyzed from a historically anchored, but synchronic readers’ perspective that emphasizes textual coherence and cotextuality rather than as potential traces of (hypothetical, and from the perspective of the readers erased and irrelevant) redactional processes that led to the creation of the present text read by the community.26 Second, all the implied authors of the prophetic books were imagined as able to write the prophetic texts in their form, including their multiple layers of meanings, familiarizing and defamiliarizing approaches, sharp changes of voices, and the like. Within a community in which the authoritative authors are imagined to write in such a style, and within which literati learned what a prophetic book should look like by reading the prophetic books In this aspect, I agree with most redactional-critical scholars. Moreover, there is clear evidence that the process worked in this way. After all, none of the textual stages proposed in the redactional critical studies of prophetic books—except the last, of course—was successfully reproduced from generation to generation and eventually reached us. 26 For very practical examples, see discussion on catchwords below. 25



in their repertoire, it is very likely that authors (and redactors) who write in such manner will emerge. In other words, one should expect that actual authors would resemble in some way implied authors. This being the case, the common markers used to detect redactional layers and redactors lose much of their validity—after all, writers have included them in their crafting of texts— and as a result, the level of uncertainty associated with redactional proposals sharply escalates. Three main concerns can be raised about the methodological approach outlined here. The first is that the culture and society of the period are reconstructed in part on the basis of our understanding of the set of texts to which the community had access and thus there is an element of circularity in this approach, because we use this reconstruction of culture and society to reconstruct their particular reading of a certain book within that set. Although there is an unavoidable level of circularity in any study of ancient culture or intellectual discourse that is based mainly on written texts, it is controlled and minimized by two factors:27 (a) the cultural/discursive background against which a particular reading of a book is construed is based on the entire intellectual discourse of the reading community not on the particular book or even genre under examination; and (b) factors other than proposed readings of texts play important roles in the historical reconstruction of the social and cultural background against which the particular text and its intended rereaderships is studied.28

27 Its full and complete elimination remains an unachievable goal. Incidentally, this holds true not only for the historical study of societies that left traces of their intellectual discourse in the form of written texts, but for all societies, but the matter requires a discussion that goes beyond the scope of this essay. 28 Among these factors one may mention social models for the production, reception, and roles of high-literacy works—to which only a small elite has direct access—archaeological evidence, approaches to questions of center and periphery in the Persian empire, studies on constructions of the past, social memory and forgetfulness, research on rereading and considerations of historical methodology. Thus, there are controls that serve to minimize the level of circular thinking.



The second, and related, concern grows out of the fact that reading competence is by definition historically contingent.29 In other words, within certain limits texts are decoded differently by various historical communities.30 Thus one may expect a multiplicity of (self-understood) intended readership communities, each associated with a particular context. To illustrate, it is not difficult to imagine a community of readers who considered itself to be the, or at least an intended community of readers of the text and whose readings were governed, for instance, by a Christological viewpoint, or by a tradition of reading these texts as haftarot,31 or by the exegetical principles manifested in the Qumranic Pesharim. It is worth stressing, however, that neither Christological approaches nor those informed by haftarot cycles or rabbinic oral torah or Pesharim hermeneutics were part and parcel of the cultural horizon of Persian Yehud. Any historical readership may imagine itself as an, or even the “intended readership” of the book, but even if they do so,32 they will not fit the concept of “intended community of rereaders” operative in this essay, because the latter is not based on a self-understanding of any particular community, but reflects the (ideal) community of rereaders characterized by the world of knowledge implied in the book as read by the primary readerships in Persian Yehud/Jerusalem. This world of knowledge is clearly historically contingent as it is a product of the society and culture of the particular period and social group. In fact, the historical contin29 I discussed these matters elsewhere. See my “Would Ancient Readers of the Books of Hosea or Micah be ‘Competent’ to Read the Book of Jeremiah?,” and from a different perspective, my Signs of Jonah: Reading and Rereading in Ancient Yehud, Sheffield, Sheffield Academic Press/Continuum (JSOTSupS 367), 2003, 129–54. 30 For an excellent example, see L. K. HANDY, “One Problem Involved in Translating to Meaning: An Example of Acknowledging Time and Tradition,” SJOT 10, 1996, 16–27. For instances related to prophetic literature see Y. SHERWOOD, The Prostitute and the Prophet. Hosea’s Marriage in Literary-Theoretical Perspective, Sheffield, Sheffield Academic Press (JSOTSupS 212), 1996; idem, A Biblical Text and its Afterlives. The Survival of Jonah in Western Culture, Cambridge, Cambridge University Press, 2000. 31 On the meanings created by the haftarot system see M. FISHBANE, The JPS Bible Commentary: Haftarot, Philadelphia, JPS (The JPS Bible Commentary), 2002, and particularly the “third level” of his commentary. 32 They have all the right to do so, within their own discourses.



gency of readings not only does not impede historical research on intellectual history that is based on the study of intended rereaderships communities as approximations to the actual communities of rereaders, but is essential to the endeavor. The third concern deals with levels of uncertainty. Any historical reconstruction is tentative, but all historical reconstructions are not equally tentative. It is pertinent to the present discussion that one is on more solid ground when one starts from the unequivocal evidence that the prophetic books in their present form emerged, or at the very least, were read and reread in such form in the Persian or early Hellenistic period than when one’s starting point is an hypothetical series of sequential (and reconstructed, not extant) texts that are built each on top of the other, as is the case with the traditional redactional critical approaches on which (the historical versions of) the TH are grounded. One is on more solid ground when one focuses on the potential readings of communities of literati that read and reread the text, again and again, and against different circumstances, than when one attempts to reconstruct the intentions of individual writers or redactors (i.e., “the great men of the past.”)33 For one, the approaches behind the TH surmise a (hypothetical) lengthy chain of sequential readings of an ongoing text, in which each link (a) is reconstructed on the precise reconstruction that is advanced for the preceding link in that chain, (b) must point at a precise understanding of the text that led to its re-writing, and (c) this re-writing was well-received by the community, endorsed by it and eventually replaced the text that the community originally received. This is, in itself, a far more problematic approach than the one advanced here, which focuses only on one community and actually existing text. Multiple authors, communities, and texts cannot but increase the level of uncertainty. To make matters worse, the necessarily sequential construction of (hypothetical) texts build33 I should note that there is a general perception in our society that history is a recounting of the deeds of the great men of the past (which includes writing). There are a number of reasons for this perception, but for the present purposes it is worth noting that it creates a longing for narratives about great men and individual agency and tends to de-select narratives about “faceless” communities. This consideration contributes to the preference of some historical narratives over others within society.



ing on each other quickly increases the likelihood that the entire reconstruction will be more likely wrong than correct.34 This effect is compounded by the fact that the reliability of the very markers used to delineate redactional layers and shape redactors is questionable, since as mentioned above, it is reasonable to assume that actual authors among the literati would resemble in some way the implied authors in the literature that they read and reread and through which they are educated. Significantly, most of the proposed redactors are quite dissimilar from these implied authors. Lest I be misunderstood, I am not denying that redactional processes took place in prophetic literature, and that literati were involved in them and that at times involved a blurring of the difference between redactor and reader, even if the outcome was always mediated by the reading communities.35 My point, however, is that the level of uncertainty associated with all these matters is far higher than the one associated with the approach I am advancing and therefore, when the results of these two methodological approaches seem to conflict, preference should granted to the data shaped by historical reading-oriented approaches and the explanations based on these data.36 34 To illustrate, should a scholar’s reconstruction of a text be twice as likely to be correct than not, and should then move only one step more and reconstruct only one previous textual stage of the text s/he has reconstructed and should also her/his proposal be twice as likely to be correct than not, her/his proposal will be correct only four times out of nine, that is, although it would remain the most likely reconstruction, it would be more likely wrong that correct. Redactional critical scholars, including those involved in the TH, often propose more than two textual stages and only a few of them are so bold as to maintain that the chances that their particular proposals are correct are double those of all the alternative proposals advanced in the field (past, present, and future) combined. 35 Cf. M. A. SWEENEY, “Sequence and Interpretation in the Book of the Twelve,” in: Reading and Hearing the Book of the Twelve (J. D. NOGALSKI and M. A. SWEENEY, ed.), Atlanta, Society of Biblical Literature (SymS 15) 2000, 49–64 (50–51). 36 To give priority to a clear reduction in levels of uncertainty carries a hefty price to pay. It involves giving up in practical terms (a) attempts to reach beyond texts more or less in the present form so as to bridge the gap between late monarchic and Persian period Israel/Judah/Yehud and (b) knowledge about the actual prophets, their words, and the reception



3. MAIN ARGUMENTS POINTING AT A HISTORICAL READING OF FIFTEEN PROPHETIC BOOKS IN YEHUD: DATA AND IMPLICATIONS As mentioned above, it is not impossible to read the twelve prophetic books as one book. It is most certainly possible, as many contemporary readers of such a book have clearly shown. Moreover, as readers approach it as one book they are bound to “discover” structures, macrostructures, general themes, and other markers of textual coherence.37 The question, however, whether the balance of the evidence suggests that ancient communities of readers in Persian Yehud were asked to approach and did read their prophetic corpus as a repertoire of fifteen related and interrelated yet separate prophetic books or as four prophetic books, or perhaps—and one must leave that option also open—as both. In other words, the question is not which meanings the literati in Yehud could have found had they read the book as a whole, but whether they did so. To address this question, arguments or approaches that imply the TH are of no use, since they would simply assume what they were supposed to prove. 3.1 One Scroll and Readings On the surface a most obvious textually inscribed marker is the presence of these twelve prophetic books in one scroll. Most likely the TH would have never developed had the books under discussion not been written in one scroll. But when were they written in one scroll and, as importantly, what did it mean in terms of readthat they were granted in the late monarchic period. I am aware that for many this might be a too high price. For that reason alone, redactional critical approaches will continue to be used, but this does not make a difference concerning the argument advanced here. 37 The most obvious case of a collection of separate works that became one book for many readerships is, of course “the Bible,” whether in the form of the Christian Bibles or the Jewish Bible/Tanakh. Readers have found structures, macrostructures, general trajectories, central themes, allusions and the like in “the Bible,” and have assigned particular meaning to the different sequences in which the individual books were arranged. It is worth noting that the concept of “the Bible” does not necessarily involve the production of the work in one scroll, as the Jewish tradition shows. I will return to that point.



ings and readings strategies for those who were used to finding these books in one scroll? To begin with, this material aspect of the production of these texts is likely related to and probably implies the use of skins as the material for writing.38 But whatever the case may be, it is impossible to know whether the practice of writing the twelve prophetic books in one scroll started in the Persian period or later. It is certainly attested in the late Second Temple period, and its aftermath.39 One may note also that the LXX translation of the XII likely came from one hand, which suggests that it was one scroll at the time. The reference in Sir 49,10 refers to the twelve prophets as a group. It is possible, though to be sure not necessary that the See M. HARAN, “Book-Scrolls at the Beginning of the Second Temple Period. The Transition from Papyrus to Skins,” HUCA 54, 1983, 111–22; idem, “Book-Scrolls in Israel in Pre-Exilic Times,” JJS 3, 1982, 161–72; cf. R. DUKE, “Parchment or Papyrus: Wiping Out False Evidence,” SJOT 21, 2007, 144–53. 39 This practice is implied in Josephus, see Ag. Ap. 1.37–43 (40); see also 4 Ezra 14,45, which refer to twenty-two and twenty-four (inspired) books respectively. Acts 7,42 reflects also the existence of a scroll including the (books of the) prophets. Although it cannot be taken for granted that all the Qumran mss. that are usually considered to be copies of the XII included all the twelve prophetic books, the case is particularly strong for 4QXIIc, 4QXIIg and 8ʗevXIIgr, and one has to keep in mind, the later Mur 88. This said, the question of whether along with the mss of the XII there were in Qumran also mss. that included only one prophetic book (e.g., Hosea, Joel, Amos) or some combination of them remains and most likely will remain open given the fragmented character of the evidence. On the Qumran evidence, and form a variety of perspectives, see R. E. FULLER, “The Twelve,” in: Qumran Cave 4. The Prophets (E. Ulrich et al., ed.), Oxford, Clarendon Press (DJD 15), 1997, 221–318; E. TOV, “The Biblical Texts from the Judean Desert—An Overview and Analysis of the Published Texts,” in: The Bible as Book: The Hebrew Bible and the Judean Desert Discoveries (E. D. HERBERT and E. TOV, ed.), London/New Castle, British Library & Oak Knoll Press, 2002, 139–166; G. J. BROOKE, “The Twelve Minor Prophets and the Dead Sea Scrolls,” in Congress Volume Leiden 2004 (A. Lemaire, ed.), Leiden, Brill, (VTS 109) 2006, 19–43; P. GUILLAUME, “A Reconsideration of Manuscripts Classified as Scrolls of the Twelve Minor Prophets (XII),” JHS 7, 2007, article 16, available online at and bibliography in these works. 38



conceptualization of such a group of twelve prophets is related to the writing of these books in one scroll. But even if this is the case, the evidence from Sirach is not necessarily relevant to the Persian period. But even if one assumes that this is the case, what can be learned from the fact that the twelve prophetic books were written in one scroll in terms of the central question discussed here, namely whether these works were read as a single book or as twelve separate books that share a particular genre and the fact that they are much shorter than the other three prophetic books? Since there is no data from the Persian period, the most reasonable way to approach the matter is to look for relevant comparative material. The one-scroll situation is clearly attested by the time of Qumran, Josephus, and rabbinic literature, but none of the three, substantially different, corpora supports the position that these books were read or drew their meanings as parts of a whole (i.e., the Book of the Twelve), rather than as “a whole” by themselves, to be sure, within a general context of repertoire of other authoritative books.40 For instance, Josephus’ reconfiguration of the story of Jonah—the book among the twelve prophetic books to which he pays the most attention—does not show any signs that he was influenced in any way by the writing of Jonah within the scroll of the twelve, quite the opposite, it focuses on a separate and well characterized individual whose story is reconfigured in the light of influences other than the supposed Book of the Twelve.41 There was no Qumranic Pesher or “Commentary” of “the Twelve,” but of separate prophetic books.42 Even a clear collection/anthology such as It is worth stressing that Sir 49,10 says nothing about whether there were four or fifteen prophetic books (reading literary units of first degree) within the authoritative repertoire of the community within which the book emerged. It simply refers to a memory of twelve prophetic personages who each of them comforted Jacob and gave the community hope. 41 See Josephus, Ant 9.208–14; see also L. H. FELDMAN, “Josephus’ Interpretation of Jonah,” AJSRev 17, 1992, 1–29. 42 There are Pesharim or commentaries of the prophetic books of Isaiah (3Q4, 4Q161, 4Q162, 4Q163, 4Q164, 4Q165), Hosea (4Q166; 4Q167), Micah (1Q14, 4Q168?), Nahum (4Q169), Habakkuk (1QpHab), Zephaniah (1Q15, 4Q170), Malachi (4Q253a, 5Q10) and Psalms (1Q16, 4Q171, 4Q173)—Psalms was considered “prophetic” within these com40



Psalms has its Pesher, but not the supposed “Twelve.”43 Moreover, there is no evidence that the use or interpretation of prophetic texts originating from the twelve prophetic books in other works (e.g., CD) was influenced by a reading of these texts based upon, or strongly influenced by the supposition that these books drew (or had to draw) much of their meaning from a “Book of the Twelve.” In fact, at the time, texts from the twelve prophetic books, as well as those from other prophetic books (e.g., Isaiah) and other biblical books, tended to be used in a way that removed them from their original meaning and context, while assuming their authority. (I will return to this point later.) Rabbinic literature, which is vastly different from the Pesharim and represents a significantly different reading tradition, is also well aware of the practice of writing the twelve books in one scroll, but time and again conceptualized them as separate books not as part of one book, or for that matter, of a collection such as Psalms or Proverbs, and certainly not as one book that as a whole stood on par with Isaiah, Jeremiah, or Ezekiel.44 The fact that different texts munities. One may note also 4Q252, 4Q253, and 4Q254, that is, Genesis Peshera, Genesis Pesherb, and Genesis Pesherc. For convenient lists, texts, translations, and bibliography, see F. GARCÍA MARTÍNEZ and E. J. C. TIGCHELAAR, The Dead Sea Scroll. Study Edition (2 vols., Leiden/New York: Brill, 1997/1998). 43 To be sure, Pesharim or commentaries did not have to deal with the entire book (see Pesher Habakkuk, which includes no reference to Habakkuk 3) but one cannot but notice that 4Q171 (4QpPsa) relates to multiple Psalms, within a single Pesher, whereas there is no comparable phenomenon within the Pesharim or Commentaries of the supposed “Book of the Twelve.” 44 A few examples suffice. The claim in b. B. Bat. 14b that the logic governing the arrangement of the prophetic books demands that Hosea should be written separately and positioned before Isaiah shows beyond any doubt that Hosea was considered and read as an independent book, just as Isaiah. Moreover, b. B. Bat 13b—which is often mentioned as supporting the idea of a BT, but see my “Twelve Prophetic Books or ‘The Twelve,’” 132–33—maintains that these prophetic books should not be considered as chapters within a unified book, nor even as integral members of a collection to be understood as whole (e.g., Proverbs, Psalms). The very same term !+ is used there for both the twelve prophetic books and the other three (Isaiah, Jeremiah, and Ezekiel). Furthermore,



shared one scroll was thus not by itself a decisive signal among anNumR 18.17 (/21) sets the Book of Jonah apart from the other eleven prophetic books ()2- !+0 6 +!). It is true that the classification of biblical books advanced in this source serves the purpose of the midrash and its need to reach the target number of fifty books, but the fact that a book within the scroll of the twelve could be easily excluded from the collection, implies a reading of the books in the scroll as independent reading units. This is clearly and unequivocally exemplified by the reading of Jonah in its entirety as the haftara for Yom Kippur as already stated in b. Meg. 31a. Here the book of Jonah was read as a separate book bearing its own meaning (not as a part of the BT that necessarily and substantively draws its meaning from the “whole,” i.e., the BT). The degree of unanimity in rabbinic literature, all across a varied spectrum of texts concerning the matter is particularly noteworthy. (To be sure, there is a masoretic note concerning the total number of verses of the Book of the Twelve, but this is not unexpected given the fact that these notes were meant to help the scribes responsible for writing the scroll and no one doubts that the twelve prophetic books were written in one scroll at that time and already for centuries. Moreover, it is not only that there are similar masoretic notes concerning literary units higher than a single book [or scroll], such as those concerning “the Prophets” [including Former and Latter Prophets], and the Miqra as a whole, but also and far more important that the masoretic tradition included a [seemingly unnecessary?] note about the verse count of each prophetic book within the scroll of the twelve, contrasting with the case of Ezra and Nehemiah, which were considered one book in the MT tradition, or with the case of anthologies such as Psalms and Proverbs in which no chapter or section has a similar note. As for the often cited reference to the middle verse of the scroll of the twelve prophetic books in Mic 3,12, it is anticipated both given its purpose in terms of writing the scroll and the strong symbolism that the text of the verse, which due to a fortunate coincidence happens to be at the middle of the scroll, conveys. In any event, the proposed existence of supposed—but at the very best only implied —messages about the BT as standing on par with the books of Isaiah, Jeremiah, and Ezekiel in the rabbinic tradition can never overweigh the numerous, explicit references to the twelve prophets as individual characters and to the books associated with them as independent books. On these matters see also E. BEN ZVI, “Twelve Prophetic Books or ‘The Twelve’,” 131–34.) Of course, the question of whether the twelve prophetic books were read as one book or as twelve prophetic books in the Persian (or early Hellenistic) period cannot be decided by the evidence of a much later tradition such as the rabbinic one.



cient readerships in terms of how they should read the texts in the scroll as a literary unity or not.45 Further considerations buttress this conclusion. For instance, the writing of twelve works in one scroll shapes by necessity a sequence. It is well known that there existed different sequences of these books.46 The point here is not to discuss them, nor to propose which one of them was original and which were secondary or to explain their potential messages;47 but rather to emphasize that This is not surprising, since works that appeared or were copied in more than one scroll could be and were read as one identifiable work (see Josephus, Antiquities), and conversely, some clearly separate works did appear in one scroll. For instance, 1QS, 1QSa, and 1QSb were all written in one single scroll. 4Q448 includes both Ps 154 and a prayer for the welfare of King Jonathan. See also the different opinions in Soferim 3 about what can or cannot be written in one scroll (e.g., individual Pentateuchal books?, Pentateuch and prophets?, etc.). One may note that there are mss. in Qumran that go beyond one biblical book and include sequences such as Gen-Exod (4Q1, 4Q11) ExodLev (4Q17); Lev-Num (4Q23). 46 Some of these differences may impinge on certain redactional critical arguments developed within the realm of the TH. Due to the reasons mentioned above, my focus, however, is on ancient readings not proposed redactional models. 47 Five different sequences of books are either clearly attested or suggested by the existing evidence: (a) the MT order, (b) the LXX order, which is also supported by 4 Ezra 1,39–40, namely Hosea, Amos, Micah, Joel, Obadiah, Jonah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, Malachi; (c) the one suggested by the Martyrdom and Ascension of Isaiah 4,22, namely Amos, Hosea, Micah, Joel, Nahum, Jonah, Obadiah, Habakkuk, Haggai, Zephaniah, Zechariah, and Malachi; (d) the one suggested by The Lives of the Prophets, namely, Hosea, Micah, Amos, Joel, Obadiah, Jonah, Nahum, Habakkuk, Zephaniah, Haggai, Zechariah, and Malachi—the list actually begins with Isaiah, Jeremiah, Ezekiel, and Daniel and given that follows the LXX in fifteen out of sixteen prophetic characters, it is doubtful that it represents a “random listing of individual names, and (e) a sequence Malachi-Jonah that might have concluded a scroll of the Twelve, and perhaps reflected an original sequence. On these matters see the discussion on 4QXIIa in R. E. FULLER, “The Twelve,” 221–32; cf. J. M. SASSON, Jonah, New York, Doubleday, (AB 24B), 1990, 13–15, and Jones, Formation, p. 6 and passim. For bibliography and a different position, see P. GUILLAUME, “The Unlikely Malachi-Jonah Sequence (4QXIIa),” JHS 7, 2007, article 15, available online at For a com45



the existence of multiple sequences of the twelve prophetic books points at readerships who understood the twelve prophetic books as separate literary works, whose order could be and was reshaped to reflect different priorities. The case is comparable with that of another collection of independent works, namely “the Writings” whose multiple sequences were possible only because “the Writings” were not read as “one book,” and therefore, readers were allowed to place a book such as Chronicles either first or last. To be sure, the MT, LXX, or any ordering of the twelve prophetic books, by necessity, leads to the creation of a sense of meaning and structure. Any attested sequence in a loose written collection would do so. For instance, a Writings sequence in which Psalms is first and Chronicles last may be said to be reflecting an envelope structure, because of the presence of Psalms in Chronicles and some conclusions would follow. Conversely, a sequence in which Chronicles is first and Ezra-Nehemiah last would certainly raise the image of a macro-structural, historically-oriented envelope and bring saliency to the conclusion of Ezra-Nehemiah. A theoretical sequence of the Writings that begins with Proverbs and ends with Qohelet, or one that begins with Lamentations and concludes with Psalms (or vice versa) would lead to relatively obvious structural and ideological links.48 It does not follow from these structural envelopes and ideological messages, however, that any community of readers read the Writings as one book, in which the parts (i.e., the different books in the Writings) drew their meaning in terms of the

parative discussion of the MT and LXX sequences and their organizational/ideological structures see, for instance, M. A. SWEENEY, Twelve Prophets, xxvii-xxix, idem, “Sequence and Interpretation.” 48 Similar arguments may be advanced in terms of prophetic literature, for instance, the sequence of Hosea first and Malachi last may lead to the emergence of a sense of literary or thematic envelope (see, for instance, G. A. TOOZE, “Framing the Book of the Twelve: Connections between Hosea and Malachi,” PhD Dissertation, Iliff School of Theology and University of Colorado, 2002, but had there been a sequence Hosea first, Ezekiel last, an even stronger sense of literary or thematic envelope would have emerged. The same holds true for many other hypothetical sequences, e.g., Nahum first, Jonah last or vice versa. Of course, when “every number is a winner,” much methodological caution is required.



whole, “the Book of the Writings.” We know, for a fact, that such was never the case.49 Turning to the twelve prophetic books, it is easy to find symbolism in the fact that the scroll ends with Mal 3,22–24, but we would have found also symbolism if the scroll would have ended with Jonah (as it might have at some point), or Hos 14,10, or Mic 7,18–20; Zeph 3,20; or for that matter a verse that does not conclude any prophetic book, but a prophetic reading within a book, such as Zech 14,9. In other words, the simple presence of a meaningful conclusion in the last book that fits well the discourse of the time does not constitute testimony that the scroll was read, at least in the main, as a whole.50 The present discussion about “one scroll” did not lead to clear evidence for substantial readings of these books in antiquity as integral parts of whole (i.e., the supposed “Book of the Twelve”) from which they drew their meaning, nor for any reading construction of the Book of Twelve as one on par with Isaiah, Jeremiah, or Ezekiel. To the contrary, there is clear evidence from antiquity of readings of the relevant books as separate, self-contained units— though related to a general authoritative corpus, just as the books of Isaiah, Jeremiah or Ezekiel are—despite their being written in one scroll. To be sure, one cannot learn from the evidence from communities later than the Persian period, whether the same holds true for Persian time communities of readers or not. To address this question, one has to study the ways in which the intended and pri-

It might be claimed that a more comparable case is the collection of Psalms in the Book of Psalms, or of Proverbs in the Book of Proverbs. Significantly, rabbinic literature unequivocally distinguishes between the level of “unity” in these books and that in the scroll of the Twelve. It is reasonable that much of the difference goes to earlier times and is related to the question of whether a single particular historical figure could be associated or even imagined as an embodiment of the book as whole. On this central issue see below. 50 The same holds true for a very general, literary envelope between Hosea and Malachi (i.e., both include references to marriage and divorce). Such an envelope may be perceived if one looks for an envelope, because of one’s assumption that one must be there since there was a “Book of the Twelve.” See above. 49



mary readerships of the books in that period construed their books and their implied authors. This said, it is worth considering the fact that no substantial readings of the scroll as a whole emerged (or at least are clearly attested) among various and substantially different groups among whom the twelve books were written in one scroll and particularly that this happens against a background of mutiple and diverse rereadings. What might have pre-empted the development of such readings? Two tentative answers will be advanced below, but a few other observations must first be discussed.51 3.2 Main Textually Inscribed Markers Signaling the Intended Readers How to Read—Part I—Superscriptions One of the most obvious textually inscribed markers telling a readership whether to read a certain work as a literary unit within a larger book or as a separate work within a general collection are those at the very boundaries, that is, at the beginning and end of the book/unit. The presence of a set of titles and incipits at the opening of each of the twelve prophetic books—and not at the opening of the scroll— clearly suggested to the intended and primary rereaders they should approach each of them as a book, not all of them together as one book.52 It is worth noting that the superscriptions of the prophetic books served both as titles and introductions to the books. As the 51 It is worth noting that communities of readers may approach even a looser collection of books than the scroll of the twelve prophetic books as a whole. The most obvious example is “the Bible,” in whatever form or version. Significantly, in this case, the collection was read as a book not because of unequivocal markers in the text of any of the separate books, or because it was read as such at the time in which each of them emerged in their present form, but because of theological presuppositions of later readers. Some of the issues (and problems) raised by readings of the book (“the Bible”) as a whole resemble matters that the TH confronts as well (e.g., “is there a thematic/ideological centre to the Bible/the Twelve?,” unavoidably, any positive answer to that question marginalizes voices). On these matters see below. 52 This is often accepted among scholars working in the TH framework, but it is either balanced by other claims concerning mainly thematic markers of coherence, or explained away in different ways. See below.



latter, they characterized it as a such, identified it as a separate book that was directly associated with a particular prophetic figure from the past, and often—but not always—served to set the world of the book in a particular time within the construction of their past held by the intended and primary readers and rereaders. In many cases, either the introduction or often a portion of it likely served as a title of the book, and as such it construed a particular image of the book, and of its character. Titles were likely used as signifiers for the book in the discourse of the literati and as a retrieval tool.53 Significantly, there are titles and/or introductions to each of the twelve prophetic books and to the books of Isaiah, Jeremiah, and Ezekiel, but not to the Twelve. A bit more than ten years ago, I wrote: The most significant and unequivocal internal evidence, namely that of the titles (or incipits) of the prophetic books, sets them on the same level of Isaiah or Jeremiah and Ezekiel, namely as separate prophetic books. Most significantly, there is no heading, nor incipit, nor introduction that may serve to orient the readers about the nature of the Book of the Twelve. In this regard, the Book of the Twelve is not only unlike other prophetic books, but also unlike clear heterogeneous collections of works such as Proverbs and Psalms. To be plausible, the proposal that those who wrote or edited the books that were eventually included in the Twelve wished to communicate to their readers that these books should be read as a unit must be able to explain why these writers provided none of the accepted discursive markers for introducing a prophetic book or even for introducing a collection of books. Moreover, it has to do so without multiplying ad hoc hypotheses.54 It is likely that they served some archival needs. One may wonder whether titles were somewhat related (or reflected in) the tags attached to the scroll in ancient libraries. Cf. M. HARAN, “Archives, Libraries, and the Order of the Biblical Books,” JANES 22 [=Comparative Studies in Honor of Yochanan Muffs], 1993), 51–61, esp. 60–61. See also G. M. TUCKER, “Prophetic Superscriptions and the Growth of the Canon,” in: Canon and Authority: Essays in Old Testament Religion and Theology (B. O. LONG and G. W. COATS ed.), Philadelphia, Fortress, 1977, 56–70. 54 E. BEN ZVI, “Twelve Prophetic Books or ‘The Twelve,’ 137. 53



To the best of my knowledge, no good explanation has been advanced to the question of why none of the sequential authors who worked in the supposed “Book of the Twelve” ever introduced an accepted discursive marker (title, incipit) for introducing a book. If they wanted their readership to read the Book as a whole, why did they consistently fail to provide it with such an obvious marker? As important as the preceding question may be, it belongs to a research approach that focuses on authors, editors, not communities of readers. I prefer to re-shape it in terms of communal systems of texts, literary repertoires and discursive worlds that raise both possibilities and constraints. Within this frame, and given that there is no general title the question would be whether something may have constrained the emergence of a single title and the associated reading of the scroll as one book that would have emerged out of it.55 At this point a caveat to the importance of the superscriptions as boundary markers and indirectly to their roles as textually inscribed markers of the separate character of each prophetic book may be raised. On the surface, it might be claimed that the mentioned openings did not have to mark the opening of a new book, or in other words, that their presence did not necessarily indicate to their intended and primary rereaders that the twelve prophetic books should not be read as literary units within a larger book. Certainly texts such as Zech 9,1 and 12,1 resemble headings of prophetic books (cf. Mal 1,1; see also the cases of Isa 2,1—and cf. Hab 1,1—and the subheading in Hab 3,1). But the situation is not analogous, since none of these subheadings include the name of a prophet that is different from the one to which the book is associated.56 We can now turn to the particularities of the prophetic openings, again, from the perspective of the indented and primary rereaderships. Clearly the openings of certain prophetic books partially resemble those in other prophetic books (compare Isa 1,1 An answer to these questions will be given below. They either include none, and thus led the intended and primary rereaderships to add virtually as it were the prophet with which the entire book is associated or explicitly mention that prophetic character. See above. 55 56



with Obad 1 [also cf. Nah 1,1]; compare Hos 1,1 with Joel 1,1; Mic 1,1; Zeph 1,1 and (LXX) Jer 1,1; compare (MT) Jer 1,1 with Am 1,1; and to some extent cf. Hab 1,1 and Nah 1,1; Hag 1,1 and Zech 1,1, see also Ezek 1,1). What would that observation mean within a rereader-centered approach? What could the intended and primary rereaders conclude from an observation of these superscriptions for what they state? The presence of a set of common, potential openings for prophetic books is nothing surprising. It goes hand in hand with genre recognition. From the perspective of the intended and primary rereaderships, the presence of such openings is a sign that they should associate the book they are reading with, among others, a figure in their collective memory, a particular literary genre, a subset of their repertoire of authoritative books or, one may say, a particular shelf of their mental library.”57 But what was in that shelf? Is there any reason to assume that the presence of the mentioned openings suggested to the readers that they should approach their collection as constituted by four books, as suggested by the TH, or for that matter in any cluster of books? This is a particularly important point since at the very end the TH constitutes a claim for assymetrical clusters of books. According to the TH, the Twelve inform each other much more than each of them informs and is informed by the books of Isaiah, Jeremiah, and Ezekiel. The TH is about privileging some potential intertextual links over others. But did the superscriptions suggest to their intended and primary rereaders clusterings governing the relative importance of these potential intertextual links and if so, would 57 It is worth noting that these openings are characteristic of prophetic literature and as such serve as genre markers, contrast with, for instance, the openings of other books (e.g., Genesis, Exodus, those in the so-called DH, Psalms, Proverbs, Chronicles. The (later) books of Nehemiah and Qohelet, both imitate for rhetorical (and ideological?) purposes the opening of prophetic books, while at the same time set themselves apart from them (they imitate only the first part of the heading, not the second) and, by doing so, show that a keen awareness of their literary roles. (For the appearance of textually inscribed pointers raising issues of similarity as well as separateness from existing authoritative books in later biblical books, see also the case of Esther, esp. Esther 1,1 and 10,2–3 and its intended interplay with the historical books.)



they point at the four book distribution advanced by the TH? The answer to this question is unequivocally negative. In fact, it is clear that the particular language of these openings did not serve as a sign to the intended and primary rereaders that they should approach books that open in a similar way as forming a cluster of closely related books informing each other and somewhat separate from other clusters. It is very difficult to maintain that the rereaders of the books of Hosea, Joel, Micah, and Zephaniah (and LXX Jeremiah) were asked to approach them as such a cluster (or those of MT Jeremiah and Amos, or Isaiah and Obadiah for that matter). The simple exchange of wording between LXX Jeremiah and MT Jeremiah also speaks against any reading position that would have paid excessive attention to the precise wording of the superscription. Certain books and particularly those that contain clear references to the reigns of kings of Judah or Israel evoked and interacted with the knowledge and images of the past that these readerships developed on the basis of their rereadings of the Book of Kings.58 These headings shared a role in serving to anchor some books in the construed past held by these rereading communities in the Persian period, but did not create a sense of close clustering, neither temporal nor wording-based.59 (One may notice that even if This is true also of some books which lack the misleadingly called “deuteronomistic” opening, and see Jonah (and cf. Josephus’ reading of that book). It is particularly worth stressing that the prophetic books do not necessarily evoke an image of the past that is supportive of that of Kings. In fact, at times they evoke images that stand in clear tension, as I argued in my “Josiah and the Prophetic Books: Some Observations,” in Good Kings and Bad Kings (L. L. GRABBE ed.), London, T & T Clark International (LHBOTS 393; European Seminar in Historical Methodology 5), 2005, 47–64. Since multiple images of the past are often attested in the same “historical” book (e.g., Joshua, Samuel, Kings, Chronicles), there is nothing surprising in the fact that some prophetic books evoked images of the past that were different from, but complementary to those in Kings within a shared ideological discourse. 59 One may claim that by reflecting on a particular period of the past, they create preferred intertextual links among several books, but books were reread for more than one purpose. Even if one were to grant that these preferred intertextual links were created, they would involve more 58



for the sake of the argument one were to argue against this conclusion, the supposed clusters would cut across the lines separating the Twelve from Isaiah, Jeremiah, and Ezekiel.) In sum, these openings suggest a collection of fifteen separate though interrelated books which all belong to the same genre/mental shelf, and for that matter, which all interact, inform, are informed, evoke and are evoked by other works within the general repertoire of authoritative texts in Yehud, be they prophetic or not. Some scholars supporting the TH have accepted the value of the objection to the TH based on the absence of a heading for the supposed “Book of the Twelve” and the presence of individual headings for prophetic books but suggested that other considerations, mainly of thematic nature, overcome this objection.60 Nogalski, for instance, has argued that the superscriptions play a macrostructural role in the organization of the Book of the Twelve.61 To be sure, this might well have been the case, had there been a Book of the Twelve, but this does not constitute proof that the literati in ancient Yehud actually read the twelve prophetic books as “The Book of the Twelve,” because the conclusion is in fact a reassertion of the main premise of the entire argument.62 3.3 Main Textually Inscribed Markers Signaling the Intended Readers How to Read—Part II Endings and Individualized Voices Another main marker of boundaries is at the end of a book. A comparative study of the prophetic books demonstrates the existhan prophetic books, and more importantly, for our present purposes, lose any relevance to the question of book unity; after all, no one claims that the Book of Kings and Micah or Jonah, for instance, constituted one single book. 60 E.g., G. A. TOOZE, “Framing the Book of the Twelve.” 61 J. D. NOGALSKI, “Intertextuality in the Twelve,” Forming Prophetic Literature, (J. W. WATTS and P. R. HOUSE, ed.), 102–24 (118–24). 62 For other works on the superscriptions from the perspective of the Twelve, see E. W. CONRAD, “Forming the Twelve and Forming Canon,” in: Thematic Threads in the Book of the Twelve (P. L. Redditt and A. Schart, ed.), 90–103 (102–3) and J. D. W. Watts, “Superscriptions and Incipits in the Book of the Twelve, Reading and Hearing the Book of the Twelve (J. D. NOGALSKI and M. A. SWEENEY, ed.), 110–24.



tence of clear genre tendencies concerning this boundary. Their conclusions tend to communicate not only a note of hope, but also a sense of uniqueness to the entire book, just as the introductions, but this time because of their tendency to include markedly unique expressions (see, for instance, Isa 66,24; Ezek 48,35; Hos 14,10; Joel 2,21; Amos 9,15; Jonah 4,11; Obad 1,21; Mic 7,20; Zech 14,21; Mal 3,24).63 These endings provided not only a clear delimitation marker for the intended and primary rereaderships but also a sense of uniqueness to each book64 that was consonant with the tendency to present a particular linguistic flavor, or a choice of particular expressions, or construction of a deeply interrelated and unique network of expressions, and the like that served to provide each book with a sense of uniqueness. For instance, the book of Hosea shows at times a somewhat “odd” flavor—which the rereaders might have associated with Israelian Hebrew. The linguistic choice of including such a flavor in the mix, helped the readers to characterize the voice and character of both Hosea and YHWH, as they were construed in that book. A different choice characterizes the book of Jeremiah, and above all, the characters Jeremiah and YHWH in that book. The same holds true for Isaiah, Ezekiel and books such as Jonah, Zephaniah, Obadiah, Micah.65 In fact, each 63 See, my “The Prophetic Book: A Key Form of Prophetic Literature,” in: The Changing Face of Form Criticism for the Twenty-First Century (M. A. SWEENEY and E. BEN ZVI, ed.), Grand Rapids, MI, Eerdmans, 2003, pp 276–97 (286). I discussed this common feature in several of my commentaries/monographs on particular prophetic books. See, for instance, E. BEN ZVI, Hosea, Grand Rapids, MI, Eerdmans (FOTL 21A, part 1) 2005, 315–16; cf. A Historical-Critical Study of The Book of Obadiah, Berlin/New York, de Gruyter (BZAW 242), 1996, 197–229. 64 See also below under “Practical Differences Concerning Openings and Endings.” 65 In some cases, this becomes particularly clear when one compares similar texts appearing in different prophetic books and notices how each is shaped to be consistent with the tone and voice of their particular books. Cf., for instance, Zeph 3,3–4 and Ezek 22,25–29 (for extensive discussion of these examples, see E. BEN ZVI, A Historical-Critical Study of The Book of Zephaniah, Berlin/New York, de Gruyter [BZAW 198], 1991, 190–205 and esp. 197–205), and Obadiah 1–7 and Jer 49,7–22 (for extensive discussion of these examples, see E. BEN ZVI, Obadiah, 99–114, and esp. 109–14, see also note 50 there), and Micah 4,1–4/5 and Isa 2,2–4/5



prophetic personage and therefore, each of the prophetic books— whether written within the scroll of the Twelve or not—is characterized and individualized by means of a particular voice, which utters unique words or expressions and creates a sense of character and book differentiation and inner coherence. Significantly, the intended readers of these books were asked to imagine YHWH as a deity that adopts different voices to interact with different characters.66 3.4 Additional Markers—Genre Genre awareness and genre expectations contribute much to the way in which readers approach texts. The TH proposes the existence of a book for which there is no par in the entire HB, whereas each prophetic book written in the scroll of the Twelve fits very well with the genre of “prophetic book.”67 Genre conventions within the literati in Yehud would have favored a reading of each of these texts as a prophetic book rather than as an integral part of a larger whole. One may claim, and many have claimed, that the BT is an anthology. Beck has discussed the genre of “anthology.”68 He is correct in noting that the term has been used without much precision. But when precision is brought to bear, it is difficult to find any parallel to the type of anthology that the TH may suggest within the HB, that is, an anthology that is not associated with a dominant (on which see E. BEN ZVI, Micah, Grand Rapids, MI, Eerdmans, [FOTL 21b] 2000, 102–3). Significantly, in each of these examples the counterpart texts are not to be found within the scroll of the twelve, but within the larger frame of the fifteen prophetic books. On this matter see below. 66 The case of the book conclusions is a helpful place to contrast the different results to which the reader centered approach that I am advancing here and some of the approaches used within the TH research realm may arrive. In fact, some book endings and their claimed to be counterpart sections in the next book in the sequence of the scroll have been claimed as support for the TH. These matters are explicitly discussed in section 4, below 67 I discussed elsewhere the genre of prophetic book, see my “The Prophetic Book: A Key Form of Prophetic Literature.” 68 See M. BECK, “Das Dodekapropheton als Anthologie,” and bibliography.



(even if symbolic) character (e.g., Proverbs, Psalms, Song of Songs)69 but twelve equally dominant characters. Again, genre considerations make it more likely that ancient readers (in Yehud and later) would approach the various prophetic books in the scroll as independent works. 3.5 The Central, Connective Weakness of the TH from an ancient Readers’ Perspective, or Why the Twelve Prophetic Books Were Not Read as a the “Book of the Twelve” Is there a central point to which all the considerations mentioned above seem to converge? Or put differently, is there a core issue that explains the clear, systemic preference for the reading of twelve prophetic books rather than of the “Book of Twelve” and from which many of these considerations follow? I propose that there is such a core issue and it concerns a main role of reading prophetic books in ancient Israel. Books embodied social memory. Reading and rereading them served the purpose of making the past present in the community. They administered as it were the presence of the past in the present.70 On the one hand, they pointed to an absence, but on the other, led the readers to visit and revisit, and to imaginatively experience places, events and above all encounter again and again personages from the past—and YHWH along with them, since the deity interacted with them.

Lamentations, of course, lacks such an explicit figure, but is “normalized” by the tradition that associates it with Jeremiah, though perhaps later than the Persian period. The normalization of Lamentations is proof positive to a discursive grammar that preferred to associate anthologies with characters of Israel’s past. This grammar is at work already in the Persian period, and in fact, it is probable that the book of Lamentations was read already at that time in a way that associated or even identifies its poetic voice with that of the prophet Jeremiah. See A. BERLIN, Lamentations, Louisville, Ky., Westminster John Knox, Press, (OTL), 2002, 30–32 and bibliography. 70 To reuse P. Nora’s language in a different setting. See P. NORA, “Between Memory and History: Les Lieux de Memoire,” Representations 26, 1989, 7–24 (20). 69



Prophetic books did not stand for themselves, but as tools for didactic instruction and as material to be activated through reading and rereading so people may visit, imagine, (and of course, construe), the past and its personages.71 It is not by chance that although all the information the readers in the Persian and later periods could gather about the prophets of old was based on the prophetic books, they constantly refer to prophets and their words, not to prophetic books reporting them (see, for instance, Jer 26,18; Zech 1,4; 7,7; 2 Chr 24,19; and cf. Sirach 44–49).72 Within this discourse, prophetic books had to be associated with characters of the past. Conversely, the “person” of the particular prophet that the readers shaped in their minds—through their shared readings and social interaction—symbolized, and embodied, as it were, the contents and messages of the book. This discursive grammar explains why explicit openings that explicitly associate a particular personage from the past were so salient in prophetic books. This explains why so much care was invested through generations of authors and redactors to maintain the particular voice of each prophet and of YHWH’s voice in relation to him, even if the same literati most likely worked on several of these books at the same time.73 It is also the force behind the construction of unique endings to prophetic books and the reason that in the prophetic books, no subunit was ever associated with a prophet other than the one mentioned in the opening of the book.74 Further it explains why there never was, or could have been, a title for the supposed Book of the Twelve. The books of Isaiah, Jeremiah, and Ezekiel had titles because the reading community 71 Through the process the very textuality and materiality of the text become secondary (or even immaterial) to the readers, except for the purpose of evoking the presence of a foundational past in the present. The world of the text becomes from the perspective of the readers identical to the world to be visited and revisited. 72 The same holds true in later traditions. 73 A point I emphasized in my chapters in The Production of Prophecy (D. V. EDELMAN and E. BEN ZVI, ed.). 74 It is well-known that the same prophet may be mentioned in subheadings within the book, but no subheading may carry a different name (see Isa 2,1; 13,1; Hab 3,1).



used them to construe and imaginatively encounter and interact with the prophets Isaiah, Jeremiah, and Ezekiel—all of whom symbolically embodied their books, in the form of their words and their messages. But no character from the past could have been imagined or construed to embody “the Twelve.” There was no possible dominant—never mind, single—character in it. Had the scroll been read as one book, it would have failed to make present within the reading community any particular prophet from the past. But when read as twelve prophetic books, it raises twelve prophets and their voices within the community. The preferred choice within this discourse is clear. Since the basic engine behind the generative grammar of readings is a matter of social memory which is related to social mechanisms that were not so susceptible to change in ancient Israel, the results of this grammar should be seen in other texts as well. For instance, historical narratives75 which were not considered prophecy from the perspective of their primary readership communities lacked textually inscribed markers such as the openings of the prophetic books. The only exception to the rule, however, makes the point since Deuteronomy was considered to a large extent a kind of “prophetic book,” the one associated with Moses, the prophet, and it contained one.76 Whether included in the Deuteronomistic or Primary (i.e., Genesis–Kings) historical collections or Chronicles, or in a parody form in Esther, or in a story as Ruth. 76 The book of Deuteronomy simultaneously occupied multiple shelves in the mental library of the Persian period literati. It was an integral part of the Pentateuchal collection, an integral part of the so-called DH collection, an integral part of the collection we may call “the primary history,” and was also a kind of prophetic book, since Moses was the prophet par excellence. I discussed elsewhere the importance of Moses as a prophet for the reception and development of the concept of prophecy/prophetic books. See my “Observations on Lines of Thought Concerning the Concepts of Prophecy and Prophets in Yehud, with an Emphasis on Deuteronomy - 2 Kings and Chronicles,” forthcoming in a FS. The question of how this tendency, which is grounded in needs of social memory (and in the way in which it is legitimized but also legitimizes the read text that evokes it), impacted books other than prophetic is beyond the scope of this essay. It suffices to say that it was at work in the case of (books later than the prophetic corpora such as) Nehemiah, Prov75



In sum, there was no Book of Twelve nor could there have been a Book of Twelve on par with Isaiah, Ezekiel, or Jeremiah, or even as a collection on par with Proverbs (in which the figure of Solomon is emphatically dominant) and there was no title associating the contents of the scroll of the twelve prophetic books to any prophetic personage, because no possible prophetic figure could have been imagined to have embodied (even if only symbolically) the twelve prophetic voices. The twelve prophetic books individually, not as a collective, stood on par with Isaiah, Jeremiah, and Ezekiel despite their substantial differences in size. Of course, the last statement raises the issue of how to explain these differences in size. A full discussion of the matter is beyond the scope of this chapter; for the present purposes it suffices to note that the relative weight of the figures of Isaiah, Jeremiah, and Ezekiel in the social memory of ancient Israel77 far exceeded those of Joel, Obadiah, or even Amos for that matter.78 The weighty place taken by the events surrounding the destruction of Jerusalem erbs, Qohelet, and Song of Songs. In all these instances, a book title served as a central interpretative key for the entire book and involved a claim of one to one association between book and personage. In the case of Nehemiah it may even have evoked a connotation of prophetic book for Nehemiah. In Proverbs this association stood in tension with claims advanced elsewhere in the text (cf. Prov 30,1; 31,1), but the dominant character of the figure of Solomon could embody the book. In the case of Lamentations, there might have been a connoted association of the book with Jeremiah already at the basic level of the construction of the voice of the speaker by the intended and primary readerships (see above); it certainly evolved later in any case. Psalms, which was considered “prophetic” in Qumran, but likely not in the Persian period, became also associated with its dominant figure, David. Needless to say, later on, tradition associated even the historical narratives with (construed) personages from the past. 77 Cf. Sirach, and concerning Isaiah and Jeremiah also the so-called DH and Chronicles. 78 To a large extent, one may assume that the length of the book associated with a character reflected, even if in a very broad way, the relative weight of the character in social memory. (Cf. the space and frequency of public display of images of historical [or present] figures today, which evoke the presence and memory of these characters. The prophetic books served as “memorials” that evoked the presence and a remembrance of characters among the literati, at least, in Yehud.)



and its immediate aftermath as well as its positive counterparts (e.g., the divine salvation of Jerusalem in the time of Isaiah and the related message of Isaiah 40–66, and that of Ezekiel 40–48) in the social memory of Yehud (or at least, its literati) contributed much to the prominent place that characters such as Isaiah, Jeremiah, and Ezekiel took in that social memory and discourse.79 Excursus The view on the matter of the twelve prophetic books vs. the “Book of the Twelve” in the light of the effort to make foundational past characters present in the community80 served to bring together the observations that emerged from the historical, reading-centered approach I advocated above. It is worth noting that a parallel and complementary development concerning the use or role of prophetic texts from the late Persian period seems to have also contributed to the systemically dis-preferred character of the type of readings advanced by the TH. The intended readership community of Chronicles was supposed to pick citations and allusions to prophetic texts from their prophetic book’s repertoire. In the vast majority of the cases, the texts were placed in times preceding the putative time of the prophetic character to whom the 79 On why the books of Isaiah, Jeremiah, and Ezekiel grew immensely while others did not, see also my comments in “De-historicizing and Historicizing Tendencies in the Twelve Prophetic Books: A Case Study of the Heuristic Value of a Historically Anchored Systemic Approach to the Corpus of Prophetic Literature,” in Israel's Prophets And Israel's Past: Essays on the Relationship of Prophetic Texts And Israelite History in Honor of John H. Hayes (B. E. Kelle and M. MOORE, ed.), London and New York, T. & T. Clark, (LHBOTS 446), 2006), 37–56. Additional considerations associated with the relatively unique character of Ezekiel (see, for instance, its utopian vision of temple, cult, and land) may have contributed to the development of Ezekiel into a massive book. The centrality of the figure of Isaiah likely reduced the need for a development of a social memory of Micah as a/the main prophetic voice during the time of Hezekiah and its successful stand against Assyria. More research in these matters is needed. 80 That is, from the perspective of social memory and its roles constructing and legitimizing characters, messages, and, above all, as providing focal points for shared social identity and constructions of continuity, and the like.



book was associated. As I mentioned elsewhere, “the words [in these cases] have a life of themselves as it were and may apply to future and past events, even in a way unbeknownst to the speakers… [t]he words themselves become a-temporal so as to become multi-temporal.”81 This mode of reading which is later attested in the Pesharim, 1 Maccabees and later NT and rabbinic literature was certainly not conducive to the readings advanced of the scroll of the Twelve advanced by the TH, because it tends to see books not as such but as a collection of verses, half-verses, or relatively brief pericopes at the most.82

4. BACK TO METHODOLOGY: PRACTICAL, CONCRETE DIFFERENCES BETWEEN TH APPROACHES AND THE ONE ADVANCED HERE IN TERMS OF QUESTIONS THEY RAISE AND HOW THEY SHAPE AND INTERPRET EVIDENCE Section 3 has shown the kind of conclusions to which the approach favored in this essay leads and how they make historical sense when approached from a memory perspective. The sharp degree to which these conclusions differ from those brought forward by the methodologies commonly used to support the TH is easily noticeable and draws much attention to the basic question of methodology at the heart of the present debate. Scholars work with the same texts, but they do see different things, because they use different research lenses and raise very different questions. Since methodology is, and will remain, at the heart of this debate, I will explore below four areas in which the methodological divergence outlined in general terms in section 2 leads to very practical, and most substantial differences in shaping and interpreting evidence and therefore, by necessity, on the pro-

81 E. BEN ZVI, “One Size Does not Fit All. How Did Chronicles Deal with the Authoritative Literature of its Time?,” forthcoming in a collected essays volume. 82 Concerning rabbinic literature see B. SOMMER, “The Scroll of Isaiah as Jewish Scripture, Or, Why Jews Don’t Read Books,” Society of Biblical Literature 1996 Seminar Papers, Atlanta: Scholars Press, 1996, 225– 242 (esp. 231).



posed conclusions.83 Needless to say, these explorations are meant to contribute to a clarification of the debate. 4.1 Practical Differences Concerning Recourse to Memory Studies The value of studies on social memory for the reconstruction of roles of readings and the general generative grammars that develop them within the discourses of ancient Israel is clear from the perspective of the approach followed here. It is fair to say, however, that the TH research framework, as much of the field in prophetic literature, does not turn to studies on memory to advance its understanding of the Twelve. This is not only, but to a substantial extent a by-product of the author/redactor centered approach that characterizes this research framework and the questions that it tends to raise. An approach centered on the community of readers as the one presented here, on the other hand, cannot but notice that continuous reading and rereading is a social act that serves particular functions in society. In the case of the prophetic books, it is difficult to imagine that such a reading and rereading would not involve shaping and enacting memories, contributing to remembering, forgetting, and bringing the past and its characters to bear into the present of the community. 4.2 Practical Differences Concerning Openings and Endings There is probably no point in which the approach advanced here and the one that characterizes much research supporting the TH differ more markedly, and with far reaching consequences, than in the case of openings and endings of prophetic books. These openings and endings were understood (see above) as markers encouraging the intended and primary rereaders to approach each book separately—though, of course, within a large discourse of the period and in a way informed by all other books in the same mental shelf (i.e., the prophetic books) and all other texts considered authoritative by the rereaders. Yet at the same time, many of the 83 In principle, one may wish for an exhaustive study of these differences, but this is an impossible goal within the frame of any article. An exploration of four areas seems like a practical compromise that suffices to raise awareness for the main issues at stake.



book endings and their claimed to be counterpart sections in the next book in the sequence of the scroll have been claimed as support for the TH. For instance, Nogalski writes “Joel begins with an extended call to repentance (1,1–2,17), precisely as Hosea ends (14,1–8).84” Nogalski gives precedence to the occurrence of two very common patterns in prophetic readings, certainly not limited to these two books over the very wording of the conclusion of Hosea and the beginning of Joel.85 It is not only that the two calls are presented as different, but that the book of Hosea actually does not end with a call for repentance, but with a very different, and unique statement in 14,10 and Joel begins not with a call of repentance but with a title/superscription in 1,1 that associates the book with a character different from Hosea. From the perspective of the intended readers, both Hos 14,10 and Joel 1,1 are clear markers of separation. From a perspective influenced by the TH and its common approaches this message is replaced by one of continuity based on similar, though very common, literary patterns. To be fair, elsewhere Nogalski points to a few catchwords that link Joel 1,1–14 back to Hos 14,5–10).86 These words are: (a) 7 “this” (Joel 1,2, twice) and % ”these” in Hos 14,10 (7 is not present in Hos 14,5–10); (b) !6! “inhabitants” (Joel 12.14—note that he splits the construct from the absolute noun, which is problematic from a cognitive semantic perspective; after all it is the entire construct chain that evokes a concept—Hos 14,5, in the context of a complex, multivocal, and certainly unique expression,87 namely E_“ ‘! The text actually reads, “14:18,” but this is certainly a typo. See J. D. NOGALSKI, “Reading the Book of the Twelve Theologically,” Int 61, 2007, 115–22 (117). 85 I leave aside the issue that the sequence Hosea–Joel is far from universal. See above. 86 See J. NOGALSKI, Literary Precursors, 21–24; idem, Redactional Processes, 13–14. 87 “The expression denotes or connotes to its intended rereadership a number of related and complementary meanings, which may be reflected in the following renderings: (a) “those who sit in (/under) his [YHWH’s] shade shall return, they will give life to [i.e., raise; cf. Isa 7,21] grain”; (b) “again, those who sit in his shade shall give life to grain” (cf. D. BARTHÉLEMY et al., Critique Textuelle de l'Ancient Testament, vol. III, Fribourg, Édi84



* ‘‘ EK ‰! FR2 ‰ !Ž _’‰ ! and contrast with 14 ‘ ‘ !Ž _F! ‰ %’Pp and !Ž _F! ‰ 14 ‘ ‘ in Joel 1,2 and 14 respectively—significantly that expression

appears in Hosea, but only in Hos 4,1, not in Hos 14,5–8/9–10); (c) *!! “wine” (Joel 1,5—this time disassociating the absolute from the construct noun [the relevant expression is * ! ‘! !7’Ž _–%P‘ ]; and doing the opposite in Hos 14,8 *F+‘ %‰ *! Ž!P‰ , which is in itself a unique expression in the HB, and one of many in Hosea 14,5–8) (d) *0 “vine” (Joel 1,7, 12 and cf. *0 ‘?# E4‰ 0‰ ! ‰ in Hos 14,8, which is again a poetic expression that appears here and nowhere else in the HB; the closest expression is in Song 6,11; 7,13, well beyond the scope of the corpus of prophetic literature); and (e) * “grain” (Joel 1,10 —see 6F4!d _!F  * ‘A‘ A _“ !P 4‘ 2‰ ! %% )‰ “ and cf. v. 11; this common triad of nouns referring to agricultural goods appears in Jer 31,12; Hos 2,10, 24; Joel 2,19; and Hag 1,11 within the corpus of prophetic literature and in many instances in the HB such as Deut 7,13; 11,14; 2 Kgs 18,32; 2 Chr 32,28; Neh 5,11, but significantly this triad is not present in Hos 14,5-8 (9/10)—and Hos 14,8 (again in * ‘‘ EK ‰! FR2 ‰ !Ž _’‰ ! E_“ ‘!). Leaving aside the question of whether Nogalski has here a solid system of catchwords that may help him to reconstruct the redactional history of these books—the matter falls outside the research frame I propose—it is most unlikely, and I would dare say unthinkable, that the intended and primary readers of these two texts not only (a) noticed the repetition of four very common words among hundreds in the relevant texts—a feat for which they would have had to look at these texts in a very a-contextualized manner—and even (b) added to these four a combination of “this” and “these” but also (c) thought that such a repetition created a sense of textual bond between these two texts that outweighed the most salient and overt differences between the two texts, including the explicit and unique language and imagery of tions Universitaires [OBO, 50/3], 1992, 623–24); (c) or “those who sit in his shade shall again revive the growth of grain” (cf. A. A. MACINTOSH, Hosea, Edinburgh, T&T Clark [ICC], 1997, 573–76); (d) “those who sit in his shade shall return, they shall revive [like] grain (see CALVIN, “Hosea,” in: J. CALVIN, Commentaries on the Twelve Minor Prophets, vol. 2 [transl. J. Owen] Grand Rapids, Baker House, p. 500; KJV); and (e) “they shall return, they shall sit in his shade, they shall make grain live.” (E. BEN ZVI, Hosea, p. 298).



Hos 14,5-8, the unique conclusion of Hos 14,10, and the superscription in Joel 1,1. A historical, reading-centered methodology that focuses on synchronic readings of texts—because at any stage they urged their readers to, and were most likely read in that way— cannot but conclude that their readers were asked and did approach them as clearly separate, not as two parts of a whole that derive meaning from that whole.88 It is worth noting also that had readers internalized a criterion according to which the simple appearance of five very common words among hundreds in two texts are proof positive that they should read them as bound together, they would have found very few texts that would have failed that test. Of course, this is not the claim that supporters of the TH advance, but proposed redactional reconstructions do not count as relevant evidence from the perspective of the historical reading approaches I advance here. Similarly, one cannot assume beforehand a (very heavy handed) understanding of the TH according to which the presence of a sequence Hosea–Joel in the MT and related texts must have been understood by readers in Yehud (or Hellenistic/Roman periods, for that matter) as a request directed to them to comb the last six verses of Hosea (why only these?) and the first fourteen of Joel for potential bridges linking the texts, so they may be read as a whole, and then use that “evidence” to support the original assumption, namely the TH, or such a drastic version of it. The case of Hosea and Joel discussed above is far from unique. For instance, redactional critical followers of the TH may observe a system of catchwords linking Obad 15–21 and Mic 1,1– 7, as maintained by Nogalski, and discuss its implications for the redactional history of the Book of the Twelve.89 A scholar following a reader-oriented approach would immediately notice that the Significantly, Nogalski himself notes “the Stichwort connection between Hosea and Joel was not noticed or not important to the translators of the versions.” See J. NOGALSKI, Literary Precursors, 24. 89 Such a system would serve as a proof for a stage in the development of the book that preceded the incorporation of Jonah. There is, however, no Obadiah–Micah sequence in any attested arrangement of the prophetic books or list of prophetic characters of which I am aware. For the analysis of this case of catchwords, see J. NOGALSKI, Literary Precursors, 31–33. 88



words involved are: 4 “mountain” 3-! “Jacob” 6 “fire” *4)6 “Samaria” and 6 “field,” all of which are very common. She or he will notice that to be counted these words had to be “abstracted” as it were from their respective literary contexts (i.e., fully decontextualized), and that both texts present themselves as extremely different to any reader. Moreover, such a scholar would immediately observe that both texts carry the usual, textually inscribed markers that separate among books, namely a title/superscription/introduction and a clear and unique ending. Finally, questions of openings and endings raise also the question of what kind of evidence do even seemingly successful bridges that involve even some shared or potentially shared keywords or concepts constitute? And evidence for what? To illustrate, a good case of such a “successful” link would involve Mal 3,22 and the following book in the “Bible” in the Hebrew traditional text, namely Ps 1,1. Moreover, one may add even a back-link between Ps 1,1 and Hos 14,10, which is the conclusion of the first of the twelve prophetic books. Yet clearly these texts were meant to be read as a unit within their primary community of readers. Significantly, even this “successful” intertextual case did not evolve into a stable sequential bond among texts, even within much later Jewish communities which would have approved the theological meanings that such a link would have engendered, as the possibility of shifting Chronicles to the beginning of the Writings clearly shows (see Leningrad Codex). From a reading perspective, even seemingly successful links do not constitute evidence that the relevant books were read as parts of a whole. Similar wordings, key words and the like among books convey allusions, evoke other texts, and the like. They do point at a broad sense of intertextuality within a certain repertoire but no more than that. 4.3 Practical Differences—Sequential vs. Branched Modes of Reading and the matter of General Web-like Constructions of the Repertoire and Knowledge of Ancient Literati A. Schart writes: The call to return within Joel presupposes that the addressees know why they should return. How can they realize that the relationship to God has been lost in the first place? Within Joel, no exam-



ple is given showing how the hearers had turned away from God or have offended God, Joel makes no effort at all to spell out what sin is at stake. Even God’s anger is never mentioned. Joel’s citation of Exod 34:6 in 2:13 makes no mention of the retributive side of God’s nature. As a result, if readers had only Joel, they would be totally at loss to explain why God brought about this extraordinary plague upon God’s people. Only readers who have first read Hosea are prepared to understand Joel’s prophecy. They know of Israel’s sin in a detailed way, as well as YHWH’s anger that motivates a devastating punishment (emphasis mine).90

Schart’s version of the TH moves, indeed, into reading considerations, but as it does so, it proposes a strict sequential mode of reading in which information is provided to the reader only through the linear sequence of his proposed Book of the Twelve. Similarly, he claims that readers of Amos alone would have encountered a serious difficulty in Amos 4,6–13, because there was no substantial prophetic call to return in the preceding chapters in Amos, but one who reads the Twelve sequentially and is aware of the information provided by Hosea and Joel will understand that YHWH has indeed called for repentance, but such a call was rejected (p. 114). Schart’s assumption of a linear sequential mode of reading that does not allow for filling seeming information gaps from outside of the world of the text—as understood by him—shapes evidence and, in turn, this evidence is presented as supporting the model. The model of reading advanced here is grounded in a web, branched discursive model. People reading the book of Amos or Joel have access not only to the preceding books as written in the scroll of the Twelve, but a large world of knowledge that they use to decode the meaning of their text. They do need to read Hosea and Joel to activate the common image of Israel not listening to YHWH’s call to repent, or to read Hosea to be able to understand Joel. After all, within their discourse the motif of YHWH’s bringing calamity to Israel activates enough images.

A. SCHART, “The First Section of the Book of the Twelve Prophets: Hosea—Joel—Amos,” Int 61, 2007, 138–52 (citation from p. 142). 90



Sequential modes of reading may be used to suggest evidence by retrojecting ideas expressed towards the end of the book into preceding chapters. A concrete example is the claim that “in the Persian period Zechariah was the lens through which the other prophets were read—and perhaps the template by which they were written or edited.”91 O’Brien explicitly acknowledges that such views were common in Persian Yehud, but still maintains that “Zechariah was the lens…” and even maintains that “the scenario I have described supports a redactional scheme in which Hosea through Zephaniah were consciously edited as a preface to Zechariah, providing a portrayal of the ‘former prophets’ useful to the writer of Zechariah (180).” Evidence is shaped around the assumption of a “Book of the Twelve” and its linear reading. From the perspective of the approach advanced here, the widespread conceptualization of the “former prophets” in the way she describes in the Persian period, which is well attested well beyond prophetic literature, provides a world of knowledge and discourse within the book of Zechariah and ample control of the reading codes necessary for “decoding” it without any need for a preface creating or spelling them out. It is worth noting that these two approaches diverge in the way in which they shape other evidence. For instance, within the model advocated here every book is an equal participant in the repertoire of the community and engages freely with all others. Thus the construction of prophecy mentioned in Zechariah is simply a manifestation of a common position along with others such as the one evoked in the book of Jonah. Within the logic of the TH model supported by O’Brien, Jonah’s voice and its own construction of prophecy becomes subordinate to that of Zechariah; after all, within this model it is only a section in a long preface to Zechariah, not an independent book whose standing is equal to it.92

J. O’BRIEN, “Nahum—Habakkuk—Zephaniah: Reading the “Former Prophets” in the Persian Period,” Int 61, 2007, 168–83 (citation from p. 172). She emphasizes the way in which prophecy is construed in texts such as Zech 1,2–6; 1,14–2,4; 7,7–14; 8,11–15). 92 O’Brien is aware that unifying readings cannot but marginalize numerous voices (see p. 182). 91



It should be noticed that the common tendency in the TH research realm to stress sequential readings93 is, however, not an essential feature of the TH. The very concept of rereading, the presence of multiple allusions, signposts, cross-references, and at times overlapping structures suggests that prophetic books were likely meant to be read by their intended readerships (and likely their primary readerships) not necessarily, and not always in a narrow linear manner. As much as this applied to all the prophetic books, it might have applied to the Book of the Twelve, had it existed as a book to be read and reread as such. An essential difference that separates the way in which evidence is shaped and interpreted between the TH and the present model concerns the classification and evaluation of textual allusions or cross-references. From the perspective a model of fifteen prophetic books within a larger repertoire of authoritative books, it The tendency is very common. Notice for instance the following argument: “If Nogalski is correct arguing for Joel’s role as the ‘literary anchor’ for the Book of the Twelve, despite presumably being composed later than the books that adjoin it, establishing the framework for the rest of the Book, then it is appropriate for Joel to be the first of the Twelve to use Exod 34,6 explicitly.” See R. L. SCHULTZ, “The Ties that Bind: Intertextuality, the Identification of Verbal Parallels, and Reading Strategies in the Book of the Twelve,” in: Thematic Threads in the Book of the Twelve (P. L. REDDITT and A. SCHART, ed.), 27–25 (38). Since the other references in “the Twelve” to which he refers are in Jonah 4,2; Micah 7,18; and Nah 1,2–3, to say that Joel is “the first” to use Exod 34,6 assumes a thoroughly linear (and text-based only) mode of reading—after all, Joel is not associated with any clear historical period in the memory of Israel. Such a sequential strategy of reading, as opposed to circular or multi-directional reading strategy even within the realm of the TH carries implications. Such a reading makes Joel 2,13 a key text in ways in which alternative readings would not be able to. The matter is important for some redactional critical studies within the TH. In other words, even within the TH frame, choices do shape evidence. The present example suffices, but they can be multiplied and particularly in reference to Joel (see, for instance, “since Joel’s superscription has no date, this implies that his message is not related to a specific situation in the history of Israel, but to something more far-reaching and thus to a more important phenomenon” [SCHART, “First Section,” p. 148]; the lack of temporal reference is given here a meaning that is different from that of other superscriptions that lack it. This makes sense if one assumes a certain version of the TH.) 93



does not matter whether the cross-reference or allusion involves books within the twelve, or within the fifteen, or for that matter, within the entire authoritative corpus available to the literati. Thus, for instance, Jer 25,30; Amos 1,2; and Joel 4,16; Zeph 3,3–4 and Ezek 22,25–29; Obad 1–7 and Jer 49,7–22; Micah 4,1–5 and Isa 2,2–5 (and see also Joel 4,10); Isa 13,6; Ezek 30,3, Joel 1,15; Obad 1,15; Zeph 1,7; Mic 3,10 and Hab 2,12; Isa 47,8 and Zeph 2,15 and the like will be treated equally.94 These will be understood as excellent examples of texts evoking texts, of texts informing and being informed by traces of other texts, of texts emerging out of a “sea” of common motifs and concerns and bringing its existence to bear on the readers of the texts, and as significant signs of a social nonfragmented memory and general discourse coherence. Within this model, however, none of them will be understood as evidence for inner book textual coherence, or as evidence supporting particular redactional proposals involving an ongoing book eventually shared by the relevant references. Within a TH approach, the latter is not the case when two or more of the relevant texts occur within the twelve prophetic books. The cases are not inherently different, but their value depends on the lens (i.e., assumptions about the twelve prophetic books) adopted by the scholar. 4.4 Practical Differences—The Question of “Main Themes” Reading the Twelve “as a book,” often involved sequential readings as mentioned above. It also often involves a sense that there must be some central themes. It is easy to understand why this is the case. Both sequential readings and central themes are normally considered markers of textual coherence and thus an integral part of what reading a book as a book is meant to be. It is not surprising that when similar approaches to the “Bible” emerged, so did the question of the theological center or centers of that book. Supporters of the TH, and even some who considered the Twelve an

94 The same holds true for Exod 20,6; 34,6–7; Num 18,16; Deut 5,9; Joel 2,13; Jonah 4,2, Micah 7,18; Ps 82,5; 103,8; 148,9 and cf. Nah 1,2–3; Jer 32,18; to which later was added Neh 9,17.



anthology,95 tend to write about main themes and usually among them, the Day of YHWH, the fate of Israel, and in the MT sequence, the role of Jerusalem and the nations.96 Significantly, none of these issues are unique to the supposed Book of the Twelve. All them are part and parcel of the discourse of the literati of Yehud and appear outside the twelve prophetic books.97 All of them can be easily explained in terms of the general discourse of the period with no recourse to the TH. But the search for unifying themes shapes evidence by making salient references to these themes within the books written in the scroll of the Twelve, by connecting them and, far more importantly, by downplaying non-unifying themes that do appear in these books. A simple example suffices: Any reference to the Day of YHWH as a main theme in the Twelve marginalizes a book such as Jonah, and so does the reading of Hosea through Zephaniah as consciously edited as a preface to Zechariah to provide the portrayal of the ‘former prophets’ advanced in that book, as discussed above. If the book of Jonah is understood as a book being read as such by itself, then it serves as a central meta-prophetic book commenting on all the other prophetic books.98 However, when 95 E.g., D. L PETERSEN, “A Book of the Twelve?,” in: Reading and Hearing the Book of the Twelve, (J. D. NOGALSKI and M. A. SWEENEY, ed.), 3–10. 96 The latter point has been argued extensively by M. A. SWEENEY. See his Twelve Prophets, xxvii-xxix, idem, “Sequence and Interpretation.” 97 This includes references to the Day of YHWH, see Isa 2,12;; 34,8; Ezek 7,19; 13,5; 30,3; Lam 1,12; 2,1.21.22; and notice the possible contrast between the relevant expressions there and the one in Isa 58,5. There is, of course, the issue of whether there was a collocation, despite all its possible variants, that pointed to a clear concept, or even concepts, of the Day of YHWH in the discourse of Yehud or later communities of readers of these books. The matter, which is open for debate has, of course, an impact value on assessments about the potential centrality of the concept of “the Day of YHWH” in the Twelve. Yet, the issue stands beyond the scope of this paper. For a recent and significant analysis on the matter see D. ISHAI-ROSENBOIM, “Is ‘ '! (the day of the Lord) a Term in Biblical Language?,” Bib 87, 2006, 395–401. For the sake of the argument here, I would simply assume that there was such a concept or range of concepts. 98 See my Signs of Jonah.



Jonah is read as a part of the Twelve that derives its meaning from the whole, its voice and its themes become by necessity, at the very least partially marginalized by the so-called unifying themes in that book. The choice of which is the case and which type of evidence Jonah provides to historians who wish to reconstruct the intellectual discourse of Yehud depends on one’s methodological approach, and the implications of that choice on the question of the TH.