The Economy of Ancient Judah in Its Historical Context 9781575064147

The dynamics of ancient Judah’s economy are among the most important, but also neglected and least understood, aspects o

185 42 2MB

English Pages 288 [280] Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Economy of Ancient Judah in Its Historical Context
 9781575064147

Citation preview

The Economy of Ancient Judah in Its Historical Context

The Economy of Ancient Judah in Its Historical Context

edited by

Marvin Lloyd Miller, Ehud Ben Zvi, and Gary N. Knoppers

Winona Lake, Indiana Eisenbrauns 2015

© 2015 by Eisenbrauns Inc. All rights reserved. Printed in the United States of America. www.eisenbrauns.com

Front cover: Photo of silver-coated bronze Persian-period coin by Pavel Shargo, the Sonia and Marco Nadler Institute of Archaeology, © Lau­ tenschlaeger Azekah Expedition survey. Azekan imitation of Greek tretra drachma. Obverse: image of Athena; reverse: owl. Mid-5th to mid-4th century b.c.e. Cover design by Andy Kerr.

Library of Congress Cataloging-in-Publication Data The economy of ancient Judah in its historical context / edited by Marvin Lloyd Miller, Ehud Ben Zvi, and Gary N. Knoppers.    pages cm Includes bibliographical references and indexes. ISBN 978-1-57506-413-0 (hardback : alk. paper) 1.  Economics in the Bible.  2.  Finance—Biblical teaching.  3.  Bible. Old Testament—Criticism, interpretation, etc.  4.  Jews— Economic conditions—History—586 b.c.–70 a.d.  I.  Miller, Marvin Lloyd, editor.  II.  Ben Zvi, Ehud, 1951—editor.  III.  Knoppers, Gary N., 1956—editor. BS1199.E35E26 2015 221.6′ 7—dc23 2015031483

The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984. ♾ ™

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  vii Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Part 1 Introduction Cultivating Curiosity: Methods and Models for Understanding Ancient Economies . . . . . . . . . . . . . . . .   3 Marvin Lloyd Miller Part 2 Economic Indications from the Hebrew Bible The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31 as a Source for Reconstructing Aspects of Thought and Economy in the Late Persian / Early Hellenistic Period . . . . . .   27 Ehud Ben Zvi More than Friends? The Economic Relationship between Huram and Solomon Reconsidered . . . . . . . . . . . . . . . . .   51 Gary N. Knoppers Agrarian Economy through City-Elites’ Eyes: Reflections of Late Persian Period Yehud Economy in the Genealogies of Chronicles . . . . . . . . . . . . . . . . . . . . . . .   77 Louis Jonker Ancient Comparisons, Modern Models, and Ezra–Nehemiah:   Triangulating the Sources for Insights on the Economy of   Persian Period Yehud . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Peter Altmann Part 3 Economic Indications from Other Literary Sources ‫ּורבּו‬ ְ ‫ ְפרּו‬and the Seventh Year: Complementary Strategies   for the Economic Recovery of Depopulated Yehud . . . . . . . . 123 Philippe Guillaume Exploitation of Depopulated Land in Achaemenid Judah . . . . . . 151 Lisbeth S. Fried v

vi

Contents

The Achaemenid Policy of Reproduction . . . . . . . . . . . . . . . . 165 Josef Wiesehöfer The Economy and Administration of Rural Idumea at the End of the Persian Period . . . . . . . . . . . . . . . . . . . . 175 Diana Edelman Part 4 Economic Indications from Archaeology Forces of Decline and Regeneration: A Socioeconomic Account of the Iron Age II Negev Desert . . . . . . . . . . . . . . . . . . . . 207 Yifat Thareani The Rural Economy of Judah during the Persian Period and the Settlement History of the District System . . . . . . . . . 237 Oded Lipschits Indexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   Index of Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . .   Index of Scripture . . . . . . . . . . . . . . . . . . . . . . . . . . . .   Index of Other Ancient Sources . . . . . . . . . . . . . . . . . . . .

265 265 270 274

Preface Gary Knoppers and Ehud Ben Zvi attended the meeting of the Canadian Society of Biblical Studies held in Montreal in 2010. One night, at a social gathering of participants in the Ancient Historiography Seminar as well as friends and students at the home of Patricia G. Kirkpatrick, Gary and Ehud talked about the lack of attention in research to ancient economies in depopulated areas, such as Yehud during the Persian period. In the course of this discussion, they also mentioned the relative lack of interaction between historians of ancient economies and their works, on the one hand, and historians of ancient Israel and general “biblicists” and their works, on the other hand. They both agreed that something must be done to address what they perceived to be a substantial gap. The outcome of this conversation was the establishment of two workshops: one held in May 2013 at the meeting at the Canadian Society of Biblical Studies (Victoria, BC, Canada) and the other on July 2013 at the meeting of the European Association of Biblical Studies (Leipzig, Germany). Marvin Miller presented an important methodological paper at the Canadian workshop. At that time, it was already evident that a volume would emerge from these workshops, so Gary and Ehud were delighted that Marvin agreed to join the editorial team for the volume and take a leading role in its production. The chapters of this book are, for the most part, revised versions of papers read in one of the two above-mentioned workshops. Although the European conference focused originally on Agrarian Economies in Depopulated Areas in Persian Antiquity, the range of issues discussed in both conferences was properly and by necessity wider; this led to the present reconceptualization of the theme as The Economy of Ancient Judah in Its Historical Context. One of the best ways of advancing a research agenda is to bring together a number of scholars with specialties in different fields and with varying perspectives to share their insights and thus to further their knowledge through conversation. This volume represents, to a large extent, a “written” conversation with multiple threads, viewpoints, and as usual, ever-shifting sets of converging and diverging lines for the reader to explore rather than merely the results of a meeting of some research “school.” Moreover, this volume is not meant to provide a single or “definitive” answer to any question but to further discussion on the topic by presenting a panorama of ways in which these questions may be approached. vii

viii

Preface

Although the editors have consistently maintained the formal independence of the chapters so that each may be read on its own, the volume as a whole is greater than the sum of its parts. To be sure, readers will sympathize more with certain approaches and topics than others. However, they will easily recognize that, together, the contributions of the participants approach a much larger, indispensable horizon than the individual perspectives. As for technical matters, we have followed the abbreviations in The SBL Handbook of Style (2nd ed.; Atlanta: Society of Biblical Literature, 2014) and the so-called social-science format. We would like to thank all of the presenters and participants in the sessions from which this volume emerged. We also thank the Canadian Society of Biblical Studies and the European Association of Biblical Studies for providing a wonderful venue for these conversations, and Jim Eisenbraun for accepting this volume for publication and for the contribution that his publishing company, Eisenbrauns, makes to our field. A particular expression of gratitude goes to Anna Cwikla, who helped us prepare the original manuscript, and to Beverly McCoy, who prepared it for publication and improved it in multiple ways with her comments and requests for clarification. Our gratitude to our respective families for keeping us “grounded” almost goes without saying. —Marvin Lloyd Miller Ehud Ben Zvi Gary N. Knoppers

Abbreviations General Abbreviations Akk. Akkadian BM tablets in the collections of the British Museum ET English translation HB Hebrew Bible IAA Israel Antiquities Authority jps Jewish Publication Society version LXX Septuagint MT Masoretic Text neb New English Bible nrsv New Revised Standard Version PF Persepolis Fortification archive PN Personal Name Syr. Syriac Tg. Targum Tg. J. Targum Jonathan Vg. Vulgate

Reference Works AA AASOR AB ABD

Archäologischer Anzeiger Annual of the American Schools of Oriental Research Anchor Bible The Anchor Bible Dictionary. Edited by David Noel Freedman et al. 6 vols. New York: Doubleday, 1992 ABRL Anchor Bible Reference Library ACCS Ancient Christian Commentary on Scripture AIL Ancient Israel and Its Literature AnBib Analecta Biblica AOAT Alter Orient und Altes Testament AoF Altorientalistiche Forschungen AOS American Oriental Series AP Cowley, A. Aramaic Papyri of the Fifth Century b.c. Oxford: Clarendon, 1923 ASOR Archaeological Reports  American Schools of Oriental Research Archaeological Reports ATD Das Alte Testament Deutsch BA Biblical Archaeologist BASOR Bulletin of the American Schools of Oriental Research BDB Brown, Francis, S. R. Driver, and Charles A. Briggs. A Hebrew and English Lexicon of the Old Testament. Oxford: Clarendon, 1906

ix

x BEATAJ

Abbreviations

Beiträge zur Erforschung des Alten Testaments und des antiken Judentum BHS Biblia Hebraica Stuttgartensia. Edited by K. Elliger and W. Rudolph. Stuttgart: Deutsche Bibelstiftung, 1983 BibInt Biblical Interpretation Series BKAT Biblischer Kommentar, Altes Testament BMAP Kraeling, E. G. Brooklyn Museum of Aramaic Papyri. New Haven, CT: Yale University Press, 1953 BN Biblische Notizen BZ Biblische Zeitschrift BZAW Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft CAD The Assyrian Dictionary of the Oriental Institute of the University of Chicago. Chicago: The Oriental Institute of the University of Chicago, 1956–2006 CBQ Catholic Biblical Quarterly CHANE Culture and History of the Ancient Near East CT Cuneiform Texts from Babylonian Tablets in the British Museum DCH Dictionary of Classical Hebrew. Edited by D. J. A. Clines. 9 vols. Sheffield: Sheffield Phoenix, 1993–2014 DJD Discoveries in the Judaean Desert DZc Darius, Suez c inscription: P. 147 in Kent, R. G. Old Persian: Grammar, Texts, Lexicon. 2nd ed. AOS 33. New Haven, CT: American Oriental Society, 1989 ErIsr Eretz-Israel FAT Forschungen zum Alten Testament FRLANT Forschungen zur Religion and Literatur des Alten und Neuen Testaments HALOT Koehler, Ludwig, Walter Baumgartner, and Johann J. Stamm. The Hebrew and Aramaic Lexicon of the Old Testament. Translated and edited under the supervision of Mervyn E. J. Richardson. 4 vols. Leiden: Brill, 1994–99 HAT Handbuch zum Alten Testament HCOT Historical Commentary on the Old Testament HSM Harvard Semitic Monographs HSS Harvard Semitic Studies HTS Harvard Theological Studies IAA Reports  Israel Antiquities Authority Reports IB Interpreter’s Bible. Edited by George A. Buttrick et al. 12 vols. New York: Abingdon-Cokesbury, 1951–57 ICC International Critical Commentary IEJ Israel Exploration Journal IG A Selection of Greek Historical Inscriptions to the End of the Fifth Century b.c., by M. N. Tod. 2 vols. Oxford: Clarendon, 1933–48 ISAP Institute for the Study of Aramaic Papyri JAOS Journal of American Oriental Society JANER Journal of Ancient Near Eastern Religions

Abbreviations

xi

JAR Journal of Archaeological Research JBL Journal of Biblical Literature JCS Journal of Cuneiform Studies JESHO Journal of the Economic and Social History of the Orient JHebS Journal of Hebrew Scriptures JNES Journal of Near Eastern Studies JNSL Journal of Northwest Semitic Languages JSOT Journal for the Study of the Old Testament JSOTSup Journal for the Study of the Old Testament Supplement Series JTS Journal of Theological Studies KAT Kommentar zum Alten Testament KS Kirjath-Sepher LCL Loeb Classical Library LHBOTS The Library of Hebrew Bible/Old Testament Studies LSTS Library of Second Temple Studies MdB Le Monde de la Bible NCB New Century Bible NCBC New Century Bible Commentary NEA Near Eastern Archaeology NEAEHL New Encyclopedia of Archaeological Excavations in the Holy Land. Edited by E. Stern. Jerusalem: Israel Exploration Society and Carta / New York: Simon & Schuster, 1992 NETS A New English Translation of the Septuagint and Other Greek Translations Traditionally Included under That Title. Edited by Albert Pietersma and Benjamin G. Wright. New York: Oxford University Press, 2007 NICOT New International Commentary on the Old Testament OBO Orbis Biblicus et Orientalis OIP Oriental Institute Publications OTE Old Testament Essays OTL Old Testament Library OTS Old Testament Studies OtSt Oudtestamentische Studiën PEQ Palestine Exploration Quarterly PJ Palästina-Jahrbuch PLO Porta Linguarum Orientalium RB Revue biblique REA Revue des études anciennes SAOC Studies in Ancient Oriental Civilizations SBLABS Society of Biblical Literature Archaeology and Biblical Studies SBLDS Society of Biblical Literature Dissertation Series SBLMS Society of Biblical Literature Monograph Series SBLSBS Society of Biblical Literature Sources for Biblical Study SBLSup Society of Biblical Literature Supplement SBLSymS Society of Biblical Literature Symposium Series SJOT Scandinavian Journal of the Old Testament

xii SR TA TAD

Abbreviations

Studies in Religion/Sciences Religieuses Tel Aviv Textbook of Aramaic Documents from Ancient Egypt, by Bezalel Porten and Ada Yardeni. Jerusalem: Hebrew University, Dept. of the History of the Jewish People, 1988–89 TZ Theologische Zeitschrift UF Ugaritic-Forschungen USQR Union Seminary Quarterly Review VT Vetus Testamentum VTSup Supplements to Vetus Testamentum WAW Writings from the Ancient World WBC Word Biblical Commentary WMAT Wissenschaftliche Monographie zum Alten Testament WMANT Wissenschaftliche Monographie zum Alten und Neuen Testament WTJ Westminster Theological Journal WUNT Wissenschaftliche Untersuchungen zum Neuen Testament ZAW Zeitschrift für die alttestamentliche Wissenschaft ZDPV Zeitschrift des deutschen Palästina-Vereins

Cultivating Curiosity: Methods and Models for Understanding Ancient Economies Marvin Lloyd Miller Summit Pacific College

The quest to understand ancient economies has crossed disciplinary boundaries, raising issues of methods. In this introductory chapter, I will consider some key theoretical matters and propose a way forward by suggesting three important processes for building an ancient economic model and recommending a way in which these methods can be applied to agrarian economies. The intent of this volume is to encourage a dialogue between experts in Ancient Israel, Biblical Studies, and cognate disciplines, such as anthropology, archaeology, history, and ancient economics and generate new approaches for the study of agrarian economies in depopulated areas. First, key terms such as market exchange, marketplace, and market system need to be defined. Using Christopher Garraty as a guide, I will refer to marketplaces as “physical locations that accommodate regularized and orderly market exchanges” (Garraty 2010: 9). Marketplaces consisted of discrete physical locations where merchants gathered regularly in villages, towns, and cities. Market exchange, on the other hand, “can occur in any number of centralized or decentralized settings” (Garraty 2010: 10). Market exchange may take place at the temple during ceremonial activities, at other pubic assemblies, family workshops, and through traveling merchants. A form of market exchange is barter, which is a system of exchange “that does not employ the media of exchange” (Garraty 2010: 8). Finally, the concept of market system “refers to a regional network of interconnected marketplaces, including the market hinterlands they provision” (Garraty 2010: 10). Second, by way of historical background, the debate about the origins of market exchange is still influenced by the work of Karl Polanyi Author’s note: An oral, earlier version of this essay was presented at the 2013 annual meeting of the Canadian Society of Biblical Studies held in Victoria, British Columbia. I would like to express my thanks to the participants for their valuable comments and to other scholars who read an earlier draft of this contribution and made valuable suggestions. I am glad to acknowledge the support of the Social Sciences and Humanities Research Council of Canada (SSHRC) for its financial support.

3

4

Marvin Lloyd Miller

(Polanyi 1957). Polanyi argued that a “market mentality” was a product of modern capitalism and viewed non-Western economies as embedded in social, political, and religious institutions. Although he believed that trade did exist in premodern economies, he contended that trade was largely an elite activity of long-distance transactions in prestige goods. His views came to be known as the “substantivist” position1 and were a reaction to the formalist model2 of classical and neoclassical economists, who attempted to seek a unifying theory of economic behavior by employing positivist approaches that claim that humans possessed a tendency for market exchange long before formal markets were established (Garraty 2010: 11).3 A recent example of a positivist position occurs in Is God an Economist? in which Sigmund WagnerTsukamoto writes that a “deeply capitalist ethic seems to organize the Old Testament” (Wagner-Tsukamoto 2009: 11).4 There arose a longstanding substantivist-formalist debate among modern historians and economists over whether ancient economies were similar to modern ones but less developed or whether they were “embedded”—that is, an economic exchange that is “integral to the whole social fabric and does not stand above it” (Meikle 2002: 245).5 In his seminal work The Ancient Economy, Moses Finley argued with greater cogency and rigor against the applicability to the ancient economy of a market-centered analysis as a “misguided application to ancient society of modern theories of investment, banking, and credit” 1.  This position should not be confused with the “modernist” model that argues that modern economic concepts like “capitalism” and “the market” could be applied directly to classical antiquity. For instance, Rostovtzeff (1936: 231–52) contends that the entire Hellenistic world was bound together in a universal price-setting market. 2. Paul Cartledge (2001: 15) gives this explanation of the formalists’ position: “For the formalists, the ancient economy was a functionally segregate and independently instituted sphere of activity with its own profit-maximizing, want-satisfying logic and rationality, less ‘developed’ no doubt than any modern economy but nevertheless recognizably similar in kind.” 3.  An early example of a positivist position is A. Smith (1776, repr. 1976: 630), who wrote in 1776 that “private interests and passions of men” create a situation “which is most agreeable to the interest of the whole society” (cited in Garraty 2010: 11). 4. See also E. F. Davis (2009), who suggests that biblical farmers were similar enough to North American farmers to justify the claim that observing modern farmers and reading modern agrarian writings can help us better understand scriptures. For other writers, who hold a positivist position and draw theological positions from the scripture, see Fick (2008) and Evans, Vos, and Wright (2003). 5.  Meikle (2002: 245) suggests that a “disembedded” economy is one “which is torn out of the social fabric to become an independent entity that dominates social decision making.”

Methods and Models for Understanding Ancient Economies

5

(Meikle 2002: 235) and drew attention to “considerations of status and civic ideology, rather than the laws of supply and demand, [which] governed economic decision-making” (Finley 1973). Calls to move beyond these dichotomizing and rather exhausted debates are becoming more insistent.6 Some scholars have reacted to these schools of thought and have embraced a neoclassical approach to economic theory embodied in the concept of Homo economicus. According to this theory, people tend to be ego-directed as opposed to being socially motivated, and behavior can be predicted based on rational, utility-maximizing agents (Stanish 2010: 190–93). The difficulty with this approach, according to Stanish, begins with the word rational, which is problematic because of the wide semantic range of the term and because people have a penchant toward social cooperation exhibiting “irrational, pro-social” behavior even at a cost to themselves (Stanish 2010: 192). Robert Axelrod describes the manner in which decisions are made in this way: “In complex situations, individuals are not fully able to analyze the situation and calculate their optimal strategy. Instead, they can be expected to adapt their strategy over time based upon what has been effective and what has not” (Axelrod 1997: 14). In sum, it seems to me that the substantivist-formalist debate has run its course, and it seems pointless to revive.7 Gone, too, are the binary positions that view people as either economically rational or as “agents acting for the social or moral good through embedded institutions” (Stanish 2010: 193). A fresh model is needed that moves beyond the anti-market and anti-anti-market divide8 that best accounts for the variety of ways that people structured their economic lives. In order to add clarity to the discussion, I believe there are three processes or activities that need to be considered in order to establish an approach that will reflect better the economic situation of Yehud 6.  The primitivist-modernist dichotomy consists of a position in which modernists understand ancient economies as being well developed and differing from modern economy only in degree; however, the primitivist view disagrees. The substantivist-formalist divide emphasizes the behaviors that determine exchanges. The formalists hold that ancient economies can be analyzed by basic behavioral assumptions, such as profits, supply, and demand; whereas, the substantivists use a different set of behavioral assumptions, such as status, patronage, etc. A formalist is most likely to be a modernist, and a substantivist is most likely to be a primitivist, but not necessarily so. 7.  For a fresh treatment of economic theory of the ancient world, see Boer (2015: 9–52). He favors the concept of régulation as a matrix to explain how stability occurs in the midst of crisis, since stability is not the norm; crisis is. 8.  This term is used by Stanish (2010: 190).

6

Marvin Lloyd Miller

in the Persian period.9 First, various types of market exchanges need to be considered; once those are understood, second, the relationship between political, religious, and market institutions can be examined. These two methodological considerations prepare the way to discuss how market economies evolve.

Economic Activity 1: Types of Market Exchanges The first consideration in model building is to clarify the structure and performance of ancient economies. To address this issue, I find helpful the insights of economic anthropologist Carol Smith, who argues that there are three fundamental types of exchange: dyadic, polyadic, and market exchange (Smith 1976: 314, 321, 334–35, 353). Her position allows for a variety of economic systems operating concurrently and envisions their evolution, a perspective that results in more complex conclusions about how ancient economies can be understood. This approach moves beyond Finley’s position that problematically favored arguments of continuity and unity of the ancient economy. Dyadic exchange consists of direct trade between equals and occurs most frequently where social stratification is minimal. A description of dyadic exchange is exchange that is “open and resembles a large, decentralized network connecting more or less equal nodes” (Braswell 2010: 130). What distinguishes this form of exchange from others is that production and exchange are organized among several local systems, without being able to take advantage of the benefits of economic integration into the regional economy (Smith 1976: 318). For example, Abraham is said to buy a burial cave for his wife from Ephron, the son of Zohar, “that he may give me the cave of Machpelah, which he owns; it is at the end of his field. For the full price let him give it to me in your presence as property for a burial place” (Gen 23:9). Although this example does not consist of an exchange between producer and nonproducer, it does provide an illustration of exchange that can occur without a central market place. This exchange is directly negotiated between two clan leaders—namely, Abraham and Ephron. In a similar manner, dyadic exchanges occurred in small agrarian villages or hamlets, where the settlement hierarchy is frequently simple and based on kinship groups. This is particularly true when resources are scarce and the survival of the group depends on labor cooperation and produce distribution (Guillaume 2012: 101). In the Persian period, the economy in the hill country was agricultural and, accord9.  By Persian period, I mean the Achaemenid or first Persian Empire (550–330 b.c.e.).

Methods and Models for Understanding Ancient Economies

7

ing to Lipschits (2006: 30), “[T]he Achaemenids, like the Babylonians that preceded them, were interested in maintaining the existing rural settlements in the hill country, which were an important source of agricultural goods that in turn were probably, in part, collected as tax.”10 Although the biblical text offers few explicit cases of dyadic exchange during the Persian period, it seems plausible that a vibrant commercial exchange occurred between kinship groups, which may not have had access to regional markets. Polyadic exchange takes place between individuals of unequal status and is “bounded, relatively small in size, and hierarchically ordered around the central hub of the chief and the chiefly village” (Braswell 2010: 130). A possible example of this kind of exchange occurs during the early monarchic period, when King David is told by Gad to “raise an altar to the Lord on the threshing floor of Araunah the Jebusite” (2 Sam 24:18). David then purchases the threshing floor and builds an altar on it in order to avert the plague. This exchange occurred between a high-status individual and a subordinate (2 Sam 24:18–24). When one is considering examples of polyadic exchanges in agrarian economies, patronage is a helpful concept. According to this perspective, a patron is in a socially superior position over a client and provides contracts and employment to him or her in exchange for honor and support. It seems reasonable to suggest that during the Persian period dyadic and polyadic types of exchange were utilized in agrarian economies. When markets evolve, market exchange tends to be more complex— urban centers grow concurrently with changes in the organization of the elite. Carol Smith suggests three kinds of market exchange: those controlled by political concerns, market variables, or competition. Of the three types, the most likely method of commerce in the Persian period is a monopolistic exchange controlled by political or religious concerns.11 In this type of exchange, the movement of trade is “controlled from a well-organized core, penetrating less organized peripheries,” and “local elites become identified with the peasant ethnic group that surrounds them, losing political and economic power to the external elites in distant metropoles” (C. A. Smith 1976: 345–46).12 Although there is an indication of some aspects of market exchange in Jerusalem, 10.  For another perspective on the typology of villages, see Faust (2003: 37–53). Hoglund’s theory of a “Persian policy” of ruralization is speculative (Hoglund 1999). 11.  C. A. Smith (1976: 345–53) refers to this type of exchange as a dendritic market system. 12.  Braswell (2010: 130) states that in a monopolistic exchange the “relationship between rural producers and middlemen is regulated by the elite, and retailing follows market principles.”

8

Marvin Lloyd Miller

it would go beyond the evidence to suggest that Jerusalem was a major urban center, even at the height of the Persian period (Lipschits 2003: 323–76; 2006: 30–34).13 Therefore, when one wants to suggest a system of monopolistic exchange where long-distance interaction may have been the norm (Braswell 2010: 132), a better geographical area to consider is along the coast, where urban and commercial centers were established (Lipschits 2006: 26; Stern 2001: 380–407). The economic goal of the Persians most likely was to secure the Via Maris and other overland trade routes in an effort to secure trade with southern Palestine and Egypt, as well as to encourage maritime trade (Ephʿal 1998: 117; Stern 2001: 371). It appears that the Achaemenids “allowed the coastal population a great degree of independence, at least in regard to internal, commercial, and economic affairs” (Lipschits 2006: 27; Elayi 1980: 25; 1982: 83–110; Stern 1990: 155). If that is the case, then these cities are what Richard Smith calls “trade states” and “existed only under special conditions in which they served as hubs or nodes for long-distance trade” (R. L. Smith 2009: 53). The fact that long-distance trade occurred between at least one coastal city and Jerusalem is supported by Neh 13:16, which states that “Tyrians also, who lived in [Jerusalem], brought in fish and all kinds of goods and sold them on the Sabbath to the people of Judah, in Jerusalem itself!” (Edelman 2006: 207–46). The amount of market interference between the coastal cities and the Persian administration is unknown; however, it appears that these “trade states” were able to establish markets and were not significantly regulated or controlled by the government.14 In order to provide a useful heuristic, I have isolated the formal characteristics of each type of exchange. However, they should be viewed as a continuum based on how an ancient economy was organized. Dyadic and polyadic exchanges are face-to-face transactions and do not require a marketplace. In contrast, the Tyrians in Jerusalem most likely sold their products at a marketplace and had access to commercialized market economies. Market exchanges were multifaceted and dynamic, and households had access to various forms of the market, depending on proximity to marketplaces and urban centers and conditional on the need for specialized items. To characterize ancient Persian period Yehud as having either a market or nonmarket economy is an oversimplification. It is more helpful to consider how provisions may have 13.  There appears to be population and agricultural growth around the city during this period; however, evidence for growth within the city seems to be lacking. See Lipschits in this volume. 14.  For a discussion on Neh 13:15–22, see Altmann in this volume.

Methods and Models for Understanding Ancient Economies

9

been procured by agriculturalists and to reflect on the extent to which households relied (or not) on marketplace exchange. What this survey shows is that a model that reflects the ancient economy must include various types of exchange and not feature one to the exclusion of others. Our study focuses predominantly on agrarian economies, and therefore a model must emphasize dyadic and polyadic exchanges that occurred most frequently between and among villages and small towns. In other words, kinship and patronage exchanges were integrated and formed an ancient social and economic network. The logical extension to understanding the structure and performance of ancient economies is to consider how political and religious institutions influenced market exchanges.

Economic Activity 2: The Relationship between Political, Religious, and Market Institutions The process of model building in the ancient Levant must include how the Persian administration and religious elite influenced the markets. The topic of how economic exchanges among individuals should be regulated or controlled by political or religious administrations is a matter of much scholarly debate. According to some economic theorists, governments are viewed as interfering with market traders;15 according to others, governments ensure peaceful marketplace interactions;16 and other historical economists view the role of governments as “order [that] is rooted in networks of social relations that promote trust and principal cause of cross-cultural variation in regional market organizations” (Garraty 2010: 20, quoting Granovetter 1985: 481–510. Unlike Philippe Guillaume (2012),17 who considers politics merely an “external 15.  This is the neo-classical view espoused by Adam Smith. 16.  This is the view of Polyani. Inspired by Polyani are J. Davis (1992) and Sahlins (1972: 279–314). Marxist models also emphasize the role of principals in promoting economic development. Xenophon writes that the troops were in Palestine to protect the imperial provinces from their enemies. Walter Miller’s translation of Cyropaedia 7.5.69 reads: And since he considered that all Babylon, too, stood in need of adequate protection, whether he himself happened to be at home or abroad, he stationed there also an adequate garrison, and he arranged that the Babylonians should furnish the money for their wages, for it was his aim that this people should be as destitute of resources as possible, so that they might be submissive and as easily restrained as possible.

17.  Guillaume (2012). He states that “Apart from particular moments when an individual may blur the distinction, patronage should be distinguished from the state” (p.  156), and “[p]olitical and economic powers are quite distinct and often compete with each other” (p. 159).

10

Marvin Lloyd Miller

variable” that may affect markets, a more comprehensive theory should consider political, religious, and market institutions as co-evolutionary (Garraty 2010: 10). Politically, the Persian government benefited from market exchange in several ways. First, taxes were a common means of extracting funds and commodities. Ezra lists three types of taxes exacted by the Persian administration ‫מנדה‬,18 ‫בלו‬, and ‫( הלך‬Ezra 4:13, 20; 7:24). Part of the tax revenue was extracted from marketplaces, where regular and centralized exchanges occurred, as well as from agricultural produce. Another way that Persian governing officials may have interfered with market exchange in ancient Levant was when there was a military threat. If the region became unstable, then the Persian administration may have interfered with or entered into the market exchange in order to convert commodities needed to feed the army and receive provisions for garrisons that were being established. Some historians of ancient cultures have seen a military threat as a positive boost to the economy, but others view it as an economic drain on marginal cultivators. Either way, a military passing through an area had an effect on the economic lives of its inhabitants. Third, the Persian government may have desired to remove some items from market exchange. For instance, when Alexander the Great entered Susa, he found 270 tons of coined gold and about 1,200 tons of silver stored as bullion (Yamauchi 1990: 102–19). The figures may be exaggerated or possibly unhistorical; however, it can be said that the hording of precious metals does divert product from entering the market. Governments frequently used products and manual labor for staterun construction projects, such as roads, temples, and other buildings, thus diverting labor and technologies from entering the market. What these three economic activities suggest is that the Persian government may have benefited from marketplace development in at least three ways: taxes, conversion, and control over product or labor distribution (Garraty 2010: 20). Therefore, it is beyond dispute that the Persian government was an active player in the markets in ancient Levant, a fact that has been used by some ancient economists to highlight that the development of mar18. This is the only time that middah (tax/tribute) appears; cf. the Akkadian mandattu > maddatu > Hebrew: midda. Mankowski (2000: 84) states that “tribute” reflects direct Akkadian borrowing. Herodotus 3.91 reads, “The fifth province was the country . . . [which] paid three hundred and fifty talents; in this province was all Phoenice, and part of Syria called Palestine, and Cyprus” (Herodotus, LCL [Godley]). This represented 2.5 percent of the treasury’s total annual income of 14,560 talents (Yamauchi 1990: 274). See also Fried and Edelman in this volume.

Methods and Models for Understanding Ancient Economies

11

kets in the Persian era consisted of a strong “top-down” emphasis. According to this position, governing agencies assumed a prominent role as revenue generators, thus increasing domestic and foreign markets (Polanyi’s view). This perspective also allows for the interests of rulers and merchants to be complementary by suggesting that merchants “needed the stability and protection a political authority could provide, and governments saw merchants as a source of revenue that was usually more willing to be taxed than other groups with surplus income such as warrior or religious elites” (R. L. Smith 2009: 51). Another approach emphasizes that markets flourished apart from centralized government intervention. According to this view, agrarian economies grew predominantly as a “bottom-up” process promoted by market traders and producers. What strong governments offer market places is to prevent insider manipulation and to ensure peace and settle disputes (J. Davis 1992: 69). What needs to be proven rather than assumed is that the Persian government had the political infrastructure to procure and police market exchanges throughout the vast Empire (Mann 1986).19 An alternative to governing agents controlling market exchange is to consider how social networks and moral codes may have ensured reliable and safe exchanges, and combated abuse and unfair pricing. Nehemiah 5 provides an example of “top-down” and “bottom-up” exchange. This account uses words of commerce, such as “mortgage,” “borrowing,” “tax,” “interest,” and “lending” and features calls for the need for assurances of fair exchange.20 Social bonds, cultural beliefs, and moral codes are appealed to in the communication between the peasants and governor. The context of the moral life of the peasants as brothers of “the same flesh and blood as [their] countrymen” (v. 5) implied that they had a set of expectations about relations between themselves and the elite. This reference expressed the common value promoted by both the poor and the elite—the sons of the borrowers were fellow Judeans as well as those of the lenders. Bilateral kinship provided a justification that those who met the expectations of patronage, consideration, and helpfulness would be treated with respect, loyalty, and social recognition (Scott 1985: 184). Not only were the peasants concerned with their domestic needs but also with social reputation. They strengthened their case for community status by stating that, even though “the poor man’s son [was] as good as a rich man’s” (v. 5), 19.  Mann (1986: 8) states that market exchange over large regional systems “spreads in a more spontaneous, unconscious, decentered way throughout a population.” 20.  For a detailed analysis of Nehemiah 5, see Fried in this volume.

12

Marvin Lloyd Miller

there existed social stratification and unfair economic treatment. This example suggests a strong “bottom-up” exchange in which peasants approach one of their own. Although the appeal to a solution to economic hardship was made to a “moral economy of kinship and a shared consciousness of religious identity” (Abbott, Smith, and Gallaga 2007: 461–84),21 this text suggests that market exchange was managed by a Persian appointed governor. Thus, on the one hand, there was a political framework that regulated the economy (as Finley suggests), and on the other hand, there is evidence that other institutions and organizations contributed to economic activity, for the appeal to fairness was based on both kinship and religious association. What would be helpful to determine is which factor of market exchange was most influential in Yehud. If the situation in Nehemiah 5 is reflective of economic influences, then a model of agrarian economies must include both “top-down” and “bottom-up” market exchanges, depending on the situation. To approach this concern, we need to consider how markets may have functioned, changed, and varied and to suggest what indicators may be used to explain market development.

Economic Activity 3: How Market Economies Evolve To help answer questions related to causal factors in market development, I have found the study conducted by Richard Blanton and Lane Fargher helpful. They have proposed three main causal factors to explain premodern market development (Blanton and Fargher 2010: 218–19).22 These will be used to suggest another step in our model building concerning how agrarian economies may have evolved in Persian Palestine. First, Blanton and Fargher predict that there is a correlation between increased markets and the replacement of the traditional elite with salaried officials. According to their model, markets tended to grow when patrimonial domains were weakened in favor of state builders, who devised suitable administrative systems to carry out their regime building goals. Shmuel Eisenstadt notes that rulers who “are stymied in their attempts to extend their domain of land, labor, and other resources are bound up in primordial social units that prevent their alienation” (Eisenstadt 1963; quoted by Granovetter 2002: 87). It 21.  This observation is made by Abbott, Smith, and Gallaga (2007: 461–84) with regard to the marketplaces developed in the Hohokam region (of Arizona) in connection with the spread of ceremonial ballcourt villages. This general observation can reasonably be applied to the situation in Nehemiah 5. 22.  Blanton and Fargher 2010: 218–19. Their analysis is from data collected from Europe, China, and Mexico and, their model is based on collection action theory.

Methods and Models for Understanding Ancient Economies

13

appears that Nehemiah’s appointment to Yehud as a Persian governor served as a political and economic strategy of the Persian administration. Replacing the leadership may have had the effect of augmenting commercialization by extending market territories, which included trade between the coastal cities and Jerusalem. Another indicator that provides additional support for the idea of increased bureaucratization of the Persian Empire is their use of a standard currency that may have promoted broader market participation and the escalation of garrisons and served military or economic purposes (Hoglund 1999: 167; Boer 2015: 187–89).23 This hypothesis helps explain the variability of the degree of market exchanges and supports a “top-down” approach that appears to have occurred in Persian Palestine. Second, markets tend to develop when agricultural production intensifies. Returning to Blanton and Fargher (2010: 218–19), they suggest that, [w]here increased production and market participation are judged to be beneficial, households located in prime agricultural areas tend to specialize in the production of grains for household consumption and market sale. In contrast, households located in marginal agricultural settings where anticipated costs of intensification exceeded expected returns, increase production by mixing extensive agriculture (usually in cash crops) with craft specialization or pastoralism. Finally, households located in urban centers, where arable land is scarce, specialize in politics, market exchange, and handicraft production.

If Blanton and Fargher’s assessment is correct,24 then agrarian economies respond to demands for surplus in different ways. Those kinship groups that were located in the plains and foothills where rainfall was relatively plentiful would intensify production by including vines and olives in production areas where wheat, barley, and oats were also grown. In contrast, in the hill areas where rainfall was sporadic, crops would have been supplemented by animal herding and craft specialization instead of increasing investment in expensive irrigation systems. Another factor that may account for the development of markets is whether the agrarian land was situated along major caravan or world-trade routes in which case economies tend to specialize in crafts and nonagricultural goods. What this analysis suggests is that diversification and specialization may not have been the result of governmental supervision but, rather, a strategy agrarian households employed to 23.  Eskenazi (2006: 526) views these garrisons as being too small for military purposes and suggests that they may have served to protect commercial routes. 24.  Their analysis is from data collected from Europe, China, and Mexico.

14

Marvin Lloyd Miller

meet their needs and to buffer against unforeseen shortfalls (Halstead and O’Shea 1989).25 Third, markets are a product of population growth and urbanization. Although the coastal region increased in urbanization and trade, the provinces of Yehud and Samaria and the city of Jerusalem were not urbanized during the Persian period (Lipschits 2006: 29–30). Persian interest in the hill country appears to have been for extracting taxes and not for developing marketplaces, long-distance trade, or for encouraging productive innovation. These data suggest that there are some indicators that market exchange may have increased during the Persian period through “top-down” policies and “bottom-up” activities. At best, it can be suggested that the advancement of markets was most likely uneven and that the data are inconclusive concerning whether agrarian economies experienced sustained growth during the 5th–4th centuries b.c.e. Up to this point, our process of model building has included the following aspects. First, market exchanges were complex and dynamic, and therefore an economic model must consider whether a household was predominantly involved in dyadic and/or polyadic exchanges and whether they had access to urban centers and trade routes. Once we study the variety of market exchanges and learn how geography and climate contribute to the type of exchanges, then we can integrate that aspect into how the Persian government and religious elite contributed to economic activity in exchange for goods and services. This feature requires that we do not view political and religious systems as merely “takers” of economic goods and services but also as contributors to agrarian economies. Finally, our model building integrates the causal factors in market development consisting of collective action in which state builders replaced local traditional elites, agricultural intensification consisting of crafts and pastoralism, and urbanization. What becomes apparent is that a model of agrarian economies must take into account an array of different production and distribution patterns.

Remodeling the Models Therefore, an attempt to find the market system is an oversimplification of how markets are organized. This leads me to pose the following question: how feasible would it be to propose a general economic framework within the complexity of multifactorial analysis without 25.  Halstead and O’Shea (1989: 3) suggest that the “buffering mechanisms” designed to lessen the impact of variability can be grouped into four categories: mobility, diversification, physical storage, and exchange.

Methods and Models for Understanding Ancient Economies

15

abandoning all general theories? My goal here is modest. Understanding the limitations of the data we now have, I simply wish to propose three potentially productive ways toward developing a general model. First, in order to address Economic Activity 1, the basic unit of analysis needs to be determined. Some economists have suggested that to understand ancient economies better we begin with considering command (Dearman 1988: 9–52, 133–34), rent capitalism (Lang 1983: 118–19), tribute, or temple economies. According to Kenneth Hirth, households are the primary unit for studying the origins and organization of marketplaces, for they “employ a wide range of subsistence strategies and non-market transactions to meet their needs, including non-maximizing ones” (Hirth 2010: 232). This perspective addresses the objections leveled against Finley and others,26 who mistakenly regard all markets and marketplaces as being the same. Households engage in reciprocal exchanges that often occur as dyadic relationships between kinfolk in non-marketplace settings through a socially meaningful context. Therefore, it can be said that households generate and mobilize resources for the benefit of kinship groups and adopt a variety of strategies to minimize risk and allow for diversification in production. According to this matrix, the primary purpose of exchange is “use value,” not “exchange value.” “Exchange value merges where competitive buyers and sellers purchase goods for resale. Use value denotes a more individual relationship between goods and their consumers and depends on the goods’ specific utility for their consumer” (Meikle 2002: 233). Ancient agrarian economies predominantly employed a system of use value, not exchange value. Studying households as the primary unit and exploring their use of goods and services is accomplished archeologically and textually by considering the range of goods, engagement of work, production specialization, and exchange of commodities. Economic Activity 2 involves the building of a framework that integrates evidence of local agrarian economies into the new Persian and religious structures that surely generated social and religious tension. What needs to be considered is how agrarian economies interacted with these new political and religious realities. Economic exchanges cannot be viewed as secondary in importance to social, religious, or political agendas. Hirth reminds us that “[e]conomic exchanges are always part of, and exchanged alongside, social transactions because they are visible reminders that relationships have been established and need to be maintained” (Hirth 2010: 229; emphasis original). A comprehensive 26.  For instance, Finley (1973: 29) states that “we must concentrate on the dominant types, the characteristic modes of behavior.”

16

Marvin Lloyd Miller

study is needed that includes how political and religious structures may vary economic activity depending on the bargains struck between Persian governors, religious elite, and the subordinates. Margaret Levi and others have suggested that, “when taxpayers or other civic groups are endowed with few resources with which to bargain, principals (rulers or other key decision makers) are predicted to provide few public goods, to exercise more coercive modes of societal domination, and to lack accountability in society” (Blanton and Fargher 2010: 211).27 Models of how households produced goods and services cannot achieve a level of sophistication if we fail “to understand how political, economic, and social activities and institutions fit together and produce the great variability of outcomes we see in history” (Granovetter 2002: 88). In view of Economic Activity 3, a characteristic of the economic development in agrarian economies must recognize that the evolution of economies is seldom unidirectional and tidy; therefore, a broad approach to ancient economies is imperative. In order to accomplish this task, both the performance and structure of markets and market exchanges need to be considered. The political and religious structures may have assisted the delivery of goods and services by providing transportation, infrastructure, coinage, and markets, but these systems may have also hindered the performance of market exchange. We must therefore be “prepared to use our imaginations in order to reconstruct the pressures and circumstances that led to their development, [for] to focus on the detailed history of systems and institutions is to create a viable and useful form of economic history” (Davies: 2001: 155). However, combining these three considerations into one volume is a challenge. The discussion thus far on ancient economies shows that the topic is a messy chowder, a pot simmering with approaches and elements familiar to all methods, such as money, governments, exchange, taxation, and so forth. In this study, we will consider several indicators that suggest how the economy in ancient Judah might have developed. Not all aspects of the foregoing discussion can be included in one volume. Therefore, including a wide range of literary approaches, both from the Hebrew Bible and other sources, as well as archaeological advances will provide the framework for this study. The group of essays included in this volume originated in two international conferences titled “Agrarian Economies in Depopulated Areas in Persian Antiquity.” The conferences were organized in the form of special sessions at the 2013 meeting of the Canadian Society of Bibli27.  Braswell (2001: 130) states that in a monopolistic exchange the “relationship between rural producers and middlemen is regulated by the elite, and retailing follows market principles.”

Methods and Models for Understanding Ancient Economies

17

cal Studies (CSBS) in Victoria, Canada, and a workshop held later that year in Leipzig, Germany, within the context of the 2013 meeting of the European Association of Biblical Studies. The idea for the study arose from a conversation that Gary Knoppers and Ehud Ben Zvi had in 2010 in Montreal, Canada, expressing the lack of attention that had been given to ancient economies in depopulated areas in Yehud and its surrounding regions during the Persian period. One of the aims of these conferences was to rectify, however partially, this neglected period and topic in historical and archaeological research, with the hope that further debates and publications will emerge. Although the title for the book was changed to The Economy of Ancient Judah in Its Historical Context to reflect the broader scope of the presentations, the initial purposes we had in mind remain unaltered. Part of the task of the contributors was not to assume capitalist structures in their treatment of ancient economies but, rather, to place their economic inquiry in a social and historical world. The first critical area of focus was to understand how economic life was embedded in social relations as portrayed in the Hebrew Bible. When we are considering an economic structure, the role of the institution of a family or household is inescapable. Therefore, the first two essays begin with a market exchange that fits with Economic Activity 1 (see above) and features the household and patron-client relationships, respectively. These economies are structured predominantly around dyadic and polyadic exchanges. Ben Zvi’s detailed analysis of Prov 31:10–31 sheds new light, not only on a remembered wise wife, but also on economic and social conditions during the late Persian/early Hellenistic period. The description of the wise wife was constructed within a community as an ideal site of memory that reflects the nonchaotic, peaceful world of royal ideology. As indicated earlier, most economic systems function by a combination of market exchanges. Accordingly, Ben Zvi describes the wise wife as seeing herself as part of a social whole and participating in agrarian aspects of the economy combined with external markets. As this lead essay demonstrates, wealth may indeed come from many sources. Proverbs 31 sheds light on economic exchanges and attitudes among affluent households, and Ben Zvi shows that a household cannot be viewed in isolation, as if separated from other economic influences, but instead must be understood as participating in other modes of exchange. There is little difference between being the head of a household and being the head of a clan—or to rephrase it, between being in a kinship and being in a patron-client relationship. A difference that can be noted is that the patron-client relationship is hierarchical and frequently

18

Marvin Lloyd Miller

crosses clan lines. Knoppers stresses this unequal relationship between market exchanges in his essay concerning the shifting sands of the perception of Solomon. According to Knoppers, the author of Chronicles reimaged Solomon’s relationship with Huram to emphasize a patronclient relationship of unequal partners. For example, in the book of Kings, Solomon allows Huram to set a price for the building materials, but in Chronicles Solomon announces to Huram that he has assigned to Huram’s servants a variety of gifts. Given the nature of treaty partnership and gift-giving, the way in which this exchange is portrayed has elements of propaganda. What Knoppers presents is that the language of patron-client in Chronicles is noticeably consistent, reflecting a change in attitude toward one of Yehud’s early heroes presented in this literary work. From an emphasis on the household and patron-client relationship, we now consider a wider market exchange, expressed above by Economic Activity  2. This exchange features the temple as an economic structure. Jonker uses Ronald Boer, who emphasizes allocative and extractive elements of an ancient economy, as a paradigm for understanding 1 Chronicles 1–9.28 The careful study of the genealogical lists in 1 Chronicles by Jonker reflects the way that this textual data can suggest where the seat of power may rest. He suggests that the temple-city complex acted as an axle of a wheel to which the different village communes were connected like the endpoints of spokes. From the textual data found in 1 Chronicles 1–9, Jonker is able to reflect on how these lists signal an emphasis on a royal connection with the cult in Jerusalem, which stands in tension with the Persian imperial center of power. The final contribution to traces of economic activities embedded in the Hebrew Bible is by Altmann. Here the net is cast much wider. He compares the market trade found in Babylon with Yehud. Altmann contends that the Babylonian temples emerged out of a structurally redistributive economy and that their administrative framework allowed individuals to be land- and slaveholders. Using Neh 13:15–22 as a test case, he employs New Institutional Economics to show that Yehud was not significantly involved in interregional market trade. Now we consider economic indicators from different domains of life, consisting of the arrival of newcomers, land allocation practices, reproductive practices, and evidence of local markets in Idumea. These topics reflect Economic Activity 3, which attempts to explain some causal factors in market development. It seems that there was a return of a portion of the Jewish diaspora to Yehud from Babylon during the Per28.  For a recent treatment on ancient economies by Boer, see Boer 2015.

Methods and Models for Understanding Ancient Economies

19

sian period—how many returned is still hotly debated. The arrival of newcomers frequently alters the balance of power and thus the activity in markets. Guillaume emphasizes that the motto “be fruitful and multiply” applied to the newly arrived Jews and was a motivation for demographic increase. This change in population strained relations with local, long-term farmers, who resented the new situation. However, the arrival of newcomers represented fresh openings for commerce for the most enterprising people. Markets evolve not only when there is an influx of new stakeholders but also when there is an intensification of agricultural production. Fried uses Nehemiah 5 as a text that illuminates the economic relationship of Yehud with Persia. Using the concept of ilku service derived from the Akkadian sources, she shows that the land was parceled out into fiefs on which ilku service was required in addition to rent. This service refers to the payment of labor to the military or other elements of the Persian administration that is owed on land given by a king or governor. It is plausible that Yehudite markets developed and production increased because of the Persian administration’s need to provision their army and seek access routes to other areas. The reproductive practices and policies of the Persian Empire is the topic undertaken by Wiesehöfer. Markets are a product of population growth and urbanization. The practices of monogamy, polygyny, and harems differed according to social strata and along cultural lines—different classes and empires held different perspectives. Although large urban centers were absent in ancient Yehud, it is informative to have a close look at the royal Achaemenid policy of reproduction. The last contribution in this section is Edelman’s focus on the economy and administration of rural Idumea. Rather than basing her conclusions on literary texts, she focuses predominantly on ostraca. She adds valuable new information concerning Persian policy regarding land ownership, tenancy, and taxes and contributes to the conversation on the routine workings of the Persian administration. From the information she has gathered, there is evidence for different forms of land holdings in the region. What has been shown in the essays in this section is that management of the lands took place through the direct supervision of the temple, palace, or Persian administration and that different circumstances required different levels of the temple or state’s economic involvement. Finally, two essays concern how archaeological evidence can be used to determine whether an economy was in decline or on the rise. Thareani divides her essay into two major parts: forces of regeneration during the 8th–7th centuries and forces of decline at the end of the

20

Marvin Lloyd Miller

7th–early 6th centuries. Part of her emphasis concerns the economic resources and systems that were present during times of economic rise and, consequently, how local elites served an imperial power. She offers various explanations to explain the colossal shift to a devastating time of economic decline. Her essay emphasizes that major economic regimes are part of the economic production of the ancient Levant and that there existed a pattern of expansion and collapse as each despot tried to bring this region under its control. Lipschits narrows the focus and uses archaeological evidence to describe the origin and history of the Persian period rural settlements and districts in Judah. In order to arrive at his conclusions, he considers the agricultural and economic potential in addition to the size and role of the districts. What becomes clear is that during the Persian period each district served different roles in the overall provincial system. The unique contributions from the authors of this volume use various methods to provide a sophisticated approach to ancient economies. Economic activity in the ancient Levant is varied and must be understood in its social setting. The aim of this volume is not to construct an economic model that can be adapted to every situation, but to emphasize the different ways that economic activity can be understood in Yehud under Persian rule.

Bibliography Abbott, D.; Smith, D. A.; and Gallaga, E. 2007 Ballcourts and Ceramics: The Case for Hohokam Marketplace Exchange in the Arizona Desert. American Antiquity 72: 461–84. Axelrod, R. 1997 The Complexity of Cooperation. Princeton, NJ: Princeton University Press. Blanton, R. E., and Fargher, L. E. 2010 Evaluating Causal Factors in Market Development in Premodern States: A Comparative Study, with Critical Comments on the History of Ideas about Markets. Pp. 207–26 in Archaeological Approaches to Market Exchange in Ancient Societies, ed. C. P. Garraty and B. L. Stark. Bolder: University Press of Colorado. Boer, R. 2015 The Sacred Economy of Israel. Library of Ancient Israel. Louisville: Westminster John Knox. Braswell, G. E. 2010 The Rise and Fall of Market Exchange: A Dynamic Approach to Ancient Maya Economy. Pp.  127–40 in Archaeological Approaches to

Methods and Models for Understanding Ancient Economies

21

Market Exchange in Ancient Societies, ed. C. P. Garraty and B. L. Stark. Bolder: University Press of Colorado. Cartledge, P. 2001 The Economy (Economies) of Ancient Greece. Pp. 11–32 in The Ancient Economy, ed. W. Scheidel and S. von Reden. Edinburgh: Edinburgh University Press. Davies, J. K. 2005 Linear and Nonlinear Flow Models for Ancient Economies. Pp. 127– 56 in The Ancient Economy: Evidence and Models, ed. J. G. Manning and I. Morris. Stanford, CA: Stanford University Press. Davis, E. F. 2009 Scripture, Culture, and Agriculture. Cambridge: Cambridge University Press. Davis, J. 1992 Exchange. Minneapolis: University of Minnesota Press. Dearman, A. 1988 Property Rights in the Eighth-Century Prophets: The Conflict and Its Background. SBLDS 106. Atlanta: Scholars Press. Edelman, D. V. 2006 Tyrian Trade in Yehud under Artaxerxes I: Real or Fiction? Independent or Crown Endorsed? Pp. 207–46 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. Eisenstadt, S. N. 1963 The Political System of Empires: The Rise and Fall of the Historical Bureaucratic Societies. New York: The Free Press of Glencoe. Elayi, J. 1980 The Phoenician Cities in the Persian Period. JANES 12: 13–28. 1982 Studies in Phoenician Geography during the Persian Period. JNES 41: 83–110. Ephʿal, I. 1998 Changes in Palestine during the Persian Period in Light of Epigraphic Sources. IEJ 38: 106–19. Eskenazi, T. C. 2006 The Missions of Ezra and Nehemiah. Pp. 509–29 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. Evans, D. J.; Vos, R. J.; and Wright, K. P., eds. 2003 Biblical Holism and Agriculture: Cultivating Our Roots. Pasadena, CA: William Carey Library. Faust, A. 2003 Judah in the Sixth Century b.c.e.: A Rural Perspective. PEQ 135: 37–53. Fick, G. W. 2008 Food, Farming, and Faith. SUNY Series on Religion and the Environment. Albany: State University of New York Press.

22

Marvin Lloyd Miller

Finley, M. I. 1973 The Ancient Economy. Los Angeles: University of California Press. Garraty, C. P. 2010 Investigating Market Exchange in Ancient Societies: A Theoretical Review. Pp.  3–32 in Archaeological Approaches to Market Exchange in Ancient Societies, ed. C. P. Garraty and B. L. Stark. Bolder: University Press of Colorado. Granovetter, M. 1985 Economic Action and Social Structure: The Problem of Embeddedness. American Journal of Sociology 91: 481–510. 2005 Comment on Liverani and Bedford. Pp. 84–88 in The Ancient Economy: Evidence and Models, ed. J. G. Manning and I. Morris. Stanford, CA: Stanford University Press. Guillaume, P. 2012 Land, Credit and Crisis: Agrarian Finance in the Hebrew Bible. Sheffield: Equinox. Halstead, P., and O’Shea, J. 1989 Introduction: cultural responses to risk uncertainty. Pp.  1–7 in Bad Year Economics: Cultural Responses to Risk and Uncertainty, ed. P. Halstead and J. O’Shea. New Directions in Archaeology. Cambridge: Cambridge University Press. Herodotus 1971 Herodotus, with a Translation. Translated by A. D. Godley. 4 vols. LCL. Cambridge: Harvard University Press. Hirth, K. G. 2010 Finding the Mark in the Marketplace: The Organization, Development, and Archaeological Identification of Market Systems. Pp 227– 47 in Archaeological Approaches to Market Exchange in Ancient Societies, ed. C. P. Garraty and B. L. Stark. Bolder: University Press of Colorado. Hoglund, K. G. 1999 The Achaemenid Context. Pp. 22–53 in Second Temple Studies 1: Persian Period, ed. P. R. Davies. JSOTSup 117. Sheffield: JSOT Press. Lang, G. 1983 Monotheism and the Prophetic Minority: An Essay in Biblical History and Sociology. Sheffield: Almond. Lipschits, O. 2003 Demographic Changes in Judah between the Seventh and Fifth Centuries b.c.e. Pp. 323–76 in Judah and the Judeans in the Neo-Babylonian Period, ed. O. Lipschits and J. Blenkinsopp.  Winona Lake, IN: Eisenbrauns. 2006 Achaemenid Imperial Policy, Settlement Processes in Palestine, and the Status of Jerusalem in the Middle of the Fifth Century b.c.e. Pp.  19–52 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns.

Methods and Models for Understanding Ancient Economies

23

Mann, M. 1986 The Sources of Social Power, vol. 1: A History of Power from the Beginning to a.d. 1760. Cambridge: Cambridge University Press. Mankowski, P. 2000 Akkadian Loanwords in Biblical Hebrew. Winona Lake, IN: Eisenbrauns. Meikle, S. 2002 Modernism, Economics and the Ancient Economy. Pp.  233–50 in The Ancient Economy, ed. W. Scheidel and S. von Reden. Edinburgh: Edinburgh University Press. Polanyi, K. 1957 The Economy as Instituted Process. Pp. 243–70 in Trade and Market in Early Empires: Economies in History and Theory, ed. K. Polanyi, C. W. Arensburg, and H. W. Pearson. Glencoe, IL: Free Press. Rostovtzeff, M. 1936 The Hellenistic World and its Economic Development. American Historical Review 41: 231–52. Scott, J. C. 1985 Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven, CT: Yale University Press. Sahlins, M. 1972 Stone Age Economics. Chicago: Aldine-Atherton. Smith, A. 1776 An Inquiry into the Nature and Causes of the Wealth of Nations. Repr. Chicago: University of Chicago Press, 1976. Smith, C. A. 1976 Exchange Systems and the Spatial Distribution of Elites: The Organization of Stratification in Agrarian Societies. Pp. 309–74 in Regional Analysis, vol. 2: Social Systems, ed. C. A. Smith. New York: Academic Press. Smith, R. L. 2009 Premodern Trade in World History. London: Routledge. Stanish, C. 2010 Labor Taxes, Market Systems, and Urbanization in the Prehispanic Andes: A Comparative Perspective. Pp. 185–205 in Archaeological Approaches to Market Exchange in Ancient Societies, ed. C. P. Garraty and B. L. Stark. Bolder: University Press of Colorado. Stern, E. 1990 The Dor Province in the Persian Period in the Light of the Recent Excavations at Dor. Transeu 2: 147–55. 2001 Archaeology of the Land of the Bible, vol. 2: The Assyrian, Babylonian, and Persian Periods 732–332 b.c.e. New York: Doubleday. Wagner-Tsukamoto, S. 2009 Is God an Economist?: An Institutional Economic Reconstruction of the Old Testament. New York: Palgrave MacMillan. Yamauchi, E. 1990 Persia and the Bible. Grand Rapids, MI: Baker.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31 as a Source for Reconstructing Aspects of Thought and Economy in the Late Persian / Early Hellenistic Period Ehud Ben Zvi

University of Alberta

Introduction Studies conducted in the last decades on Prov 31:10–31 have contributed a great deal to a better understanding of the text and the portrayal of the successful, wise, worthy wife at its very center.1 They have dealt with literary composition and structure (Lichtenstein 1982: 202– 11; Hurowitz 2001: 209–18)2 and addressed, inter alia, gender (Camp 1985: 90–94, 186–208; Fischer 2005: 237–53; cf. Valler 1995: 85–97), imagery (Szlos 2000: 97–103; Novick 2009: 107–13), anthropology (Lang 2004b: 188–207), socioeconomics (Yoder 2001; 2003: 427–47), text criticism (Rofé 2002: 145–55),3 form criticism (Wolters 1988: 446–57; Nwaoru 2005: 41–66), and various combinations of these matters.4 In addition, they have been informed by comparative studies with ancient Near Eastern texts (Hurowitz 2005: 221–34), classical texts (esp. Xenophon, Oeconomicus; Waegeman 1992: 101–7; Lang 2004b),5 Ottoman households (Lang 2004a: 140–57), Yoruba oriki/recitals (Nwaoru 2005), and Indian parallels (Luke 1991: 131–32). My contribution to this vast area of studies is from the perspective of a historian of the world of the thought and memory of ancient Israel. 1.  For an excellent study of this pericope, which also includes a survey and critique of most of these studies, see Fox 2009: 882–917. 2.  It may be mentioned that readings of Prov 31:10–31 as being closely linked to Prov 31:1–9 have a long history. Yefet ben ʿElī (10th century) maintained that the author of Prov 31:10–31 was the mother of Lemuel—whom he identified with Bathsheba. The proposal for a female authorship of Prov 31:10–31 is noteworthy. On Yefet ben ʿElī’s reading of this pericope, see Sasson 2013: 175–76. 3.  Compare with Rofé’s contribution by the same title published in Hebrew in Talshir, Yona, and Sivan 2001: 382–90. 4.  In fact, all the works mentioned thus far could easily have been included in more than one category. This is the case because, in reality, none of these categories excludes the others, and scholars usually deal with multiple issues and approaches. 5. On Xenophon’s Oeconomicus, see, e.g., Pomeroy 1994.

27

28

Ehud Ben Zvi

Moreover, since the world of thought does not exist in a vacuum, totally unrelated to historical contingencies, and since even imagination must emerge from a “real” world, I conclude with some observations about the ways in which this ideal image may shed light on economy and society in Yehud. It is obvious that the imagined, remembered, and certainly utopian (from the perspective of the readers) the ‫ אׁשת־חיל‬portrayed in Prov 31:10–316 did not provide a representative portrayal of the activities of historical, average wives in the late Persian / early Hellenistic period in Yehud/Judah—or for that matter, in any period.7 Instead, it constructed within the community an ideal site of memory. The ‫ אׁשת־חיל‬became the embodiment of an exemplar.8 It served processes of socialization within the relevant social group and as a socially approved, guiding lighthouse for single men navigating the sea of marital matchmaking, and likely also for future brides (Crook: 1954: 137–40; Luke 1991; Nwaoru 2005; Fox 2009: 905) within the same group, learning about what they should do to become wives who both were a treasure and created a treasure 6.  As the title of this contribution shows, I am rendering ‫ אׁשת־חיל‬a “successful, wise, worthy wife.” This is just an attempt to convey a more substantial portion of the semantic realm of ‫אׁשת־חיל‬. Rofé (2002) influences this translation to some extent. Many other renderings have been offered, e.g., “capable wife/woman,” “valiant wife/woman,” “wife/woman of valor,” “wife/woman of substance,” “wife/woman of worth,” “wife/woman who is a treasure,” “wife/woman of strength,” “excellent wife/woman,” or one may even think of “wonder-wife/woman.” Within the pair “wife”–“woman,” the preferred term is often “woman,” but since the married status of the ‫ אׁשת־חיל‬plays an important part in her characterization and in the ways in which she was imagined and remembered in the late Persian/early Hellenistic period, I find the term “wife” to be historically more precise. This said, I am aware of concerns related to today’s use of biblical text—in settings other than academic attempts to reconstruct a historical past—that have been raised about translations such as the one I am suggesting. See, e.g., Yilibuw 1996: 30–32. The goal of this contribution, however, is to advance a (construction of a historically reasonable) reconstruction of the ancient world of thought. Within the world of thought (and memory and imagination) of the literati of Persian or early Hellenistic Yehud/Judah, the ‫ אׁשת־חיל‬was certainly imagined as a married heterosexual woman and a mother as well. For some lists of translations and brief discussion, see, e.g., Hurowitz 2005: esp. p. 221. 7.  The point is actually partially conceded by the text itself, as it frames the beginning of the pericope with a rhetorical question. On this issue, compare, for instance, Fox 2009: 890–91; Hausmann 1992: 262; and Fischer 2006: 149. 8.  The community reading and rereading the text remembers her, her attributes, actions, and her world. As the process continues, the ‫ אׁשת־חיל‬becomes a site of memory, and I would add, an important site of memory within the community. See below.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

29

and to inculcate a direct correspondence between the two—namely, being and creating a treasure. Before we explore some aspects of the underlying world of thought reflected and communicated by ‫אׁשת־חיל‬, two preliminary considerations are in order. First, there is the matter of dating the relevant text. Obviously, if it did not exist during the late Persian / early Hellenistic period in Judah/Yehud, it cannot help us shed much light on this period. But there is a general agreement on dating this text within that range.9 Of course, there are many attempts to narrow the range for the composition of the text, but the arguments in favor of more-precise dates are not necessarily conclusive (Fox 2009: 899–902 ). In any event, for the present purposes, given the long-term, basic continuity of the relevant socioeconomic setting, a wide range such as “late Persian / early Hellenistic” does not represent a substantial problem.10 9. See Yoder (2001), who dates MT Proverbs 1–9 and 31:10–31 to “a date between the beginning of the sixth century b.c.e. and the end of the third century b.c.e.— most likely sometime in the Persian period” (p. 38). Lang (2004b) suggests that the text may be “roughly contemporaneous with Xenophon’s Oeconomicus” and in any case no earlier than the 5th or the 4th centuries b.c.e. (2004b: 188–89; 2004a: 140). Wolters (1985: 577–87) proposes a date in the 3rd century (but notice also Fox’s critique on purported crucial evidence for the dating [Fox 2009: 897]; and the comments in Yoder 2001: 33). Finally, one may notice that Waegeman suggests a date no later than the 2nd century b.c.e. (Waegeman 1992: 101). In general, positions concerning these more narrow temporal ranges tend to be argued on matters such as the connection between Prov 30:10–31 and Proverbs 1–9, observations about similar social (and clearly gendered) settings in the Persian period or in Xenophon’s Oeconomicus (see, e.g., 3.10–15; 6.9; 7.1–41; 8.11–17; 10.10–13), and some linguistic considerations (among other matters). For some additional considerations, see n. 21 below. For the position that Prov 31:10–31 is a premonarchic text, see Lyons 1987: 237– 45. Lyons contends that the basic “pioneer” conditions of the premonarchic period that she associates with this text were also present in the Persian period but decides to support a date in the premonarchic period because of her position that “the extended family did not reappear a self-sufficient economic unit” at any time in Judah, including the Persian period, after the beginning of the monarchy (Lyons 1987: 241). Neither the crucial position nor the related dating has received much support in current research. 10.  What about proposals to date the text outside these general parameters? Not only is there is no evidence supporting a Hasmonean date for the text, but both Ben Sira’s reliance on Proverbs and commonly suggested dates for LXX Proverbs make such a date less likely. See Cook 1993: 25–39; 1999: 448–61; 2008: 67–85. A monarchic date for Prov 31:10–31 is also difficult, given, inter alia, the multiple interrelations between this text and other texts in the book. But, for the present purposes, even if one were to argue that there was some forerunner of Prov 31:10–31, the ‫אׁשת־‬ ‫ חיל‬evoked through acts of reading Proverbs (again notice, how the relevant text

30

Ehud Ben Zvi

The second preliminary observation is that, although (a) the wife in Prov 31:10–31 evoked images of Lady Wisdom (e.g., Prov 31:10; cf. 3:15, 8:11), and (b) was likely imagined as partially embodying some of the attributes of Lady Wisdom, and (c) at times an ad hoc potential (though limited) overlap between the two was connoted (e.g., Prov 31:25–26) for the sake of stressing how good she was, the fact remains that the ‫ אׁשת־חיל‬was still construed and remembered primarily as a wise wife, not an elevated, above-human wife. Unlike the ‫אׁשת־חיל‬, Lady Wisdom is not actually and exclusively married to a particular man. Moreover, and most significant, no matter how much the husband of the ‫אׁשת־חיל‬ was honored among his peers, he certainly did not marry the “daughter of Yhwh” who stood before creation (cf. Prov 8:22–36).11 The ‫אׁשת־‬ ‫ חיל‬was, after all, an ideal but very human, wife, mother, 12 manager, and entrepreneur whose memory served as an exemplar for other (human) women—not a kind of goddess or just an allegory for Wisdom (McCreesh 1985: 25–46).13 Since the ‫ אׁשת־חיל‬was remembered as an ideal human wife, she could embody and communicate what the community—or better, what the well-off sector of the community—considered ideal economic beis deeply and carefully interwoven with others) could not have preceded the book itself, which in any case leads us back to the Persian–early Hellenistic period. 11.  Of course, when the text is read with very different “lenses,” other meanings appear, e.g., when the ‫ אׁשת־חיל‬is the Church and Christ is her husband (see Wright 2005: 186–87), or as a type of Mary (a quite common approach, even today in some religious circles), or the “torah” (see Midrash Proverbs—implicitly, Yhwh is her husband), or “Practical Wisdom” or “matter” that serves “Philosophical Wisdom,” or “the intellect” (see Ralbag, Proverbs 31; ‫)החומר המשרת אל השכל שרות שלם‬, or Israel (Pesiq. Rab. 31, and Zohar 3.42.b), or the Shekinah and Queen Sabbath (as often understood in traditional Jewish liturgy, influenced by Kabbalistic traditions), or any number of biblical women (e.g., Sarah—passim in rabbinic sources, the wife of Noah, Midrash Eshet Hayil at the beginning), or even, at times, men such as Moses, or the ideal student of Torah. According to Saadia Gaon, the text recounts “the attributes of [excellent] men” (though metaphorically), whereas according to Yefet ben ʿElī, it recounts the attributes of excellent people, whether men or women. See Sasson 2013: 173–76. On rabbinic sources, see Valler 1995; and on general Jewish traditional understandings, see Fox 2009: 905–7; and bibliographical references cited in these two works. For a discussion of the place of the Eshet Hayil in contemporary Jewish, liberal and feminist liturgy, see Falk 1999: 451–52. 12.  Certainly not to be identified with the only woman explicitly characterized as an ‫ אׁשת־חיל‬in the Hebrew Bible—namely, Ruth—when she was still destitute (Ruth 3:11). Compare and contrast with Goh 2014: 487–500. 13.  For a different take on the relation between Lady Wisdom and the ‫אׁשת־חיל‬ along with a relevant exploration of the concept and role of “personification,” see Camp 1985: 96–97, 186–222. For a critique of McCreesh’s position and arguments, see Gilbert 1999: 147–67.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

31

havior at the level of a single household. In fact, thinking about and through the ‫ אׁשת־חיל‬was a way for the community to explore and express its views about ideal economic activity at the level of a single affluent household. This level is particularly important, not only for all the usual reasons for taking into account the activities of agents other than central political leaders or whatever is said about them, but also because it is most likely that the individual household was and was considered by the community to be the basic social and economic unit.14 In what follows, I will discuss some aspects of these concepts and their potential significance.

Economic Power Brings Honor and the Strongest Endorsement for the Pursuit of Profit and for Profit Itself The ‫ אׁשת־חיל‬produced material gain (v. 11) for her husband and household. Not only is the wife’s strenuous pursuit of gains—that is, her pursuit of profit (and profit itself)—glorified within the world of thought evoked by the text, but her acquisition of wealth (i.e., economic capital) provides her and him with additional honor, social capital, and power rather than vice versa. This is a rhetorical and mnemonic (at least) partial reversal of the expected situation: namely, that power is the source of wealth, rather than vice versa. In fact, it is often maintained that, whereas wealth is the source of power in capitalistic societies, the opposite was true in more traditional societies (Amin 1991: 349–50). Given the relation between power and social capital, the transformation of her profits into her husband’s/household’s social capital/ honor, the observation made above about wealth’s being a source of power holds true in the “story” of the ‫אׁשת־חיל‬, even if one were to argue that the community did not construe this wife as one who pursued profit for profit’s sake and wealth for wealth’s sake. The same holds true when attention is explicitly drawn to the patronage system headed by the ‫ אׁשת־חיל‬herself (not her husband) and certainly not restricted to “women and children,” but to the poor and ָׂ ‫) ַ ּכ ָפּּה ָפ ְּר‬. Producing needy in general (v. 20; see ‫ֶביֹון‬ ְ ‫שה ֶל ָענִי ְויָדֶי ָה ִׁש ְלּחָה ָלא‬ and accumulating economic power was conceptualized as leading to increased social capital and power, rather than vice versa, and these matters were explored and embodied in the ‫אׁשת־חיל‬. The reference to her husband and his increased honor (v. 23) may imply an additional redistribution/patronage system, this time managed by the elders of the city. 14.  There is no house of the (local) king to compete for socioeconomic power in the Persian period, and it is unclear whether the Jerusalemite temple played a direct and central role in the production of wealth within the province.

32

Ehud Ben Zvi

These references are not insignificant. They serve an ideological need to “normalize” or “balance” her pursuit of profit and wealth (well beyond that required for covering sufficiently the needs of her household) by transforming them into a pursuit that serves purposes other than increasing wealth and by suggesting, implicitly, a potential redistributive system. To them, one may add the likely later shift in the characterization of the ‫ אׁשת־חיל‬from a “wise” woman (see LXX Prov 31:30) to a woman who “fears Yhwh” (MT), which in another way “normalizes” her memory.15 But, to be sure, remembering the ‫ אׁשת־חיל‬meant assigning much more social mindshare to her work and its constant pursuit of accumulation of wealth within the boundaries of her household and to its partial, glorification, rather than to the social redistribution of the wealth accumulated through an ongoing, constant pursuit of profit, and nowhere is it stated that the main reason for all her labors was to fund the poor and needy. A final observation: this reversal of expectations—namely, that wealth is construed as leading to power, rather than vice versa—may well be only partial. The ‫ אׁשת־חיל‬and her husband and those who identified themselves as the addressees of the opening rhetorical question were imagined as members of a high socioeconomic group that already owned considerable resources. One may argue, then, that the society in Yehud as a whole had a power structure that, although it was allowed material wealth to bring about relatively minor shifts of social clout/power among its top echelon, due to restricted social mobility reflected a system in which as a whole, power (i.e., the already existing power of the top echelon of household) led to its increased wealth rather than vice versa. This may well be correct, but one must keep in mind the case of Ta(pe)met and her daughter Yehoshima, which represents a very well attested instance of social mobility in Elephantine.16 Such cases might have been uncommon and, in any event, one cannot extrapolate from Elephantine to Yehud. Nevertheless, they raise questions about any claims that social mobility fueled by acquisition of some relative measure of wealth was

15.  An additional potential case of “normalization” that is clearly not a late addition may be discerned in v. 31, if one follows Fox and understands the texts as envisioning “two kinds of recognition: material and verbal” (Fox 2009: 899). 16.  Compare with, for instance, the situation at the time of the mother’s wedding as reflected in B 36 = TAD B3.3 with that of the daughter in B 41 = TAD B3.8. (Social mobility through outstanding service to the crown, esp. but not exclusively in the military, had long been a common path to social mobility in the region and has been well attested in, for instance, Egypt. But we are referring here to a social mobility that is not dependent on the decision of important figures in the court.)

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

33

an impossibility in Yehud.17 It is worth noting in this context that Prov 30:10–31 at the very least creates the illusion that any household led by such an ‫ אׁשת־חיל‬will do well (see pp. 38–39 below) and that remembering her was still remembering that an increase in profit through the judicious management of a wife was likely to lead to increased social power.

Pursuing Profit Is Wise and It Involves Wisdom about How to Make a Profit: Articulating the Construction of an Economy How did the imagined, remembered, and, ideally, to be imitated as much as possible ‫ אׁשת־חיל‬pursue her profit and increase her wealth? The text suggests that this question is important. In fact, within the multiple sites of memory encoded in the authoritative repertoire of the community, thinking about and through the image of the ‫ אׁשת־חיל‬provided one of the few potential grounds for exploring these matters at the most-relevant level for the community—that is, the single household. To be sure, the text refers to the proper administration of resources (esp., food and clothing, including the appropriate purchase and production of goods for household use, e.g., food, v.  15; and clothes for herself that communicated her high status, v. 22).18 But in terms of creating a profit, the stress is first on the sale of value-added products. In particular, she produces clothes and sells (luxury?) linen garments and sashes (or loincloths) to merchants and with her earnings she buys good fields and plants vineyards (v. 16) that, in turn, are supposed to increase her profit. As profit and wealth accumulate, she is able to produce even more profit and wealth. Her position is secure, and she may “laugh” (v. 25) at the time to come (i.e., whatever the future may bring). 17.  It is worth stressing that there was (some degree of) social mobility in both Egypt and Mesopotamia, and not only at the top (e.g., “commoners” who became royal or high-level military or administrative leaders with “humble” origins). These matters are beyond the scope of this essay, but see, e.g., Steinkeller 1987: 73–115, e.g., p. 100; Stone 1999: 208; Frood 2010: 478–79; Vandorpe 2010: 159–79. For the present purposes, it suffices to note that assumptions about an airtight closure of the possibility of social mobility in Persian/early Hellenistic Yehud/Judah are problematic. 18. Imagining her making status clothes for herself makes sense within the world evoked by reading the text among the literati of the time but seems to have raised some uneasiness among some later readers. What about her husband? The LXX “normalizes” the text: “She duplicated cloaks for her husband and for herself clothes of fine linen and purple” (NETS). Another example of the LXX’s tendency to “normalize” the text appears in v. 21, which in the LXX reads, “Her husband has no concern for his household, when he spends time somewhere, for all that are hers are being clothed” (NETS). LXX Proverbs emerged in a society that was not identical to that of the original text and in which sensitivity to the partial “absence” of the husband likely led to additional, explicit textual “mitigation,” when compared with the MT.

34

Ehud Ben Zvi

The ‫ אׁשת־חיל‬was imagined as an entrepreneur, as an extremely industrious person who worked day and night (v. 18; compare and contrast with Josh 1:8). She was remembered as “seeking” ‫ דרשה‬wool and flax (v. 13; compare and contrast with Ezra 7:10), as a master of her trade (vv. 13, 19), a person well aware of quality control (v. 18), and a mistress who wisely and kindly taught her servant-girls to produce very good wares (v. 26; cf. 15). She was recalled as a wise trader, intelligent buyer of farmland,19 and a planter and owner of vineyards. A set of concepts about ideal economic behavior was thus shaped, formulated, and expressed. Of course, necessarily, the same holds true for a world of thought and attitudes concerning socioeconomic matters. Within this world, the image of a trade ship evoked strong positive responses, and trade in general not only was viewed in very positive terms but was construed as necessary for the wise management of well-off households. The sense of a Yehudite household’s ability to behave wisely in both the economic sphere and in (local and foreign) trade was construed and communicated. Needless to say, this represents a very different view from the commonly attested ancient (essentially) moralistic, aristocratic disdain for trade and its corrupting influence, as well as for profit produced from trade—all of which were seen as potentially destabilizing forces from the perspective of an established, land-based aristocracy.20 It is perhaps, not surprising that trade is viewed so highly in a text in which a wife is made the exemplar of productive, wealth-creating leadership.21 ָׂ ‫ָממָה‬ 19. Note the choice of words in v. 16—namely, ‫ּקחֵהּו‬ ּ ָ ‫שדֶה ו ִַת‬ ְ ‫ז‬, “she plans (/schemes) and executes.” Whatever she carefully plans, she executes. These are not impulsive but well-calculated purchases, and within this world of memory and imagery, there is nothing the other side can do to stop her. 20.  In terms of ancient Israel, see, for instance, the negative characterization of trade in the oracles against Tyre or other Phoenician cities (e.g., Isaiah 23; Ezekiel 27), but see also Nah 3:16 and see the negative reference to Babylon in Ezek 17:4, as a city of merchants. Note that there is no comparable negative, moralizing account of farming in any of the so-called oracles against the nations (OAN). On Ezekiel 27, see recently Wilson 2013: 249–62. This tendency existed in many other ancient Mediterranean societies. See the polemic title in Horden and Purcell, The Corrupting Sea: A Study of Mediterranean History (2000); and cf. Bang 2006: 56–58. It is worth noting that, in Babylonia, “There is no indication that enterprise was considered ‘dirty business’ or something to be delegated to an underling as in Roman times” (Wunsch 2005: 367–79). 21.  There is an element of “transgression” on all these matters. Transgressions or challenges to socially accepted viewpoints are often (construed as) gendered.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

35

Within the world evoked by the ‫אׁשת־חיל‬, productive reinvestment was considered a strong virtue, but significantly and not surprisingly, within an agrarian society and economy, land holdings were brought to bear. Within the world of the ‫אׁשת־חיל‬, wealthy, well-managed households were supposed to increase their land holdings through well-considered purchases. In this world, fields did not evoke images of ancestral inheritance (cf. 1 Kgs 21:3)22 but of goods up for purchase by successful, wise household leaders, including wives such as the ‫אׁשת־חיל‬.23 It should be noticed that, within the “story” of the ‫ אׁשת־חיל‬and the world it construes, there is nothing the seller can really do to stop her from buying the plot that she planned to purchase. She intelligently planned and industriously carried out her plan. A seller is obviously implied in this story and world, but not worthy of being explicitly mentioned or remembered, since he (or “she”?) had no significant role to play in the story, and his (or her?) perspective was deemed irrelevant.

Wives as Economic Agents and an Economy Imagined as the Arena for Wives’ Heroism Since one is to assume that men were also economic agents, and they also bought fields, the question of preference for the memory of a wife over a husband in the most memorable case of successful management of a household within the community cannot be avoided. Certainly, the Sitz im Buch of Prov 31:10–31 played a role in such a preference (see the multiple links between Proverbs 1–9, 31:1–9, and 31:10–31),24 and the same holds true for tendencies to recall and incorporate some of the attributes of Lady Wisdom in the ‫אׁשת־חיל‬. Prov 31:10–31 was read and the ‫ אׁשת־חיל‬was imagined and remembered in ways informed by the book of Proverbs as a whole and, vice versa, the way the book was read was informed by the concluding and memor­ able figure of the ‫אׁשת־חיל‬. Just as Lady Wisdom is a caring, reliable provider, so is the ‫אׁשת־חיל‬. Within this basic plot, there is not much room for a male provider. 22.  Note also Meyers’s observation: “In contrast with the detailed Pentateuchal legal materials dealing with restitution of property, there are no laws that regulate land transfer except for inheritance” (Meyers 1997: 20). This absence served to construct an ideal world in which these transfers are not worth thinking about or remembering. 23.  Contrast the trader with whom the ‫ אׁשת־חיל‬continuously collaborates and who is necessary for her creation of wealth. Traders and, for that matter, her “girls” (i.e., maidservants) are construed as being in a situation of ongoing interdependence with the ‫ ;אׁשת־חיל‬sellers of fields are not. 24.  For the former (i.e., between Proverbs 1–9 and 31:10–31) see, e.g., Yoder 2001; for the latter (i.e., between Prov 31:1–9 and 31:10–31) see, e.g., Hurowitz 2001.

36

Ehud Ben Zvi

The preference for a wife figure led (within the relevant cultural context) in turn to other preferences, such as an emphasis on clothing (and on the preparation of food and internal administration of the household) or that she be imagined as selling (luxury) clothes (Lang 2004b). Given this emphasis, it is only to be expected that there will be references to her “girls” (rather than her “boys”). Moreover, since within this conceptual world, the household served for production and reproduction, just as the Greek oikos, it is not only expected that the successful wife would be imagined as a mother. In addition, since the characterization of the ‫ אׁשת־חיל‬centers on her success and wisdom in managing the household and increasing its wealth, it is only to be expected that matters of sexual pleasure and desire or the woman’s physical looks were construed as irrelevant and thus not worthy of being mentioned.25 As usual in all these instances, it is what was not necessarily anticipated that deserves particular attention. In this case, it is the fact that the community was asked to imagine the ‫ אׁשת־חיל‬as a wife who used her profits to purchase fields wisely and plant vineyards. To be sure, unlike Athenian women, elite Persian women owned (and managed) land,26 but the ‫ אׁשת־חיל‬was imagined within a community in which the ideal world expressed in its “legal” texts was very unclear about whether wives, under normal circumstances, could buy fields on their own and held them as their property,27 and in which, within this ideal “legal world” (commercial), land transfers were not worthy of much thought (and attention), unless it was in the context of redemption (e.g., Leviticus 25; Westbrook 1991: 11; see also pp. 90–117). But the text and the memory of the ‫ אׁשת־חיל‬that it shaped not only stressed the wife’s agency, inside and outside the physical boundaries of her “domestic” space, and into the “male” realm of farming, but also depicted her in terms associated with the masculine/warrior sphere. To some extent, remembering the ‫ אׁשת־חיל‬involved activating an underlying image of economic activity as war and her agency as heroic behavior. The use of terms such as ‫( אׁשת־חיל‬connoting a pair with ‫גבור‬ ‫“ חיל‬mighty warrior” in v. 10), ‫( שלל‬connoting “booty” in v. 11), the 25.  This is true despite the fact that sexual desire (and to be more precise, the desire of a husband for his wife) was considered very important in other wisdom texts and within the very book of Proverbs (see Prov 5:15–19). Compare and contrast with the social constructions of similar matters in Greece. See, e.g., Corner 2011: 60–85. 26.  As is widely known, Spartan women could own land. See, e.g., Hodkinson 2003. 27. See Westbrook (1991: 65) and note, “It is not certain that women could own property at all.” See also this: “[T]he limitation of women’s property rights is the economic linchpin of patriarchal structure. . . . The basic fact that women did not normally own land made them economically dependent on men—first on their fathers, then on their husbands, and ultimately on their sons” (Kensky 2008: 983).

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

37

language of v. 17 (namely, ‫“ חגרה בעוז מתניה ותאמץ זרעותיה‬she girds her loins and strengthen her arms”)28 all point in this direction. One may even say that there was a connoted, secondary military image when the community remembered that her husband “trusts in her”; for him and his household, she is their fortress and “army,” upon whom they can reliably lean (cf., e.g., Wolters 1988; Lawrence 2009: 341–43). The partial “masculinization” of the necessarily “feminine” image of the ‫ אׁשת־חיל‬involved multiple processes of “otherization” and “mirroring” and the creation of a conceptual shared “in-between” realm populated by those construed as agents in the production and accretion of wealth, including both wise males and females.29 A study on these matters cannot be carried out within the present essay, but for the present purposes the accompanying characterization of economic activities aimed at increasing wealth as the equivalent of war is of particular significance, and so is the attribution of heroic features to those succeeding in making a profit. It is worth noting that the social background of the text is a local elite who lives in a community without a local king or army and cannot aspire to heroism in battle, but only (as the text suggests) to increasing their wealth. Remembering the ‫אׁשת־‬ ‫ חיל‬works well in the community, because such action embodies in a female body a social mindscape concerning economy and wealth creation shared by the elite, including directly or indirectly (as members of a retainer group) the male literati who read and reread these texts and who may have read them to others, especially nonliterati members of the elite. Obviously, the ‫ אׁשת־חיל‬is construed as an ancient super-wife and super-mom.30 This being the case, and despite the many evident differences, contemporary critiques and cross-cultural, social anthropological comparisons between the ‫ אׁשת־חיל‬and today’s image of the (North American) “super-mom” may be heuristically helpful for approaching some aspects of the social and ideological setting in which these “types” emerge and become popular. The ubiquity of the concept/image of the present-day “super-mom” is not dependent at all on whether actual North American mothers can fulfill the too-high expectations or on the systemic reorganization of resources that would be required to allow more than just a few to come close to fulfilling them, and certainly not on whether this image is an oppressive burden on some (elite?) women, because it creates expectations that cannot be matched in reality by most women (even elite women). Instead, its popularity is due to its ability to 28.  On this expression, see Novick 2009. 29.  Note also at the basic level of the implied “narrative,” the explicitly necessary partnership of the ‫ אׁשת־חיל‬and male traders, who shared the same goal of acquiring wealth through their respective “trades.” 30.  The point is quite explicitly expressed in, for instance, v. 29; cf. v. 31.

38

Ehud Ben Zvi

embody a set of values and ideals about work, pursuit of profit, wealth, and “heroic” agency in the economic world that characterizes a substantial sector of today’s society and that, in turn, influences others. It stands to reason that the situation in the case of our ancient “superwife/mom” was not substantially dissimilar. If this is true, the social mindset embodied in the ‫ אׁשת־חיל‬was not far from that of a significant sector of individuals in the Persian or early Hellenistic period in Judah who remembered her, as they read, reread, and imagined her.

A Different Kind of Potential Cross-Cultural Observation about Wives with Some “Masculine” Features Yehud was a poor province, and even its elite were relatively poor. The same may be said of early Hellenistic Judah. These circumstances may have facilitated a tendency among the male literati who read these texts and shaped these worlds of imagination and memory to accept and even idealize a slightly “masculinized” feminine image. Crossculturally societies that live in harsher environments are more predisposed to developing a male preference for women who show some “masculine” features, because the latter tend to be construed as communicating a better ability to compete for resources, which is exactly the area in which the ‫ אׁשת־חיל‬excels (Marcinkowska et al. 2014).

Secure, Predictable Society and Principles of Selection Governing Memories The ‫ אׁשת־חיל‬is a site of memory that conjures an entire economic world of activity. The world evoked by the site presupposes a peaceful and stable society. The ‫ אׁשת־חיל‬could laugh at what the future might bring because, in her world, the future was predictable, controllable, and construed in continuity with the present. She did not need to worry about calamities, about invading armies that might destroy her household or pillage her wealth, or any other chaotic event (e.g., drought). In her world, her wise use of wealth has resulted in secure wealth for years to come.31 Moreover, since chaos does not play any substantial role in this world, intelligence and industriousness lead necessarily to wealth, blessing, praise, and honor. What characterizes the ‫ אׁשת־חיל‬is that she is superintelligent and superindustrious (passim), so she surpasses others (v. 29); but all these wives share in the same secure, nonchaotic world. One could mention also that the ‫ אׁשת־חיל‬is a super-reliable wife/ entrepreneur/mom. Everything is in order in her household, and chaos has no leg to stand on. Her household and the world in which she lives 31.  Of course, it is the kind of security and peace that encourages investments in production, training of future workers, and allows for intelligent, long-term planning.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

39

are partially construed as microcosms/macrocosms of each other. If this is true, who is the counterpart of the ‫ אׁשת־חיל‬in the larger world? One answer would be “Wisdom” (cf. Proverbs 1–9 and the ways in which it informs and is informed by Prov 31:10–31), but how is Wisdom manifested in the macro-world? Moreover, within the discourse of the period, would it not be possible to imagine Yhwh as the wise leader of a household that contains the entire world and from which chaos is removed? What about the leadership of a temple with cosmic significance (Van Leeuwen 2007; Ben Zvi 2014)? Or the Persian king? Or multiple possible combinations of the above? Most significantly, this issue is not explored in Prov 31:10–31 (see below). To be sure the secure, nonchaotic, reliable world mentioned above could only exist if there were social and political structures to maintain it. It required a sense of social cohesion but also actual political power and ability to control the area to eliminate any substantial chaotic event. In some ways, this orderly, permanently peaceful world was a local reflection of the nonchaotic, peaceful world of Achaemenid royal ideology in which the Persian king provided happiness to humankind (cf. Lincoln 2012). But remembering the ‫ אׁשת־חיל‬meant, not only viewing the world and economic activity in it from the perspective of a single household, but bracketing matters that were not and could not be controlled by such a household (e.g., taxes, Persian imperial rule, temple leadership, and even Yhwh, who from the perspective of the community governed the world).32 As mentioned above, it is likely that v. 30 did not include the reference to “fearing Yhwh” (see the LXX), but even if it is not the case, Yhwh is not explicitly portrayed as an active agent in the “story” of the ‫אׁשת־חיל‬. The principle of selection governing the memories to be evoked by the ‫ אׁשת־חיל‬and her activities was that, whatever helped to depict her agency was to be remembered, for the pericope was about her and what she did. But the presence of such a rhetorically and mnemonically helpful focus does not mean that the community remembering such a “great” world in which the ‫ אׁשת־חיל‬lived and which, to a substantial extent, seemed to “mirror” theirs was not being implicitly socialized into a world of Persian imperial rule, of temple leadership, of Yhwh’s rule, or of taxes. “Natural” preconditions are most often not worth mentioning, but still images and “memories” of a world that implicitly assumes them socialize as much as or more than explicit references.33 32.  As mentioned above, it is likely that v. 30 did not include the reference to “fearing Yhwh” (see the LXX), but even if this is not the case, Yhwh is not explicitly portrayed as an active agent in the “story” of the ‫אׁשת־חיל‬. 33.  This peaceful, nonchaotic world seems less likely to have been construed as “natural” when the weight of the chaos that followed the fall of the Achaemenid

40

Ehud Ben Zvi

‫אשׁת־חיל‬, Socialization, and Mutually Balancing Voices The preceding sections explored aspects of intellectual thought, world view, and a general social mindscape reflected and communicated by the “story” of the ‫ אׁשת־חיל‬and embodied in her. They have shown that remembering this wife involved socialization into a world of ideas concerning the role of elite wives and the economy in general. It involved inculcating certain approaches concerning profit, trade, industriousness, wealth, land-acquisition and ways to achieve honor in society. It involved also a partial appropriation of the concept of heroism that emphasized, inter alia, how difficult the ongoing pursuit of profit was. It involved a focus on households as the central socioeconomic unit and a general view of society that is devoid of substantial anxiety concerning the power of chaotic powers. Society was construed as stable, predictable, and ordered. Remembering the ‫ אׁשת־חיל‬involved recalling and stressing the role of Wisdom. It also involved balancing other texts and messages that existed within the world of thought and knowledge of the remembering community. One may mention, for instance, how the very presence of a view from a “private” household, as opposed to the usual view from the perspective of political figures and centers (kings, governors, etc.) mutually balances, complements, and informs the community’s approach to many other texts. The same can be said for remembering a female hero rather than a male hero, of remembering the less heroic husband (see below) and needless to say, of the mentioned viewpoints concerning economic activity, trade, pursuit of profit, goods, agents, which, from a perspective of the discourse of the community as a whole, mutually complement, inform and balance other views, widely expressed in other texts.

Observations on the Utopian ‫ אשׁת־חיל‬and Her Orderly World as an Expression of Lacks and Longings The ‫ אׁשת־חיל‬and her world represented a utopia. Of course, it was the utopia of a particular group in society and others (e.g., her “girls” or those who sold their fields to her) may not have shared it. Still, it was a utopian world, and such worlds often provided societies ways Empire began to affect the social mindscape of the community in Yehud (which usually occurs one generation or two after the events). In my opinion, Prov 31:10–31 seems more at home in the 5th or 4th centuries b.c.e. than the 3rd century. This consideration for dating seems to me far more important than the lack of Greek elements. For proposed dates for this text, see n. 9 above.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

41

to address present lacks and express their longings. Wives are not like the ‫ אׁשת־חיל‬though, significantly, the ancient male readers would have liked them to be. The latter is not an insignificant observation, given the expanded role that she exerted, the secondary role of the husband in the leadership of the household (note that nowhere is it stated that her husband “allowed” or “commanded” her to do her role; and see the different situation in Oeconomicus), and common transcultural attitudes such as those expressed in Sir 25:22 (“There is wrath and impudence and great disgrace when a woman supports her husband”). Similarly, households are not as prosperous as that of the ‫אׁשת־חיל‬, and subsistence goods play a larger role in the economy. Even affluent families in an agrarian society located in a stable polity might suffer from lack of rain or pestilence. Farmers rarely “laugh” at what the future may bring, but they would like to. Of course, the ‫ אׁשת־חיל‬served some of the transcultural social roles associated with a super-hero, and in this case and most significantly, a masculinized but saliently female super-hero. But who are her “enemies,” the “super-villains” that she must confront. They are neither the “others” (whether construed as internal or external to her social group) nor natural forces (e.g., drought, pestilence, etc.), mythological animals, or even “the sword.” The “villains” with whom this super-hero has to contend seem to be laziness, lack of understanding, or the proper socialization of those under her. Of course, none of these are substantial hindrances to a person with her wisdom. A super-hero with such enemies is on the one hand the kind of super-hero that emerges from the book of Proverbs as a whole, but on the other hand is a super-hero for a society that even affluent households could only dream of.34

From a Utopian ‫ אשׁת־חיל‬and Her Orderly World to Socioeconomic Realities in Persian or Early Hellenistic Yehud The interwoven network of images mentioned in the preceding section along with their orderly, predictable, stable and wisdom-full world represent an example within a wide range of sets and arrays of sets of “ideal” images in ancient Israel (Ben Zvi 2006; 2013). All of them were characterized by the absence of some (substantial) “lack,” which varied from set to set and array to array, and yet all of them emerged out of a utopianist generative grammar, whose particular manifestations (i.e., the mentioned sets and arrays of sets) informed and balanced each 34.  There is nothing to fear in this world and nothing standing against making a profit and the associated accumulation of wealth and honor, except rejection of the wisdom exemplified in the ‫אׁשת־חיל‬.

42

Ehud Ben Zvi

other. Whereas a discussion of these matters goes beyond the scope of this essay, the same does not hold true for some observations on an additional (and complementary) facet of the study of utopian or imaginary worlds in ancient Israel that is particularly relevant to the contribution that the study of images and (utopian) memories evoked by the ‫ אׁשת־חיל‬makes to the elucidation of some of the matters to which this volume is devoted. Utopias cannot but be based, to some extent, on existing worlds. There are constraints on social imagination and social communication that affect the production of utopias. The “story” of the ‫אׁשת־חיל‬ “worked” because the target readership was aware of what merchant ships, traders, luxury clothing, maidservants, fields, and vineyards were. All of them existed in the world of the text and in that of the remembering community. Likewise, the literati in Yehud who were imagining and remembering the ‫ אׁשת־חיל‬could not have construed her and her world as ideal if they were not willing to accept, at least to some extent, the socio­ economic values that implicitly governed and were reflected and communicated by the ‫ אׁשת־חיל‬in their own “historical” world—not just in the world evoked by remembering the ‫אׁשת־חיל‬. A society in which relatively affluent groups value industriousness and the ongoing pursuit of profit, and in which wealth is meant to lead to more wealth and to increased honor, in which trade is a positive feature and so is increased land acquisition—this society does not have to be a “pre-capitalist” society. Such societies most likely did not exist in antiquity. Instead, it was a society in which traditional agrarianism was combined with the existence of markets and one in which, despite “moralistic” claims by entrenched land aristocracies, wealth might indeed come from trading value-added products;35 moreover, wealth 35.  Even a cursory debate on the place of markets in ancient agrarian societies is well beyond the scope of this essay. On these issues, see, e.g., Bang 2006: 51–88; 2011 and the debate it initiated. See also, e.g., Hunt 1991: 153–68, esp. §3.1.6.1 “Market Exchange,” pp. 158–61. For the present purposes, it suffices to state that markets existed for millennia but that their existence did not turn the relevant economies into “market” or “capitalist” economies. Likewise, concepts that resemble present-day concepts in some form (e.g., loans, interest, shared investments, shared profits) existed in some ancient societies—the obvious case of Old Assyria comes to mind, and see, e.g., Veenhof (1997: 336–66), but this is just one example (for a study of another, see Wunsch 2010: 40–61). But none of this means any of the relevant societies were modern capitalist societies. At the same time, the existence of these ancient concepts that somehow resemble contemporary concepts should not be denied solely for fear of seemingly advancing an anachronistic, “modernistic” view.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

43

from these sources might at times bring in new landowners or further elevate existing ones (see Prov 31:16, 28). Likewise, although the references to the activities of the ‫ אׁשת־חיל‬are clearly hyperbolized for obvious literary and mnemonic reasons, it is safe to assume that they reflect in some way actual activities of wives within the relevant socioeconomic circle in Yehud. The fact that women in substantially different but more-or-less contemporary societies of the period fulfilled similar roles (see esp. Xenophon, Oeconomicus, but also the evidence from Elephantine and other areas in the Persian period; e.g., Waegeman 1992; Lang 2004b; Yoder 2001: 59–72) not only reinforces this position but also demonstrates beyond doubt that these socioeconomic patterns were not exclusive of wealthy or “tradeoriented” (groups within) societies (e.g., Athens, Babylonia). The same point emerges from cross-cultural comparisons with societies that were by no means contemporary with the society in Yehud, as the case of households in the blurry boundary between urban and rural in the Ottoman period shows (Lang 2004a).36 Studies of socioeconomic realities in mid- to late Yehud (or its continuation in early Hellenistic Judah) have focused, with good reason, on the evidence from the archaeological data. But textual evidence may also contribute a great deal. Unlike other textual evidence often advanced that either focuses on putative, one-time events or scribal legislation,37 Prov 31:10–31 sheds (indirectly but) very good light on ongoing economic activities and attitudes among affluent households. Moreover, unlike other textual sources, it sheds light not on the center (e.g., the impact on the economy of a governor or of the temple, as a central institution) but on the activities and dreams of multiple, though wealthy agents. In some ways, one can say that Prov 31:10–31 may be one of the very few sources that allows one to come a bit closer toward something even partially resembling what a limited microhistory of the

36.  Of course, the ‫ אׁשת־חיל‬lives and works in that blurry boundary between rural and urban as well. 37.  Usually the main texts brought to bear are Ezra–Nehemiah, esp. Nehemiah 5 and, though less often, some pentateuchal literature. The first is substantially shaped by common historiographic tendencies/metanarratives (e.g., those involved in the quasi royalization of Nehemiah); the second deals with literary/ ideological examples of legislation that were not necessarily drafted for the purpose of actual implementation. These considerations do not necessarily preclude their use as sources for reconstructing the economic history of Yehud but raise a number of issues that must be addressed—although, for obvious reasons, in a separate study.

44

Ehud Ben Zvi

Persian (and early Hellenistic) period in Judah might have looked like, had we sufficient resources to develop one.38 38.  On microhistory, see, for instance, Szijártó 2002: 209–15; and the “classical” work in Levi 1992: 93–113. For an evaluation of the tendency toward microhistory among a substantial number of contemporary historians, see Marcos 2009: 80–93 (University of Barcelona), “Tendencias historiográficas actuales,” published online and available freely at Cultura Histórica, http://www.culturahistorica.es. As I mentioned elsewhere, the attention that characters such as Ta(pe)met or Mibtahiah of Elephantine have received suggests that, had significant, relevant sources been available, a substantial number of historians of ancient Israel would have developed at least some microhistorical practices when reconstructing the history of Persian period Judah/Yehud (see Ben Zvi forthcoming).

Bibliography Amin, S. 1991 The Ancient World-Systems versus the Modern Capitalist WorldSystem. Review Fernand Braudel Center 14: 349–85. Bang, P. F. 2006 Imperial Bazaar: Towards a Comparative Understanding of Markets in the Roman Empire. Pp. 51–88 in Ancient Economies, Modern Methodologies: Archaeology, Comparative History, ed. P. F. Bang, M. Ikeguchi, and H. G. Ziche. Pragmateiai 12. Bari: Edipuglia. 2011 Roman Bazaar: A Comparative Study of Trade and Markets in a Tributary Empire. Cambridge: Cambridge University Press. Ben Zvi, E. 2006 Utopias, Multiple Utopias, and Why Utopias at All? The Social Roles of Utopian Visions in Prophetic Books within Their Historical Context. Pp. 55–85 in Utopia and Dystopia in Prophetic Literature, ed. E. Ben Zvi. PFES 92. Helsinki: Finnish Exegetical Society / Göttingen: Vandenhoeck & Ruprecht. 2013 Reading and Constructing Utopias: Utopia/s and/in the Collection of Authoritative Texts/Textual Readings of Late Persian Period Yehud. SR  42: 463–76. 2014 Exploring Jerusalem as a Site of Memory in the Late Persian and Early Hellenistic Period. Pp. 197–217 in Memory and the City in Ancient Israel, ed. D. V. Edelman and E. Ben Zvi. Winona Lake, IN: Eisenbrauns. forthcoming  Clio Today and Ancient Israelite History: Some Thoughts and Observations at the Closing Session of the European Seminar for Historical Methodology. In ‘Not Even God Can Alter the Past’: Reflections on Sixteen Years of the European Seminar in Historical Methodology, ed. L. Grabbe. LHBOTS/ESHM 10. London: T. & T. Clark.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

45

Camp, C. V. 1985 Wisdom and the Feminine in the Book of Proverbs. Bible and Literature 11. Decatur, GA: Almond. Cook, J. 1993 The Septuagint as Contextual Bible Translation: Alexandria or Jerusalem as the Context of Proverbs. JNSL 19: 25–39. 1999 The Law of Moses in Septuagint Proverbs. VT 49: 448–61. 2008 Semantic Considerations and the Provenance of Translated Units. Pp. 67–85 in XIII Congress of the International Organization for Septuagint and Cognate Studies: Ljubliana, 2007, ed. M. K. H. Peters. Atlanta: Society of Biblical Literature. Corner, S. 2011 Bringing the Outside In: The Andrōn as Brothel and the Symposium’s Civic Sexuality. Pp. 60–85 in Greek Prostitutes in the Ancient Mediterranean, 800 bce–200 ce, ed. A. Glazebrook and M. Henry. Madison, WI: University of Wisconsin Press. Crook, M. B. 1954 The Marriageable Maiden of Proverbs 31:10–31. JNES 13: 137–40. Falk, M. 1999 The Book of Blessings. Boston: Beacon. Fischer, I. 2005 Gotteslehrerin: Ein Streifzug durch Spr 31,10–31 auf den Pfaden unterschiedlicher Methodik. BZ 49: 237–53. 2006 Gotteslehrerinnen: Weise Frauen und Frau Weisheit im Alten Testament. Stuttgart: Kohlhammer. Fox, M. V. 2009 Proverbs 10–31: A New Translation with Introduction and Commentary. AB 18B. New Haven, CT: Yale University Press. Frood, E. 2010 Social Structure and Daily Life: Pharaonic. Pp.  469–90 in A Companion to Ancient Egypt, vol. 1, ed. A. B. Lloyd. 2 vols. Malden, MA: Wiley-Blackwell. Gilbert, M. 1999 La Donna Forte di Proverbi 31,10–31. Pp. 147–67 in “Ritratto o Simbolo,” Libro dei Proverbi: Tradizione, redazione, teologia, ed. G. Bellia and A. Passaro. Casale Monferrato: Piemme. Goh, S. T. S. 2014 Ruth as a Superior Woman of ‫ ?חיל‬A Comparison between Ruth and the “Capable” Woman in Proverbs 31.10–31. JSOT 38: 487–500. Hausmann, J. 1992 Beobachtungen zu Spr 31,10–31. Pp.  261–66 in Alttestamentliche Glaube und Biblische Theologie: Festschrift für Horst Dietrich Preuss, ed. J. Hausmann and H.-J. Zobel. Stuttgart: Kohlhammer. Hodkinson, S. 2003 Female Property Ownership and Status in Classical and Hellenistic Sparta. In Women and Property in Ancient Near Eastern and

46

Ehud Ben Zvi

Mediterranean Societies, ed. D. Lyons and R. Westbrook. Washington, DC: Center for Hellenic Studies. http://chs.harvard.edu/CHS/ article/display/1219. Horden, P., and Purcell, N. 2000 The Corrupting Sea: A Study of Mediterranean History. Oxford: Black­well. Hunt, R. C. 1991 The Role of Bureaucracy in the Provisioning of Cities: A Framework for an Analysis of the Ancient Near East. Pp.  153–68 in The Organization of Power: Aspects of Bureaucracy in the Ancient Near East, ed. M. Gibson and R. D. Biggs. 2nd ed. SAOC 46. Chicago: Oriental Institute. Hurowitz, V. A. 2001 The Seventh Pillar: Reconsidering the Literary Structure and Unity of Proverbs 31. ZAW 113: 209–18. 2005 The Woman of Valor and a Woman Large of Head: Matchmaking in the Ancient Near East. Pp.  221–34 in Seeking Out the Wisdom of the Ancients: Essays Offered to Honor Michael V. Fox on the Occasion of His Sixty-Fifth Birthday, ed. R. L. Troxel, K. G. Friebel, and D. R. Magary. Winona Lake, IN: Eisenbrauns. Kensky, T. F. 2008 Parasht Pinchas: Another View. P. 983 in The Torah: A Women’s Commentary, ed. T. C. Eskenazi and A. L. Weiss. New York: Women of Reform Judaism. Lang, B. 2004a The Hebrew Wife and the Ottoman Wife. Pp. 140–57 in Anthropology and Biblical Studies: Avenues of Approach, ed. L. J. Lawrence and M. I. Aguilar. Leiden: Deo. 2004b Women’s Work, Household and Property in Two Mediterranean Societies: A Comparative Essay on Proverbs XXXI 10–31. VT 54: 188–207. Lawrence, B. 2009 Gender Analysis: Gender and Method in Biblical Studies. Pp. 333–48 in Method Matters: Essays on the Interpretation of the Hebrew Bible in Honor of David L. Petersen, ed. J. M. LeMon and K. H. Richards. Atlanta: Society of Biblical Literature. Levi, G. 1992 On Microhistory. Pp.  93–113 in New Perspectives on Historical Writing, ed. P. Burke. University Park, PA: Pennsylvania State University Press. Lichtenstein, M. H. 1982 Chiasm and Symmetry in Proverbs 31. CBQ 44: 202–11. Lincoln, B. 2012 ‘Happiness for Mankind’: Achaemenian Religion and the Imperial Project. Acta Iranica 53. Leuven: Peeters. Luke, K. 1991 The Ideal Wife (Prov 31:10–31). Jeevadhara 21: 118–32.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

47

Lyons, E. L. 1987 A Note on Proverbs 31.10–31. Pp. 237–45 in The Listening Heart: Essays in Wisdom and the Psalms in Honor of Roland E. Murphy, ed. K. G. Hoglund, E. F. Huwiler, J. T. Glass, and R. W. Lee. JSOTSup 58. Sheffield: JSOT Press. Marcinkowska, U. M., et al. 2014 Cross-Cultural Variation in Men’s Preference for Sexual Dimorphism in Women’s Faces. Biology Letters. http://dx.doi.org/10.1098/ rsbl.2013.0850. Marcos, F. S. 2009 Tendencias historiográficas actuals. Pp.  80–93 in Cultura Histórica. http:// www.culturahistorica.es. McCreesh, T. P. 1985 Wisdom as Wife: Proverbs 31:10–31. RB 92: 25–46. Meyers, C. 1997 The Family in Early Israel. Pp. 1–41 in Families in Ancient Israel, ed. L.  G. Perdue, J. Blenkinsopp, J. J. Collins, and C. Myers. Louisville: Westminster John Knox. Novick, T. 2009 “She Binds Her Arms”: Rereading Proverbs 31:17. JBL 128 (2009): 107–13. Nwaoru, E. O. 2005 Image of the Woman of Substance in Proverbs 31:10–31 and African Context. BN 127: 41–66. Pomeroy, S. B. 1994 Xenophon, Oeconomicus: A Social and Historical Commentary. Oxford: Clarendon / New York: Oxford University Press. Rofé, A. 2002 The Valiant Woman, γυνὴ συνετή, and the Redaction of the Book of Proverbs. Pp.  145–55 in Vergegenwärtigung des Alten Testaments: Beiträge zur biblischen Hermeneutik. Festschrift für Rudolf Smend zum 70. Geburtstag, ed. C. Bultmann, W. Dietrich, and C. Levin. Göttingen: Vandenhoeck & Ruprecht. Sasson, I. 2013 The Book of Proverbs between Saadia and Yefet. Intellectual History of the Islamicate World 1: 159–217. Steinkeller, P. 1987 The Foresters of Umma: Towards a Definition of Ur III Labor. Pp. 73– 115 in Labor in the Ancient Near East, ed. M. A. Powell. New Haven, CT: American Oriental Society. Stone, E. C. 1999 The Constraints on State and Urban Form in Ancient Mesopotamia. Pp.  203–27 in Urbanism and Economy in the Ancient Near East, ed. M. Hudson and B. A. Levine. Cambridge: Peabody Museum of Archaeology and Ethnology.

48

Ehud Ben Zvi

Szijártó, I. 2002 Four Arguments for Micro History. Rethinking History: the Journal of Theory and Practice 6: 209–15. Doi: 10.1080/13642520210145644. Szlos, M. B. 2000 A Portrait of Power: A Literary-Critical Study of the Depiction of the Woman in Proverbs 31:10–31. USQR 54: 97–103. Talshir, Z.; Yona, S.; and Sivan, D., eds. 2001 Homage to Shmuel: Studies in the World of the Bible. Jerusalem: Bialik Institute / Beer-sheva: Ben-Gurion University Press. [Hebrew] Valler, S. 1995 Who Is ‘Ĕšet Ḥayil’ in Rabbinic Literature? Pp.  85–97 in A Feminist Companion to Wisdom Literature, ed. A. Brenner. Sheffield: Sheffield Academic Press. Vandorpe, K. 2010 The Ptolemaic Period. Pp. 159–79 in vol. 1 of A Companion to Ancient Egypt, ed. A. B. Lloyd. 2 vols. Chichester: Wiley-Blackwell. Van Leeuwen, R. C. 2007 Cosmos, Temple, House: Building and Wisdom in Mesopotamia and Israel. Pp.  67–90 in Wisdom Literature in Mesopotamia and Israel, ed. R.  J. Clifford. SBLSymS 36. Atlanta: Society of Biblical Literature. Repr. pp. 399–421 in From the Foundations to the Crenellations: Essays on Temple Building in the Ancient Near East and Hebrew Bible, ed. M. J. Boda and J. Novotny. AOAT 366. Münster: Ugarit-Verlag, 2010. Veenhof, K. R. 1997 “Modern” Features in Old Assyrian Trade. JESHO 40: 336–66. Waegeman, M. 1992 The Perfect Wife of Proverbia 31:10–31. Pp.  101–7 in Goldene Äpfel in silbernen Schalen: Collected Communications to the XIIIth Congress of the International Organization for the Study of the Old Testament, Leuven 1989, ed. K.-D. Schunck and M. Augustin. Frankfurt am Main: Peter Lang. Westbrook, R. 1991 Property and the Family in Biblical Law. JSOTSup 113. Sheffield: JSOT Press. Wilson, I. D. 2013 Tyre a Ship. ZAW 125: 249–62. Wolters, A. 1985 Ṣôpiyyâ [Prov 31:27] as Hymnic Participle and Play on Sophia. JBL 104: 577–87. 1988 Proverbs XXXI 10–31 as Heroic Hymn: A Form Critical Analysis. VT 38: 446–57. Wright, J. R., ed. 2005 Proverbs, Ecclesiastes, Song of Solomon. ACCS. Downers Grove, IL: InterVarsity.

The “Successful, Wise, Worthy Wife” of Proverbs 31:10–31

49

Wunsch, C. 2005 The Šangû-Ninurta Archive. Pp.  367–79 in Approaching the Babylonian Economy: Proceedings of the START Project Symposium Held in Vienna, 1–3 July 2004, ed. H. D. Baker and M. Jursa. AOAT 330. Münster: Ugarit-Verlag. 2010 Neo-Babylonian Entrepreneurs. Pp. 40–61 in The Invention of Enterprise: Entrepreneurship from Ancient Mesopotamia to Modern Times, ed. D. Landes, J. Mokyr, and W. Baumol. Princeton, NJ: Princeton University Press. Yilibuw, D. 1996 Tampering with Bible Translation in Yap. Pp. 21–38 in Hidden Truths from Eden: Esoteric Readings of Genesis 1–3. Semeia 76. Atlanta: Scholars Press. Yoder, C. R. 2001 Wisdom as a Woman of Substance: A Socioeconomic Reading of Proverbs 1–9 and 31:10–31. BZAW 304. Berlin: de Gruyter.

More Than Friends? The Economic Relationship between Huram and Solomon Reconsidered Gary N. Knoppers

The University of Notre Dame

In a recent monograph, Roger Nam (2012) discusses patterns of symmetrical and asymmetrical trading between Israelite monarchs and their foreign counterparts in the book of Kings. One such symmetrical system of economic exchange that Nam (2012: 76–84) discusses at length is the bilateral treaty between Solomon and Hiram of Tyre. This international diplomatic friendship enables Solomon to build the temple and to endow it with finely crafted Phoenician furnishings. In this horizontal relationship of economic reciprocity, Solomon annually dispatches agricultural products to Hiram, and Hiram, in turn, sends Solomon cedar and cypress timber, as well as technical expertise to construct a royal palace and sanctuary in Jerusalem. Yet, Solomon’s international trade deals, as portrayed in Kings, also have negative consequences for the body politic. Because of his indebtedness to his Tyrian treaty partner, the Israelite king is eventually forced to cede 20 towns in Galilee to satisfy his international obligations (1 Kgs 9:11–14). The subject of this volume deals with agrarian economies in depopulated areas in Persian antiquity. In this essay, I would like to examine how the literati of one such depopulated area, namely Yehud, completely transform the presentation of an earlier well-populated era— namely, Solomon’s reign—as bequeathed to them in the book of Kings. Given that Persian period Yehud is generally understood to be a time of economic impoverishment, it is all the more important to understand how elite scribes in that society (re)imagined a much earlier period of pan-Israelite solidarity and boundless prosperity. In particular, I am interested in how Chronicles transforms one critical international economic partnership—Solomon’s parity treaty with Author’s note: Earlier, oral versions of this essay were delivered at the 2013 annual meetings of the Canadian Society of Biblical Studies and the European Association of Biblical Studies. I am grateful for the helpful comments and questions I received at each of these meetings.

51

52

Gary N. Knoppers

Hiram (Huram) of Tyre—into a patron-client relationship.1 The friendship across borders, cast in Kings as a fictive kinship relationship among “brothers,” becomes in Chronicles an asymmetrical system of trans­ national economic exchange.2 In a late Persian or early Hellenistic period setting, Chronicles completely rewrites the horizontal relationship of economic reciprocity between two potentates into a vertical relationship of unequal economic exchange.3 Whereas in Kings Hiram refers to Solomon as “my brother” (‫ ;אחי‬1 Kgs 9:13), in Chronicles Huram refers both to David and to Solomon as “my master” (‫ ;אדני‬2 Chr 2:13–14).4 In Kings, a series of mutually beneficial commodity transactions lead Solomon and Hiram to formalize their bilateral relationship diplomatically: “There was peace between Hiram and Solomon and the two of them ratified a covenant” (‫ ;ויכרתו ברית שניהם‬1 Kgs 5:26). In Chronicles, there is no such pact between the two states. It is not as though Chronicles exchanges a parity treaty for a vassal treaty because Chronicles mentions no treaty at all.5 1.  In speaking of “Huram” (‫)חורם‬, I am following the Qere. In some cases, the Kethiv and the Versions read the expected “Hiram” (‫)חירם‬. So, e.g., the MT and Tg. 2 Chr 9:21, ‫ ;חורם‬LXX (Χιράμ), Syr., and Vg., “Hiram.” See also the variants in 1 Chr 14:1 (Knoppers 2004b: 594). 2.  The usage may be compared with the concept of (Akkadian) aḫḫūtu, “brotherhood,” in structuring international relations (Kalluveettil 1982; Liverani 1990). Within the Hebrew Bible, see also 1 Kgs 20:32–33; Amos 1:9. 3.  If Chronicles was written in the early Hellenistic era rather than in the late Persian period, it would not materially affect the larger argument staked here because most Persian administrative arrangements in the southern Levant seem to have been maintained, in spite of the transfer of power from one empire to another (Lemaire 1996; 1999; 2001; Lipschits and Vanderhooft 2011). 4.  In Samuel, Hiram sends messengers to David with cedar timber, masons, and carpenters to construct a palace for him (2 Sam 5:11–12). David interprets the international recognition bestowed upon him by one of Israel’s key northern neighbors as a sign “that Yhwh had established him as king over Israel (and) that his kingship was highly exalted for the sake of his people Israel” (2 Sam 5:11–12 // 1 Chr 14:1–2). The story told in Samuel about Hiram’s shipment of cedar, masons, and carpenters to build David’s palace is retold with a few changes, but Hi/uram’s diplomatic gesture carries special weight in Chronicles, a work in which commendations, tribute, and gifts from other leaders and nations unmistakably signify divine blessing (Dillard 1984b). Later in his reign, the Chronistic David is able to set aside “innumerable cedar logs” for another major public works project, the Jerusalem temple, because the Sidonians and the Tyrians had sent him so much timber (1 Chr 22:4). The anecdote, unparalleled in Samuel, indicates that the David/Huram relationship, like the Solomon/Huram relationship, is one of unequals. David does not request the added lumber; it is sent to him as a gift. 5.  Of course, a formal treaty was not a prerequisite for an international clientpatron relationship to exist or function properly (Westbrook 2000: 40).

The Economic Relationship between Huram and Solomon

53

The metamorphosis of a bilateral treaty partnership to a non-treaty patron-client association raises its own set of questions. At a time in which the province of Yehud was only a fraction of the size and population of the projected united kingdom, why expend the effort to transform the relationship between Israel and Tyre? One could cite the well-known Chronistic aversion to international covenants as unwelcome encroachments on Israel’s primary covenantal relationship with Yhwh (Knoppers 1996). This may well be part of the story, but it is not all of it, because such an anti-treaty stance does not explain why Chronicles devotes consistent attention to the Solomon-Hiram collaboration. It would have been easy to follow the example of the authors of Ezra, who briefly state that the returned exiles paid food, drink, and oil to the Tyrians and the Sidonians to bring them cedar from Lebanon for reconstructing the temple (Ezra 3:7). The writers of Ezra attempt neither to specify the mechanics of the relationship between the relevant parties nor to pursue the Judean-Tyrian connection.6 But Chronicles depicts Huram continuing to cooperate with Solomon, even intensifying his support for Solomonic international commercial ventures. If Huram ostensibly does not stand under any international legal obligations to his Israelite counterpart, why does he maintain the relationship after he has duly fulfilled his pledge to provide material and technical expertise for the construction of Solomon’s sanctuary and royal palace? I would like to begin with a couple of observations on the Solomon narrative found in Chronicles. Near the end of his tenure, Solomon comes to rule “over all the kings from the River (Euphrates) to the land of the Philistines and to the boundary of Egypt” (2 Chr 9:26). This remark, whether borrowed from 1 Kgs 5:1 relating to an earlier part of Solomon’s reign or from a Vorlage that resembled the order found in the LXX (1 Kgs 10:26a) relating to the end of Solomon’s reign, is significant.7 6.  That such a commercial connection existed, at least at some points in the postmonarchic period seems evident by the anecdote in Neh 13:16. Recently, see Edelman 2006. More broadly, see Elayi 1987; 1990; and Jigoulov 2010. 7.  Klein (2012: 13–47) argues that the Chronicler’s Vorlage in this and in several other cases in the Solomon narrative more closely resembled LXX Kings than it did MT Kings. It should be recalled that Chronicles lacks the other generalization about the geopolitical dimensions of Solomon’s rule, which appears earlier in Kings (1 Kgs 5:4[4:24]), namely, that “he was holding sway over all of Transeuphratene (‫הוא רדה‬ ‫ )בכל עבר הנהר‬from Tiphsah until Gaza— over all the kings of Transeuphratene—and he had peace on all sides round about him.” LXX 3 Kgdms 2:46f–g, reads similarly but substitutes enigmatic Raphi for Tipshah, “[H]e was head everywhere across the river from Raphi to Gaza among all the kings across the river (ὅτι ἦν ἄρχων ἐν παντὶ πέραν τοῦ ποταμοῦ ἀπὸ Ραφὶ ἕως Γάζης, ἐν πᾶσι τοῖς βασιλεῦσι πέραν τοῦ ποταμοῦ), and he had peace on all sides round about him” (καὶ ἦν αὐτῷ εἰρήνη ἐκ πάντων τῶν μερῶν αὐτοῦ

54

Gary N. Knoppers

Coming near the end of Solomon’s reign, the capstone comment is significant because it portrays Solomon as holding sway over the various potentates in what early readers would recognize as the province of ʿăbar–nahărâ.8 To be sure, the designation ebir nāri, “Beyond the River,” is already attested in 7th-century b.c.e. Assyrian sources as a geographical term for the land that constituted, from an eastern perspective, the area west of the Euphrates River (Parpola and Porter 2001: 8; Rainey and Notley 2006: 164, 278–83). During the reign of Xerxes (486–465 b.c.e.) or shortly thereafter, the territory of Transeuphrates (Aramaic ‫ ;עבר נהרא‬Hebrew ‫ )עבר הנהר‬became an organized political province separate from its earlier association with Babylon (Stolper 1989; 1992). It is this geo­political designation and not the earlier geographic designation of Neo-Assyrian and Neo-Babylonian times that would be familiar to early readers of Chronicles. The association of Solomonic authority with the satrapy of Transeuphratene is especially noteworthy because the summary remark does not deny the existence of a variety of kingdoms and polities coexisting within the context of a larger overarching regional political entity. It simply claims that for an indeterminate time Solomon came to exercise authority over these other states. The comment about Solomonic hegemony in Transeuphratene fits its literary context nicely in Chronicles as the climax to Solomon’s rule. There is a progression to the peaceable king’s tenure that builds on the successful construction of the Jerusalemite temple and its integration into national life.9 Having completed and dedicated the central sanctuary that Yhwh directly elected him to build, Solomon does not rest on his laurels.10 He continues to build up the infrastructure of his domain and expand its trade with other polities in the ancient Near East. I would like to suggest that the writers cast Solomon, the “man of rest” (‫ ;איש מנוחה‬1 Chr 22:9), as a kind of satrap who comes to exercise authority within and benefit from the larger geopolitical sphere he inhabits. κυκλόθεν). The testimony of LXX 1 Kgs 5:4[4:24] is less specific: “For he was head of Transeuphratene and he had peace on all sides round about” (ὅτι ἦν ἄρχων πέραν ποταμοῦ, καὶ ἦν αὐτῷ εἰρήνη ἐκ πάντων τῶν μερῶν κυκλόθεν). This summary comment is also not found in Chronicles. 8.  The Vorlage of Kings employed by the writers of Chronicles may have had the notice at this location (see LXX 1 Kgs 10:26a), so it is by no means a given that the writers of Chronicles deliberately repositioned it (Klein 2012: 135). 9.  On the stress on peace as a hallmark of Solomon’s tenure, see Jonker 2008. 10.  The theme of Solomon’s divine election appears prominently in the transition from David’s reign to that of Solomon (1 Chr 28:5, 6, 10; 29:1), but the individuation of the Davidic promises (1 Chr 17:11–14) already points in this direction (Knoppers 2004b: 661–77).

The Economic Relationship between Huram and Solomon

55

Patron-Client Relationships in the Persian Empire In arguing that Solomon is presented as a quasi-Persian-period satrap in the area “Beyond the River,” I am assuming that the writers of Chronicles were familiar with different levels of Achaemenid authority and the means by which Persian rulers and their subordinates both administered their far-flung realm and related to each other. Other scholars have compared the treatment of Solomon in Kings, whose reign or, rather, the first part of it is characterized by international peace and harmony, with the realm of peace and prosperity promoted in Persian royal ideology (de Pury 2003; Römer 2008).11 Court propaganda does not trumpet the forceful conquest of a host of recalcitrant enemies. Rather, Achaemenid decorative palace reliefs feature images of various nations living harmoniously together and joyfully submitting to the imperial center (Root 1979; Briant 2002: 172–95). The Pax Persica, as it has become known, was promoted by Persia’s elite as a benefit to all who were fortunate enough to live within its orbit of influence.12 At the head of this international realm sat the king, “who contained within his person and his office the welfare of the empire” (Young 1992: 236). Of particular importance for understanding the distinctive Achaemenid statecraft of creating and enhancing conditions of peace, over against the records of previous ancient Near Eastern imperial regimes, was the elaborate system of gift-exchange that Achaemenid leaders developed over the course of Persian rule and that became a major feature of royal, satrapal, and local administration (Wiesehöfer 1980; 1989; Briant 1989; Sancisi-Weerdenburg 1989; Bruno 2010; Miller 2010). Gift-giving was, of course, a constituent feature of earlier Near Eastern and Egyptian royal administrations, but successive Achaemenid rulers developed the custom in elaborate new ways.13 For those looking to advance their careers within their local administrations or satrapies, bestowing gifts was an indispensable means to ingratiate oneself with 11.  Indeed, it has come to my attention that Dalby (2013) compares the Solomon of Kings with a Persian period satrap. 12.  For many peoples who considered the Persians to be unwelcome intruders and actively resisted foreign occupation, the Pax Persica was, of course, anything but peaceful. For the residents of Lower Egypt, for example, the Persian period from the time of Cambyses (525 b.c.e.) through the reign of Artaxerxes III was largely disruptive and destructive (Redford 2006: 256–68). For other sobering examples, see Lincoln 2012: 393–405. 13.  Several aspects of Achaemenid imperial practice in deploying gift-giving as a means of engaging, controlling, and manipulating other powers are anticipated in the Amarna period (Cohen and Westbrook 2000; Westbrook 2000; Naʾaman 2000; Zaccagnini 2000).

56

Gary N. Knoppers

one’s superiors. Given the vast numbers of lower-level officials in proportion to the limited number of governors, satraps, and royal advisers, it is readily understandable why lower-level officials would want to curry favor with their supervisors. But what seems unusual about the Achaemenid regime was the major deployment of gift-giving by the king to enhance his power. The Great King reinforced his position atop the sociopolitical hierarchy by means of ritualized gift-exchange (Lewis 1989; Briant 2002: 302–47; Klinkott 2007). As Mitchell observes with reference to the emperor’s dealings with his Greek clients in the West: the “King stood at the centre of this matrix of gift and countergift, and the function of the gifts was to fix the King in his position of regal superiority . . . the King always took care to pay his own debts and to give a gift in return that was of greater value than the gift given, thus putting the giver in a state of debt to the King” (Mitchell 1997: 113). Gift exchange became such a prominent feature of Persian life that Herodotus (3.89) claimed that there was no fixed tribute during the reigns of Cyrus and Cambyses. Payment was made instead in gifts. Although Herodotus’s comment does not comport with epigraphic and archival evidence, it highlights the historical importance that gift-giving acquired in forming, reinforcing, or transforming political relationships within the context of a much broader Achaemenid government. Xenophon writes in his obituary of Cyrus the Great that Cyrus was a man worthy to become king, because he not only received more gifts than any man alive but also gave more and better gifts than anyone else (Anabasis 1.9.20–28).14 With respect to his subjects, the Achaemenid monarch stood at the top of a vast, multilayered, and complex hierarchy of patron-client relationships. Through mechanisms of access, gifts, and friendship, he patronized his underlings, who were in turn indebted to him for advancements made in their careers. Satraps, in turn, employed the material favors bestowed to them from on high to patronize their own friends and clients (Nussbaum 1959: 16–29; 1967: 33; Roy 1967; Briant 2002: 308–25, 340–45, 394–98; Trundle 2004: 109–10; 2010: 227–52). The officials serving under satraps were bound, much as the satraps were bound to the king, by the terms of personal, ritualized attachments to their superiors. While the king may have had no direct relationship to most of the provincial governors serving in the network of satrapies 14. Thus for Xenophon, Cyrus inspired a superior kind of loyalty among his followers than his rival (Artaxerxes) enjoyed, whom he was attempting to replace (1.9.29–31). “Using the gift-giving mechanism of philia, Cyrus sought to make it so that prominent Persians were more indebted to him than to Artaxerxes, in the process usurping the King’s place at the top of the Persian socio-political hierarchy” (Rop 2013: 141). See also Xenophon, Anabasis, 1.1.5.

The Economic Relationship between Huram and Solomon

57

within his vast domain, each minor provincial governor was nevertheless part of a long chain of commitments and obligations that ended with the king.

Solomon and Hi/uram: Temple-Palace Negotiations Briefly examining the particular dynamics of Persian period patronclient relationships brings us to the Huram-Solomon connection. Following his ascent to the kingship during a period of widespread peace, Solomon moves quickly to capitalize on the blessed condition within which Israel finds itself.15 Being an astute student of Deuteronomy, as the law of the king (Deut 17:14–20) enjoins him to be, Solomon realizes that the construction of the central shrine is incumbent upon the body politic as soon as Israel is safely settled within the promised land (Deut 12:10–12). Hence, immediately after his accession, with the kingdom militarily secure and prosperous, Solomon “gave the order to build a house for the name of Yhwh” (2 Chr 1:18). This means, among other things, that a good deal of material found in Kings about Solomon’s wisdom, career as a sage, state administrative reorganization, and court provisions is wanting in Chronicles. Having decided to build the temple, Solomon conducts a muster (2  Chr 2:1), evaluating the number of laborers, quarriers, and supervisors available to him. He then sends a detailed letter to Huram of Tyre to enlist his strategic assistance (2 Chr 2:2–9). Because of the major Chronistic reorganization and consolidation of Solomon’s early tenure, the Huram story takes on greater prominence in Chronicles than it does in Kings.16 That the sequence and content of the Solomon-Huram correspondence differ in pivotal ways from the sequence and content in Kings 15.  The work does not indicate how Solomon became such a prominent player in international trade within the eastern Mediterranean world (2 Chr 1:14–17) so soon after his accession to “the throne of the kingdom of Yhwh” (1 Chr 29:25). Precisely because no such explanation is provided, readers are compelled to draw inferences from the information provided to them. The juxtaposition of the Gibeonite theophany with the narrative notices about Solomon’s military powers, international trading ventures, and Jerusalemite wealth suggests a connection between them. Some of God’s promises to Solomon at Gibeon—namely, riches, wealth, and honor (2 Chr 1:12–13), like unto which no previous king or later king would ever enjoy—find immediate fulfillment as Solomon begins his reign. Nevertheless, Solomon is not at a point early in his career where he can be said to rule over the entire Mediterranean littoral west of the Euphrates (cf. 1 Kgs 5:1, 4). His newfound prestige, although impressive, is more modest. 16.  On the reorganization of Solomon’s reign in Chronicles, see Dillard 1984a; 1987: 1–24; Williamson 1982: 192–202; Japhet 1993: 521–47; Dyck 1998: 152–56; Klein 2012: 18–41.

Gary N. Knoppers

58

Table 1   Kings

 Chronicles

David’s fatherly advice

Topic

1 Kgs 2:1–10

[cf. 1 Chr 22; 28]a

Introductory notices

1 Kgs 2:11–12

1 Chr 29:23–25

Consolidation of kingdom

1 Kgs 2:13–46

2 Chr 1:1b

High places and pharaoh’s daughter

1 Kgs 3:1–3



Theophany at Gibeon

1 Kgs 3:4–15

2 Chr 1:2–13

Chariots and horsemen

[cf. 1 Kgs 10:26]

2 Chr 1:14

Prosperity

[cf. 1 Kgs 10:27]

2 Chr 1:15

Horse and chariot trade

[cf. 1 Kgs 10:28–29]

2 Chr 1:16–17

Tale of two prostitutes

1 Kgs 3:16–28



Administrative (re)organization

1 Kgs 4:1–19



Huge population

1 Kgs 4:20



Rule over Transeuphratene

1 Kgs 5:1

[cf. 2 Chr 9:26]

Daily court provisions

1 Kgs 5:2–3



Rule in all Transeuphratene

1 Kgs 5:4



Safety and security

1 Kgs 5:5



Horse stalls and horsemen

1 Kgs 5:6

[cf. 2 Chr 9:25]

Monthly provisions from prefects

1 Kgs 5:7–8



Sapiential compositions

1 Kgs 5:9–14



Solomon to build temple



2 Chr 1:18c

a.  The Israel Solomon experiences when he ascends to kingship is unimpeded by internal factions or strife. Solomon’s legitimacy is stressed by repeated references to him as David’s rightful successor and to his appointment to kingship over all Israel by divine election (1 Chr 17:11; 23:1; 28:6; 29:23–25; 2 Chr 1:1; 6:10; 7:17–18). Israel’s authoritative national institutions take shape during the united monarchy, and the combined reigns of David and Solomon effectively establish a template by which later periods are judged (Throntveit 1988; 2003; Schweitzer 2007: 76–93; Tiňo 2010: 35–107). b.  There is no need to pacify, eliminate, or exile domestic foes (1 Kings 1–2), because David has already consolidated domestic support for Solomon (1 Chronicles 22–29). c.  The placement of this notation may be compared with the declaration of David in 1 Chr 13:1, expressing his wish to bring the ark into Jerusalem.

(see table 1) makes the Chronistic version of the Israelite-Tyrian exchange all the more fascinating. In many instances, Chronicles manifests an awareness of the claims made in Kings but completely reverses their tenor and thrust. Even in cases in which the work selectively borrows from Kings, the recontextualization, rearrangements, and supplementation of the inherited material generate many new meanings.

The Economic Relationship between Huram and Solomon

59

Table 2  Kings

Chronicles

Decision to build a temple and a palace

Topic



2 Chr 1:18

Muster: laborers, quarriers, and supervisors



2 Chr 2:1

Hiram sends envoys to Solomon

1 Kgs 5:15



Initial message to Hi/uram

1 Kgs 5:16

2 Chr 2:2–3

David disallowed from construction

1 Kgs 5:17–19a



Purpose of temple



2 Chr 2:4–5

Solomon’s request for artisan

[cf. 1 Kgs 7:13]

2 Chr 2:6

Solomon’s request for materials

1 Kgs 5:20a

2 Chr 2:7–8

Open payment offer

1 Kgs 5:20b



Allocation of gifts

[cf. 1 Kgs 5:25]

2 Chr 2:9

Hi/uram’s positive reaction

1 Kgs 5:21

2 Chr 2:10–11

Huram’s dispatch of Huram-abi



2 Chr 2:12–13

Hi/uram’s compliance

1 Kgs 5:22–25

2 Chr 2:14–15

Solomon’s payment to Hiram

1 Kgs 5:25

[cf. 2 Chr 2:9]

Parity treaty established

1 Kgs 5:26



Israelite corvée

1 Kgs 5:27



Muster and corvée: resident aliens



2 Chr 2:16

Implementation: relays to Lebanon

1 Kgs 5:28



Laborers, quarriers, supervisors

1 Kgs 5:29–30

2 Chr 2:17

Temple-building preparations

1 Kgs 5:31–32



a.  In Chronicles, such an explanation is unnecessary, because readers know that David himself made a series of speeches both to Solomon (1 Chr 22:7–17; 28:9–10, 20–21), and to the Israelite leadership he assembled in Jerusalem (1 Chr 22:17–19; 28:2–8; 29:1–5, 20) during the latter part of his reign, carefully preparing for the transition to the rule of his divinely chosen son. Indeed, one of David’s speeches quotes Yhwh prophesying (and punning on) the name of David’s son, “the man of rest,” who will succeed the king, who “shed much blood and waged great battles. . . . Solomon (šĕlōmōh) will be his name, and peace (šālôm) and quiet (šeqeṭ) I shall give to Israel in his days” (1 Chr 22:8–9).

Only some of the major differences can be highlighted here. In 1 Kgs 5:15[1], Hiram takes the initiative (see table 2). “Hiram sent messengers to Solomon, when he heard that Solomon had been anointed as king in place of his father, with whom Hiram was always a friend” (‫כל־‬ ‫)אהב היה חירם לדוד הימים‬. Solomon, in turn, responds to the Tyrian initiative by sending a message back to Hiram, negotiating for assistance in building Jerusalem’s new sanctuary (1 Kgs 5:16–20). Later in his reign, when the construction phase is complete, Solomon requests of Hiram a

60

Gary N. Knoppers

crafts specialist to work on the temple furnishings (1 Kgs 7:13–14). But in Chronicles, a well-organized and far-sighted Solomon includes the artisan as part of his requisition from the Tyrian king. The two accounts may be contrasted as in table 2. By omitting Hiram’s initial message of congratulations (1 Kgs 5:15), Chronicles presents Solomon as the one who takes the initiative and dominates the interaction.17 In his communication to Huram, Solomon affirms Huram’s earlier assistance in shipping cedar to David, which enabled him to build a royal palace. Solomon then proceeds to tell the Tyrian monarch all he needs for his project. He commands Huram to send a skilled crafts specialist to Jerusalem to work alongside Solomon’s own specialists and to send him an abundance of cedar, cypress, and almug from Lebanon (2 Chr 2:6–8). Like the Kings narrative, the Chronicles narrative reflects the important role the Phoenicians played in international trade and commerce. Phoenician sea traders exported cedar southeast to the city-states of Mesopotamia and down the Mediterranean coast south all the way to Egypt. Moreover, when the Chronistic Solomon details the skill set that the artisan should possess, thus melding the temple construction materials with those of the tabernacle, many of the materials he mentions—gold, silver, bronze, iron, purple, crimson, and violet—are commodities Tyrian traders were renowned for brokering (Rainey and Notley 2006: 28–29; Pohlmann 1996–2001: 2.374–95; Joyce 2007: 174–80; Block 1997–98: 2.28–121). This shows some awareness of Phoenician commerce on the part of the writer. It is clear, then, that Chronicles restructures the Solomon-Huram diplomatic exchange in its source to recast their horizontal system of economic exchange as a vertical system of economic reciprocity. Other features of the international correspondence buttress the point. In Kings, Solomon wisely or unwisely allows Hiram to set whatever price he desires for the cedar and professional services that Solomon requests (1 Kgs 5:25[11]).18 But in Chronicles, Solomon announces to Huram that he has assigned for Huram’s servants the following gifts: 17.  The introduction is borrowed from 1 Kgs 5:16[2] “And Solomon sent to Hiram saying.” The skeletal framework of the Chronistic presentation (2 Chr 2:2–15) bears some resemblances to the exchange of messages between Solomon and Hiram depicted in 1 Kgs 5:15–23. In both cases, the two monarchs dispatch messages to one another. Indeed, there are certain phrases from Kings that reappear in Chronicles, but the content and social dynamics of these borrowed phrases take on a remarkably divergent meaning in Chronicles. 18. “Solomon relinquishes any negotiating leverage with his statement .  .  . ‘There is not one among us, who can cut timber like the Sidonians’ ” (1 Kgs 5:20; Nam 2012: 78).

The Economic Relationship between Huram and Solomon

61

“twenty thousand kors of crushed wheat, twenty thousand kors of barley, twenty thousand baths of wine, and twenty thousand baths of oil.” The choice of commodities and their large sums requires some commentary. First, the Kings account makes no mention whatsoever of the huge quantities of barley and wine. In the projected united monarchy of Chronicles, which encompassed, among other areas, the Judean and Samarian hill country, the Shephelah, and the coastal region, the range of agricultural products makes good sense. The addition of barley and wine to the list of Kings complements the wheat and olive oil mentioned in the Chronistic Vorlage. The text thus underscores the profoundly agricultural base of the Israelite economic infrastructure, as imagined by the author. Second, the commodity amounts seem to be astronomically large. Indeed, at first glance, the quantities listed in Chronicles (2 Chr 2:9) appear to dwarf the already fantastic numbers appearing in (MT) 1 Kgs 5:25[11]: “twenty thousand kors of wheat as provisions for his household and twenty kors of pressed oil.”19 The kor (‫ ;כר‬cf. Akk. kurru) is a measure of dry capacity of unknown quantity. Estimates range from 320 to 450 liters, but it may be that the kor was not an entirely uniform measure.20 Whatever the exact amount, the total is remarkable. The amount listed in Kings would be almost double what Solomon’s court is said to have received in a year from its Israelite subjects (1 Kgs 5:2). Within the world of high numbers in Chronicles, the large quantities are, however, not altogether extraordinary. For instance, following his victory over the Ammonites, King Jotham receives from the Ammonites a tribute of 100 talents of silver, 10,000 kors of wheat, and 10,000 kors of barley for three years (2 Chr 27:5). Given that high, rounded numbers dominate the Chronistic Sondergut, specific, small, and odd numbers are the surprise (Knoppers 2004b: 569–71). As for those commodities unique to Chronicles, the quantities are likewise colossal round numbers. The bath (‫ )בת‬is often thought to be a liquid measure equivalent to an ephah (Ezek 45:11) and thus 40–45 liters in capacity (1 Kgs 7:26, 38; Isa 5:10; HALOT 166b). By a similar reckoning, Klein (2012: 36) estimates that the Chronicler has Solomon 19.  While MT 1 Kgs 5:25[11] reads, ‫ועשרים כר שמן כתית‬, “twenty kors of pressed oil,” LXX 1 Kgs 5:25[11] has, καὶ εἴκοσι χιλιάδας βεθ ἐλαίου κεκομμένου, “twenty thousand baths of pressed oil.” Similarly the MT, “twenty thousand baths of oil” (‫אלף‬ ‫ )ושמן בתים עשרים‬and LXX 2 Chr 2:9 (καὶ ἐλαίου μέτρων εἴκοσι χιλιάδας), “twenty thousand measures of oil.” 20.  Kuhrt (2010: 884) lists the capacity of the Babylonian kurru (= 5 panu) as 180 liters and the capacity of the West Semtic kor as 360 liters (= 30 seah) but notes that such measurements are approximate.

62

Gary N. Knoppers

dispatching 116,000 gallons of wine and 116,000 gallons of oil for Huram’s lumberjacks. But Lipschits, Koch, Shaus, and Guil (2010) caution that the bath may not be a strict liquid measure but, rather, a Judahite storage jar, the well-known ‫ למלך‬jar. They argue (Lipschits et al. 2010: 458) that it is an open question whether Iron Age Judah possessed a constant, much less-fixed system of liquid volume measurements. Similar things may be said of the “twenty thousand baths of (olive) oil” (‫)ושמן בתים עשרים אלף‬.21 To put matters in practical terms, one could summarize that Solomon gave Huram 20,000 ‫ למלך‬jars of wine and 20,000 ‫ למלך‬jars of oil. No wonder, then, that Huram responded so gratefully and enthusiastically. Who would not be happy to receive 40,000 ‫ למלך‬jars of wine and olive oil? Third, it is unclear whether Chronicles presents Solomon as being even more generous than the Solomon of Kings. To be sure, Chronicles includes gargantuan quantities of wine and oil lacking in Kings, but 1 Kgs 5:25[11] adds that “Solomon used to give (this) to Hiram every year” (‫)שנה בשנה‬. By contrast, the implication of the Chronicles story is that Solomon’s contribution was a one-time payment. Such an inference is confirmed by the fact that the Solomon narrative lacks the later observation found in 1 Kgs 9:11 that Solomon made his payments to Hiram for a period of 20 years. In other words, having offered to pay Hiram whatever Hiram wanted, Solomon finds himself committed to supplying such foodstuffs to his treaty partner for two decades. In summary, Solomon plays the role of a very generous patron and Huram that of a very grateful client, but it does not seem that Solomon grossly overpays. Solomon initiates the exchange of communiqués, dictates what he needs from the Tyrian king, and itemizes enormous sums of agricultural produce to be given to Huram’s employees, who become, in turn, Solomon’s employees (2 Chr 2:13–14).

The Gift That Keeps on Giving: Solomon and Huram in Galilee If the temple/palace arrangements between Solomon and Huram establish an international patron/client relationship, written in Persianperiod terms, the verbal response of Huram confirms Solomon’s higher status. Huram is even more fulsome in his gratitude and enthusiastic Yahwistic praise (2 Chr 2:10–11) than his alter ego is in Kings (1  Kgs 5:21[6]). As Ben Zvi (2006: 270–88) observes, the sentiments Huram expresses about Yhwh “creating the heavens and the earth” could have 21.  The reference is to olive oil, one of the staples of Israel’s agricultural produce (Borowski 2003). Kings stipulates “pressed oil” (‫)שמן כתית‬. See further HALOT 1567a–1569b.

The Economic Relationship between Huram and Solomon

63

Table 3 Topic Summary of building period

 Kings

Chronicles

1 Kgs 9:10

2 Chr 8:1

Solomon’s ceding of towns

1 Kgs 9:11



Hiram’s unhappiness

1 Kgs 9:12–13



Hiram’s shipment of gold

1 Kgs 9:14



Fortification of Huram’s ceded towns



2 Chr 8:2

Capture of Hamath–zobah



2 Chr 8:3

been uttered by a pious Israelite. And that may be the point. Huram presents himself as one of Solomon’s own courtiers might, eager to please his master and continue the client-patron relationship. I would argue that Huram’s desire to ingratiate himself with his superior explains why, following 20 years of successful temple and palace construction, Huram gives 20 towns in Galilee to Solomon (2 Chr 8:1). Huram honors Solomon with one town for every year of construction. The towns are thus a sign of how much the Tyrian king respects and values his Israelite counterpart. As for the gifted towns, Solomon proceeds to rebuild them and settle Israelites within them (2 Chr 8:2). This version of the story marks, of course, a complete reversal of the claims made in Kings (table 3). In Kings, Solomon cedes 20 towns in the land of Galilee (‫)בארץ הגליל‬ to Hiram, because of “all of the cedar, cypress timber, and gold that Hiram had sent him” (1 Kgs 9:11). According to a later text, the amount of bullion that Hiram (had) sent to Solomon was substantial: one hundred and twenty talents of gold (1 Kgs 9:14).22 But when Hiram went to inspect the towns granted to him, “they were not pleasing in his eyes” (‫)ולא ישרו בעיניו‬. As a result, he questioned Solomon, his “brother,” 22.  Würthwein (1977: 106–7) and Nam (2012: 84) follow some earlier commentators in seeing this verse as a late gloss triggered by the appearance of “gold” in 1 Kgs 9:11 (or, more broadly, by the negative material in 1 Kgs 9:11–13). The gloss ostensibly presents Hiram as repaying Solomon for the 20 towns. Cogan (2001: 300) views the text, not as a gloss, but as an authorial footnote to the earlier claim made in 1 Kgs 9:11a that Hiram had sent gold to Solomon. This is possible, given the absence of a parallel notice in Chronicles. To complicate matters, 1 Kgs 9:14 (. . . ‫ )וישלח‬does not make any overt connection with the incident recounted in 1 Kgs 9:11–13. One could guess that the shipment of bullion was sent to cover the difference between the assessed value of the towns and what was due to Hiram, but the cost seems exorbitant. Alternatively, the notice may signal a commitment, whether satisfied or not, on Hiram’s part to remain in the treaty relationship (Knoppers 1993–94: 1.124–32).

64

Gary N. Knoppers

about them, labeling them as “a land of Cabul” (‫ ;ארץ כבול‬1 Kgs 9:13).23 Given the symmetrical relationship of reciprocity between Hiram and Solomon, Hiram was entirely within his rights to press such a commercial claim against his treaty partner.24 As we have seen, Chronicles shares the assumption that the towns needed work, but the relationship between the two parties is completely reversed. As a dedicated and grateful client, Huram decides to turn over 20 towns to his patron. By rebuilding the towns and resettling them, Solomon extends and strengthens his kingdom. The implication is the opposite of that in Kings. Instead of overreaching and having to concede part of his northern domain to his Phoenician financier, Solomon enhances the expanse of his domain by exploiting a generous gift made by his Tyrian subordinate. Having acquired these 20 towns in Galilee, Solomon proceeds to expand his reach to the Euphrates by going to Hamath-zobah (‫ )וילך שלמה חמת צובה‬and prevailing over it (‫ויחזק‬ ‫ ;עליה‬2 Chr 8:3).25

The Ties that Bind: Solomon and Huram Incorporated Subsequent interactions between the two potentates confirm Huram’s commitment to his patron. When Solomon embarks on a new economic venture by traveling south to Ezion-geber and to Eloth, Huram dispatches a fleet staffed by skilled merchant marines for the use of Solomon’s retainers (2 Chr 8:17). Huram’s sailors accompany Solomon’s servants to Ophir, and together they procure 450 talents of gold26 for the Israelite king (2 Chr 8:18).27 Again, Chronicles departs

23.  The precise meaning of the presumably derogatory term Cabul is, however, uncertain (Frankel 1992: 797; HALOT 458b). 24.  To say that the Solomon of Kings “appears as a cheat and swindler” (Nam 2012: 81) seems, however, to be too strong a claim to make. As Nam himself observes (2012: 66–76), international trading relationships between equals (or near equals) in the ancient Near East were characterized by an ongoing process of claims, counterclaims, and renegotiation. 25.  The notice, unparalleled in Kings, explains how Solomon came to rule over Transeuphrates (2 Chr 9:26). Nevertheless, it seems at odds with the divine pledge to David: “A son has been born to you; he will be a man of rest” (‫ ;איש מנוחה‬1 Chr 22:9). 26. So MT and LXX 2 Chr 8:18. MT 1 Kgs 9:28, “four hundred twenty.” The lemma of LXX 1 Kgs 9:28, “one hundred twenty,” assimilates to the figure provided in 1 Kgs 9:14; 10:10 (so also Klein 2012: 118). 27.  On the international reputation of Ophir as a source of gold, see 1 Kgs 10:11; Isa 13:12; Ps 45:10; Job 22:24; 38:16; 1 Chr 29:4. The location of Ophir is disputed: southwestern Arabia or the land of Punt (the northeastern African coast; Knoppers 2004a: 277).

The Economic Relationship between Huram and Solomon

65

Table 4  Kings

Chronicles

Solomon builds fleet at Ezion-geber

Topic

1 Kgs 9:26



Solomon travels to Ezion-geber and to Eloth



2 Chr 8:17

Hiram sends sailors for fleet

1 Kgs 9:27



Huram sends fleet with sailors



2 Chr 8:18

Hi/uram’s sailors travel with Solomon’s servants

1 Kgs 9:27

2 Chr 8:18

Gold talents from Ophir brought to Solomon 1 Kgs 9:28

2 Chr 8:18

from its source text, although not as radically as with the case of the 20 towns in Galilee. In 1 Kgs 9:26 “King Solomon makes a fleet (‫ )ואני עשה המלך שלמה‬at Ezion-geber (‫)בעציון־גבר‬, which is near Eloth (‫)אשר את־אלות‬.”28 Hiram then sends his experienced merchant marines to accompany Solomon’s servants on their travels (1 Kgs 9:27; see table 4). In both Kings and Chronicles, the Tyrian monarch enables Solomon to expand the development of Israel’s international economic position by making his employees’ naval expertise available to the Israelite king. But in Chronicles Huram voluntarily takes on added capital risk, because his ships are at stake in the joint operation.29 His generosity seems to be underscored, because the mercantile venture enriches Solomon enormously. 28. Interpreting ‫ את‬as signifying proximity (e.g., Judg 3:19; 4:11; 2 Kgs 9:27; HALOT 101b). 29.  The story assumes Tyre’s international reputation in trade and shipping, but it is initially unclear how the writers envisage the operation’s logistics. If Huram had sent the fleet by sea from Tyre to Ezion-geber or Eloth in the Gulf of Aqaba, the route would have been long and circuitous in the 10th century b.c.e. The proposals to explain this difficulty include sending the ships around Africa; sending the ships overland (Welten 1973: 37–38); disassembling the ships, sending the parts overland, and reassembling them on site (Ikeda 1991: 114–15); and sending them via canal from the Nile to the Red Sea and up to the Gulf of Aqaba (Posener 1938; Williamson 1982: 233). The most likely explanation is the last one, because the writers are living sometime after the reign of Darius I (522–486 b.c.e.), during whose tenure a waterway was established between Bubastis and the Red Sea (Herodotus 1.56–57; 2.158–59; cf. 4.44; Diodorus 1.33.9–10). In one of Darius’s Suez inscriptions, he claims to have had a canal dug facilitating maritime trade between Egypt and Persia (Kent 1989; DZc §3.7–12). Although Diodorus credits the completion of the Suez project to the time of Ptolemy, his dating is tendentious (Briant 2002: 476–77). See further Redmount (1995). In short, the rewriting evident in Chronicles assumes conditions extant in the 5th century b.c.e. (or later).

66

Gary N. Knoppers

By contrast, it is difficult to see how Huram benefits, unless his aim is to remain on good terms with his liege lord. In an aside within the Queen of Sheba story, the work revisits the joint Israelite-Tyrian mercantile venture.30 The employees of Huram and the employees of Solomon who brought gold from Ophir also brought almug wood and precious stones (2 Chr 9:10).31 From the almug wood, the king made ramps for the House of Yhwh and the royal palace, as well as lyres and harps for the musicians (2 Chr 9:11).32 Again, the version of Chronicles departs somewhat from its Vorlage (see table  5).33 In Kings, it is simply the fleet of Hiram that brought 30.  2 Chr 9:10 follows 1 Kgs 10:11–12 in providing a parenthetical notice set off from what precedes it by the use of the conjunctive and the adverb (‫)וגם‬. The digression, prompted by the register of goods brought by the Queen of Sheba, focuses on the imported almug wood that was used to construct particular palace and temple fixtures, as well as musical instruments for the singers. 31. In speaking of almug wood, I am following some Hebrew manuscripts: ‫אלמוגים‬, as in MT 1 Kgs 10:11–12; Song 3:10. The spelling of MT 2 Chr 9:10–11 (and MT 2 Chr 2:7) ‫אלגומים‬, “algum wood,” reflects metathesis. The nature of the wood is not yet known, but it is native to the Mediterranean area (2 Chr 2:7; cf. Ug. ʾalmg; Akk. elammakku, “building timber” [the latter is evidently a loanword into Akk.; CAD E 75–76; HALOT 58a]). LXX ξύλα πεύκινα, “fir wood.” MT 1 Kgs 10:11 adds ‫הרבה‬ ‫מאד‬, “(in) great abundance.” 32. Vis-à-vis ‫ מסלות‬in 2 Chr 9:11, my translation is based on the LXX (ἀναβάσεις), “ascents,” and ‫סללה‬, “assault ramp,” from the root ‫( סלל‬2 Sam 20:15; 2 Kgs 19:32 // Isa 37:33; Jer 6:6; Ezek 4:2; 17:17; 21:27; 26:8; Dan 11:15; HALOT 757b). The meaning of MT ‫ מסלה‬is much debated. A ‫ מסלה‬is often thought to refer to a track or highway firmed with fill or stones (HALOT 606a–b), but Dorsey (1991: 385–91) argues for the meaning of a gateway or entranceway, based on the use of Akk. mušlālu, “gatehouse, gate” (CAD 10 277). But see the response to Dorsey by Tidwell (1995), which views the ‫ מסלה‬as an approach road leading from the base of a mound to the main citygate. Within the other literary contexts in Chronicles where the term is found, a ‫ מסלה‬seems to form part of a larger architectural complex. Because our specific text refers to Solomon’s employing almug wood to build a ‫מסלה‬, a road does not seem to be in view. Hence, some prefer “bannisters” (e.g., Rudolph 1955: 222; cf. Vg. 1 Kgs 10:15, fulchra, “rails”) or “steps” (Vg. gradus, “stair”; Klein 2012: 141). Yet, the normal term for steps is ‫( מעלות‬e.g., 2 Chr 9:18). BDB reads with 1 Kgs 10:12, ‫מסעד‬, a hapax legomenon, perhaps referring to a support or buttress (so LXX, ὑποστηριγματα; Tg. J.) or, less likely, to a parapet (Noth 1968: 228). Weiss (1968: 130) thinks the Chronicler’s Vorlage read (or was understood as) ‫מצעד‬, “step.” Yet, if so, it is unclear why the text speaks of a ‫מסלה‬, rather than a ‫מצעד‬. On the use of the term ‫ מסלה‬elsewhere in Chronicles, see Knoppers 2004b: 864–65. 33.  See further Japhet 1993: 632–34, 638; Klein 2012: 130–31, 140–41. Klein comments that the stories about Huram’s trading adventures with Solomon (2 Chr 8:17– 18 // 1 Kgs 9:26–28 and 2 Chr 9:10–11 // 1 Kgs 10:11–12) form an inclusio around the story about the Queen of Sheba (2 Chr 9:1–9 // 1 Kgs 10:1–10), although it should be observed that the Sheba story continues in 2 Chr 9:12 (// 1 Kgs 10:13). Regard-

The Economic Relationship between Huram and Solomon

67

Table 5  Kings

Chronicles

Hiram’s fleet fetches almug and precious stones

Topic

1 Kgs 10:11



Huram/Solomon servants fetch almug and stones



2 Chr 9:10

Almug used for supports and musical instruments 1 Kgs 10:12

2 Chr 9:11

Incomparable almug

1 Kgs 10:12



Incomparable musical instruments in Judah



2 Chr 9:11

almug wood in great abundance and precious stones.34 Solomon’s retainers and Huram’s retainers go unmentioned. It would seem that the writers of Chronicles stipulate, in line with their previous assertion about the cooperation of Solomon’s and Huram’s forces, the involvement of both groups in this venture. If the writers blindly followed Kings, they would be tacitly admitting to Solomon’s total reliance on his Tyrian ally. It is one thing to cast Solomon as an important power in relation to Huram. It is another thing, however, to cast Solomon as a complete dependent. The writers of Chronicles maintain the relative power structure between the two. There is one final mention of the Solomon-Huram relationship. When waxing about how “all the drinking vessels of King Solomon were made of gold and all of the utensils of the Lebanon Forest Palace were made of refined gold and how silver was considered as nothing in the days of Solomon” (2 Chr 9:20), the writers supply an explanation: “For the king had ships sailing to Tarshish (‫כי־אניות למלך הלכות‬ ‫ )תרשיש‬with the servants of Huram. Once every three years, the ships of Tarshish would return bringing elephant ivories, gold, silver, apes, and baboons” (2 Chr 9:21). Chronicles thus hints at the establishment of a new phase in Solomon’s reign. Having enjoyed success in other commercial ventures, Solomon launches a new commercial initiative with Huram of Tyre that entails the involvement, if not creation, of Solomon’s own merchant marine navy. By means of this international enterprise, Solomon becomes even more of a player in and a beneficiary of transnational maritime trade (see table 6). less, the literary presentation in both works closely connects the activities of these foreign potentates. Each royal (in Chronicles) specifically praises Yhwh, the God of Israel, lavishes compliments on Solomon, and enhances the material well-being of the Israelite state. See further Ben Zvi 2006: 272–76. The “blessing other” (= “other who blesses”) motif should be considered part and parcel of the larger idealization of Solomon’s tenure (Knoppers 2015: 139–68). 34.  MT 1 Kgs 10:11 reads ‫אני‬, “ship” or collectively, “fleet.” Cf. 2 Chr 9:21 (MT ‫ ;אניות‬LXX = ‫)אני‬.

Gary N. Knoppers

68

Table 6  Kings

Chronicles

Gold surplus; silver worthless

Topic

1 Kgs 10:21

2 Chr 9:20

Solomon’s Tarshish ships and Hiram’s fleet

1 Kgs 10:22



Solomon’s ships to Tarshish with Huram’s crews —

2 Chr 9:21

Maritime bonanza

2 Chr 9:21

1 Kgs 10:22

The claims made in Kings are more modest. There, the text presents Solomon with his own Tarshish fleet plying the seas with the merchant navy of Hiram (1 Kgs 10:22). In this case, Tarshish functions as an attributive adjective (‫ )אני תרשיש‬describing the kind of ships Solomon possessed. In the narrative setting of Kings, both potentates possess their own merchant vessels but cooperate in a long-range lucrative trade. Such an arrangement comports with the existence of a bilateral treaty between the two realms. Alternatively, in the Chronistic narrative in which Solomon enjoys priority, Huram’s sailors serve on Solomon’s fleet traveling to Tarshish. But in both cases, the result is similar: “once every three years, the ships of Tarshish would return bringing gold and silver, elephant ivories, apes, and baboons.”35

Conclusion We have seen that in both Kings and Chronicles Solomon renews the diplomatic relationship that existed between David his father and the king of Tyre. But whereas the Deuteronomistic work clearly defines the Solomon-Hiram relationship as a parity treaty in which the two partners become “brothers” in a fictive kinship affiliation, Chronicles defines the relationship as a patron-client association in which Solomon plays the role of a master and Huram the role of a loyal adjutant. In both cases, Solomon strengthens the Israelite-Tyrian connection by tendering extremely generous compensation for those cutting and shipping the high-quality timber to Jerusalem. But in Kings, Solomon eventually becomes so indebted to his “brother” in Tyre that he is forced to relinquish 20 Galilean towns to satisfy his international obligations.36 The 35.  So MT 2 Chr 9:21. Compare with “Once every three years, the fleet of Tarshish would return carrying gold and silver, elephant ivories, apes, and baboons” (1 Kgs 10:22). 36.  Such a concession and the capture of Gezer by the pharaoh, donated as a send-off (‫ )שלחים‬to his daughter, Solomon’s wife (1 Kgs 9:16), complicate the picture of Solomon’s territorial independence in the post-temple-dedication phase of his tenure in Kings.

The Economic Relationship between Huram and Solomon

69

opposite is true in Chronicles. Having received Solomon’s large quantities of material gifts, Huram acts repeatedly to ingratiate himself with his patron. His gift of 20 Galilean towns to Solomon enables Solomon to expand his domain substantially in the north. His donation of Tyrian transport ships for Solomon’s commercial use effectively expands Israel’s reach in transnational trade. The Solomon-Huram relationship is, of course, only one facet of the much larger Chronistic reimagination of Solomon’s reign, but it is nonetheless significant. Commercial relations between the two leaders consistently enrich the Israelite state, because Solomon’s patronage of Huram allows Solomon to gain access to and exploit international markets. Huram plays the role of a grateful client, ever eager to assist his patron in expanding his economic reach. Recasting the SolomonHuram relationship in terms that would be familiar to their readers, the writers present Solomon as a quasi satrap who comes to exercise authority within the Levantine area of Transeuphratene. Much like the satrap of Beyond the River might employ Phoenician ships and merchant marines to enrich his domain (and that of his imperial king), Solomon exploits his ties to Huram to enrich his state and project power abroad. For his part, Huram supports each of Solomon’s trade initiatives, even risking his own mercantile assets in the process. Capitalizing on his international contacts and successful investments, Solomon becomes a major player in regional commerce. Flourishing global trade becomes, in turn, an unambiguous driver of Israelite wealth, and Solomon’s regime attracts the wealth of nations.37 However optimistic this utopian ideal may be, it is important to recognize what it implicitly concedes. The very framing of the alternative reality set in the past is couched in terms that acknowledge Israel’s traditional economic limitations. What Solomon as head of a large, militarily powerful, thriving, and united Israel bestows to his Tyrian ally are the traditional types of agricultural produce his land offers—wheat, barley, oil, and wine. Lacking in this list are precious metals (gold, silver), precious stones, spices, and other valuable natural commodities. 37.  Indeed, in one important comment found in Kings (1 Kgs 10:23), reproduced in Chronicles (2 Chr 9:22), Solomon is said to have exceeded all the kings of the land “in riches and wisdom.” The notice, which marks the fulfillment of the divine pledge to Solomon at Gibeon of incomparable wealth and wisdom (2 Chr 1:12–13; cf. 1  Kgs 3:12–13), places Solomon in a category that transcends the limits of the Transeuphratene satrapy. Inasmuch as “all the kings of the land were seeking the presence of Solomon to hear his wisdom . . . bringing each one his tribute” (2 Chr 9:23–24 // 1 Kgs 10:24–25), Solomon’s reputation extended beyond the borders of the region over which he is said to have ruled.

70

Gary N. Knoppers

In other words, for other important material resources—not just cedar, cypress, and almug wood—but also for precious metals, precious stones, spices, and exotic animals, Israelite leaders require control over or at least access to overland trade routes, critical ports, and international sea lanes. The very construction of a Solomonic ideal presupposes, therefore, the limitations of Israel’s indigenous natural resources. There is another important way in which the work concedes the economic and geopolitical realities of its own times. Within the Chronistic construction of monarchic history, Solomon appears as a unique individual, the monarch whom Yhwh made “exceedingly great in the sight of all Israel” (1 Chr 29:25; Knoppers 2004b: 957; Schweitzer 2007: 76–88). Compared with the rump state that survives disunion during the reign of his son, the large and influential kingdom of Solomon is clearly exceptional. In short, economically speaking, the Solomonic state is the exception that proves the rule of Judean material, economic, and geopolitical limitations. Nevertheless, the recasting of Solomon as a regional overlord who moves Israel beyond its traditional agricultural base to commandeer a variety of resources to become a leading regional participant in international trade is itself significant. If the united kingdom is a “better alternative reality,” as Schweitzer (2007: 76–88) calls it, that model reflects aspirations on the part of Judean scribes for something better than the realities of their own economic and political conditions.38 In this respect, the work employs a postexilic imperial political instrument to transcend the inherent limitations of that model. An overarching economic and political authority to which Solomon might himself be subject, such as the Great King, is never alluded to, unless one takes that authority to be the God of Israel.39 The utopian ideal is, therefore, not a small depopulated and economically challenged province ruled by others within the context of a vast empire but a people within its own land, whose lead38. My interpretation differs from that of Riley (1993: 168–204), Schweitzer (2007: 125–27), Klein (2006: 380–82), et al. in that I regard Davidic leadership as a constituent feature of the Chronistic ideal reality. Other political authorities, such as the Persian monarchs, may fulfill the duties of the Davidic Dynasty for a time in Jerusalem’s cultic worship and do so under divine blessing (2 Chr 36:22–23), but that situation is not cast as a utopian ideal. On this issue, see also Japhet 1999 and Boda 2009. 39.  The identification of the kingdom or kingship of Yhwh (1 Chr 17:14; 28:5; 29:11; 2 Chr 13:8), the throne of Yhwh (1 Chr 28:5; 29:23; 2 Chr 9:8), and the Temple of Yhwh (1 Chr 17:14; 28:6; 2 Chr 9:8) with the kingdom (or kingship), throne, and temple of David and Solomon is an integral part of the distinctive manner by which Chronicles depicts the united monarchy (Japhet 1989: 395–411; Kuntzmann 1993; Tiňo 2010; Knoppers 2015).

The Economic Relationship between Huram and Solomon

71

ers advance an economic agenda that enables Israel to leverage, if not transcend, its available resources to the enhancement of its own society.

Bibliography Ben Zvi, E. 2006 History, Literature and Theology in the Book of Chronicles. Bible World. London: Equinox. Block, D. I. 1997–98  The Book of Ezekiel. 2 vols. NICOT. Grand Rapids, MI: Eerdmans. Boda, M. J. 2009 Identity and Empire, Reality and Hope. Pp.  219–47 in Community Identity in Judean Historiography: Biblical and Comparative Perspectives, ed. G. N. Knoppers and K. A. Ristau. Winona Lake, IN: Eisenbrauns. Borowski, O. 2003 Daily Life in Biblical Times. SBLABS 5. Atlanta: Society of Biblical Literature. Briant, P. 1989 Table du roi: Tribut et redistribution chez les Achéménides. Pp. 35–44 in Les Tributs dans l’empire perse: Actes de la Table Rond de Paris, 12–13 décembre 1986, ed. P. Briant and C. Herrenschmidt. Paris: Peeters. 2002 From Cyrus to Alexander: A History of the Persian Empire. Winona Lake, IN: Eisenbrauns. Briant, P., and Herrenschmidt, C., eds. 1989 Le tribut dans l’Empire perse: Actes de la table ronde de Paris, 12–13 décembre 1986. Travaux de l’Institut d’études iraniennes de l’Université de la Sorbonne nouvelle 13. Paris: Peeters. Bruno, J. 2010 Höfischer Lebensstil und materielle Prachtentfaltung. Pp.  377–410 in Der Achämenidenhof, ed. B. Jacobs and R. Rollinger. Wiesbaden: Harrassowitz. Cogan, M. 2001 1 Kings. AB 11. Garden City, NY: Doubleday. Cohen, R., and Westbrook, R., eds. 2000a Amarna Diplomacy: The Beginnings of International Relations. Baltimore: Johns Hopkins University Press. 2000b Introduction: The Amarna System. Pp.  1–12 in Amarna Diplomacy: The Beginnings of International Relations. Baltimore: Johns Hopkins University Press. Dalby, A. 2013 Les sources du luxe du roi Salomon. Pp.  311–25 in Le banquet du monarque dans le monde antique, ed. C. Grandjean, C. Hugoniot, and B. Lion. Tables des hommes. Rennes: Presses universitaires de Rennes.

72

Gary N. Knoppers

Dillard, R. B. 1984a The Literary Structure of the Chronicler’s Solomon Narrative. JSOT 30: 85–93. 1984b Reward and Punishment in Chronicles: The Theology of Immediate Retribution. WTJ 46: 164–72. 1987 2 Chronicles. WBC 15. Waco, TX: Word. Dorsey, D. A. 1991 The Roads and Highways of Ancient Israel. Baltimore: Johns Hopkins University Press. Dyck, J. E. 1998 The Theocratic Ideology of the Chronicler. BibInt 33. Leiden: Brill. Edelman, D. 2006 Tyrian Trade in Yehud under Artaxerxes I: Real or Fictional? Independent or Crown Endorsed? Pp. 207–46 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. Elayi, J. 1987 Recherches sur les cités phéniciennes à l’époque perse. Naples: Istituto Universitario Orientale. 1990 Economie des cités phéniciennes sous l’empire perse. Naples: Istituto Universitario Orientale. Frankel, R. 1992 Cabul. Pp. 797 in vol. 1 of ABD. Ikeda, Y. 1991 King Solomon and His Red Sea Trade. Pp.  113–32 in Near Eastern Studies: Dedicated to H.I.H. Prince Takahito Mikasa on the Occasion of His Seventy-Fifth Birthday, ed. M. Mori, H. Ogawa, and M. Yoshikawa. Bulletin of the Middle Eastern Culture Center in Japan 5. Wiesbaden: Harrassowitz. Japhet, S. 1989 The Ideology of the Book of Chronicles and Its Place in Biblical Thought. BEATAJ 9. Frankfurt am Main: Peter Lang. 1993 I and II Chronicles. OTL. Louisville: Westminster John Knox. 1999 Exile and Restoration in the Book of Chronicles. Pp. 33–44 in The Crisis of Israelite Religion: Transformation of Religious Tradition in Exilic and Post-Exilic Times, ed. B. Becking and M. C. A. Korpel. OtSt 42. Leiden: Brill. [Repr. pp. 331–41 in From the Rivers of Babylon to the Highlands of Judah. Winona Lake, IN: Eisenbrauns, 2006.] Jigoulov, V. S. 2010 The Social History of Achaemenid Phoenicia. London: Equinox. Jonker, L. C. 2008 The Chronicler’s Portrayal of Solomon as the King of Peace within the Context of the International Peace Discourses of the Persian Era. OTE 1: 653–69. Joyce, P. M. 2007 Ezekiel: A Commentary. LHBOTS 482. New York: T. & T. Clark.

The Economic Relationship between Huram and Solomon

73

Kalluveettil, P. 1982 Declaration and Covenant. AnBib 88. Rome: Pontifical Biblical Institute. Kent, R. G. 1989 Old Persian: Grammar, Texts, Lexicon. 2nd ed. AOS 33. New Haven, CT: American Oriental Society. Klein, R. W. 2006 1 Chronicles. Hermeneia. Minneapolis: Fortress. 2012 2 Chronicles. Hermeneia. Minneapolis: Fortress. Klinkott, H. 2007 Steuern, Zölle, und Tribute im Achaimenidenreich. Pp.  263–90 in Geschenke und Steuern, Zölle und Tribute: Antike Abgabenformen in Anspruch und Wirklichkeit, ed. H. Klinkott, S. Kubisch, and R. MüllerWollermann. CHANE 29. Leiden: Brill. Knoppers, G. N. 1993–94  Two Nations under God: The Deuteronomistic History of Solomon and the Dual Monarchies. HSM 52–53. Atlanta: Scholars Press. 1996 ‘Yhwh Is Not with Israel’: Alliances as a Topos in Chronicles. CBQ 58: 601–26. 2004a I Chronicles 1–9. AB 12. New York: Doubleday. 2004b I Chronicles 10–29. AB 12A. New York: Doubleday. 2015 Judah, Levi, David, Solomon, Jerusalem, and the Temple: Election and Covenant in Chronicles. Pp.  139–68 in Covenant and Election in Exilic and Post-Exilic Judaism: Studies of the Sofja Kovalevskaja Research Group on Early Jewish Monotheism, vol. 5, ed. N. MacDonald. FAT 2/79. Tübingen: Mohr Siebeck. forthcoming  When the Foreign Monarch Speaks about the Israelite Tabernacle. In History, Memory, Hebrew Scriptures: A Festschrift for Ehud Ben Zvi, ed. I.  D. Wilson and D. V. Edelman. Winona Lake, IN: Eisenbrauns. Kuhrt, A. 2010 The Persian Empire: A Corpus of Sources from the Achaemenid Period. London: Routledge. Kuntzmann, R. 1993 Le Trône de Dieu dans l’oeuvre du Chroniste. Pp. 19–27 in Le Trône de Dieu, ed. M. Philonenko. WMANT 69. Tübingen: Mohr. Lemaire, A. 1996 Nouvelles inscriptions araméennes d’Idumée au Musée d’Israël. Transeuphraténe Supplement 3. Paris: Gabalda. 1999 Der Beitrag idumäischer Ostraka zur Geschichte Palästinas im Übergang von der persischen zur hellenistischen Zeit. ZDPV 115: 12–23. 2001 Nouvelles Tablettes Araméennes. Hautes Études Orientales 34. Moyen et Proche-Orient 1. Geneva: Droz. Lewis, D. M. 1989 Persian Gold in Greek International Relations. Revue des études anciennes 91: 227–34.

74

Gary N. Knoppers

Lincoln, B. 2012 ‘Happiness for Mankind’: Achaemenian Religion and the Imperial Project. Acta Iranica 53. Leuven: Peeters. Lipschits, O.; Koch, I.; Shaus, A.; and Guil, S. 2010 The Enigma of the Biblical Bath and the System of Liquid Volume Measurement in the First Temple Period. UF 42: 453–78. Lipschits, O., and Vanderhooft, D. S. 2011 The Yehud Stamp Impressions: A Corpus of Inscribed Impressions from the Persian and Hellenistic Periods in Judah. Winona Lake, IN: Eisenbrauns. Liverani, M. 1990 Prestige and Interest: International Relations in the Ancient Near East circa 1600–1100 b.c.e. Padua: Sargon. Miller, M. C. 2010 Luxury Toreutic in the Western Satrapies: Court-Inspired GiftExchange Diffusion. Pp. 853–97 in Der Achämenidenhof, ed. B. Jacobs and R. Rollinger. Wiesbaden: Harrassowitz. Mitchell, L. 1997 Greeks Bearing Gifts: The Public Use of Private Relationships in the Greek World, 435–323 b.c. Cambridge: Cambridge University Press. Naʾaman, N. 2000 The Egyptian-Canaanite Correspondence. Pp. 125–38 in Amarna Diplomacy: The Beginnings of International Relations, ed. R. Cohen and R. Westbrook. Baltimore: Johns Hopkins University Press. Nam, R. S. 2012 Portrayals of Economic Exchange in the Book of Kings. BibInt 112. Leiden: Brill. Noth, M. 1968 Könige 1. BKAT 9/1. Neukirchen-Vluyn: Neukirchener Verlag. Nussbaum, G. B. 1959 The Captains in the Army of the Ten Thousand. Classica et Mediaevalia 20: 16–29. 1967 The Ten Thousand: A Study in Social Organization and Action in Xenophon’s ‘Anabasis.’ Social and Economic Commentaries on Classical Texts 4. Leiden: Brill. Parpola, S., and Porter, M. 2001 The Helsinki Atlas of the Near East in the Neo-Assyrian Period. Helsinki: Neo-Assyrian Text Corpus Project. Pohlmann, K.-F. 1996–2001  Das Buch des Propheten Hesekiel/Ezechiel. 2 vols. ATD 22. Göttingen: Vandenhoeck & Ruprecht. Posener, G. 1938 Le canal du Nil à la Mer Rouge avant les Ptolémées. Chronique d’Égypte: Bulletin périodique de la Fondation égyptologique reine Élisabeth 25–26: 259–73. Pury, A. de 2003 Salomon et la reine de Saba: L’analyse narrative peut-elle se dispenser de poser la question du contexte historique? Pp. 213–38 in La

The Economic Relationship between Huram and Solomon

75

Bible en récits: L’exégèse biblique à l’heure du lecteur, ed. D. Marguerat. MdB 48. Geneva: Labor et fides. Rainey, A., and Notley, S. 2006 The Sacred Bridge: Carta’s Atlas of the Biblical World. Jerusalem: Carta. Redford, D. B. 2006 A History of Ancient Egypt: Egyptian Civilization in Context. Dubuque, IA: Kendall/Hunt. Redmount, C. A. 1995 The Wadi Tumilat and the “Canal of the Pharaohs.” JNES 54: 127–35. Riley, W. 1993 King and Cultus in Chronicles: Worship and the Reinterpretation of History. JSOTSup 160. Sheffield: Sheffield Academic Press. Römer, T. C. 2008 Salomon d’après les Deutéronomistes: Un roi ambigu. Pp.  98–130 in Le roi Salomon, un héritage en question: Hommage à Jacques Vermeylen, ed. C. Lichtert and D. Nocquet. Livre et le rouleau 33. Brussels: Lessius. Root, M. C. 1979 The King and Kingship in the Achaemenid Art. Acta Iranica 19. Leiden: Brill. Rop, J. 2013 All the King’s Greeks: Mercenaries, Poleis, and Empires in the Fourth Century b.c.e. Ph.D. dissertation. The Pennsylvania State University. Roy, J. 1967 The Mercenaries of Cyrus. Historia: Zeitschrift für Alte Geschichte 16: 287–323. Rudolph, W. 1955 Chronikbücher. HAT 21. Tübingen: Mohr. Sancisi-Weerdenburg, H. 1989 Gifts in the Persian Empire. Pp.  129–45 in Le tribut dans l’Empire perse: Actes de la table ronde de Paris, 12–13 décembre 1986, ed. P. Briant and C. Herrenschmidt. Travaux de l’Institut d’études iraniennes de l’Université de la Sorbonne nouvelle 13. Paris: Peeters. Schweitzer, S. J. 2007 Reading Utopia in Chronicles. LHBOTS 442. London: T. & T. Clark. Stolper, M. 1989 The Governor of Babylon and Across-the-River in 486 bc. JNES 48: 283–305. 1992 Babylonian Evidence for the End of the Reign of Darius I: A Correction. JNES 51: 61–62. Throntveit, M. A. 1988 Hezekiah in the Books of Chronicles. Pp. 302–11 in Society of Biblical Literature 1988 Seminar Papers, ed. D. J. Lull. Atlanta: Scholars Press. 2003 The Relationship of Hezekiah to David and Solomon in the Books of Chronicles. Pp. 105–21 in The Chronicler as Theologian: Essays in Honour of Ralph W. Klein, ed. M. P. Graham, S. L. McKenzie, and G. N. Knoppers. JSOTSup 371. London: T. & T. Clark.

76

Gary N. Knoppers

Tidwell, N. L. 1995 No Highway! The Outline of a Semantic Description of mesillâ. VT 45: 251–69. Tiňo, J. 2010 King and Temple in Chronicles: A Contextual Approach. FRLANT 234. Göttingen: Vandenhoeck & Ruprecht. Trundle, M. 2004 Greek Mercenaries: From the Late Archaic Period to Alexander. New York: Routledge. 2010 Coinage and the Transformation of Greek Warfare. Pp. 227–52 in New Perspectives on Ancient Warfare, ed. G. Fagan and M. Trundle. History of Warfare 59. Leiden: Brill. Weiss, R. 1968 Textual Notes. Textus 6: 125–32. Welten, P. 1973 Geschichte und Geschichtsdarstellung in den Chronikbüchern. WMANT 42. Neukirchen-Vluyn: Neukirchener Verlag. Westbrook, R. 2000 International Law in the Amarna Age. Pp.  28–42 in Amarna Diplomacy: The Beginnings of International Relations, ed. R. Cohen and R. Westbrook. Baltimore: Johns Hopkins University Press. Wiesehöfer, J. 1980 Die “Freunde” und “Wohltäter” des Grosskönigs. Studia Iranica  9: 7–21. 1989 Tauta gar en atelea: Beobachtungen zur Abgabenfreiheit im Achaimenidenreich. Pp. 183–91 in Le tribut dans l’Empire perse: Actes de la table ronde de Paris, 12–13 décembre 1986, ed. P. Briant and C. Herrenschmidt. Travaux de l’Institut d’études iraniennes de l’Université de la Sorbonne nouvelle 13. Paris: Peeters. Williamson, H. G. M. 1982 1 and 2 Chronicles. NCB. Grand Rapids, MI: Eerdmans. Würthwein, E. 1977 Die Bücher der Könige: 1 Könige 1–16. ATD 11/1. Göttingen: Vandenhoeck & Ruprecht. Young, T. C., Jr. 1992 Persepolis. Pp. 236–37 in vol. 5 of ABD. Zaccagnini, C. 2000 The Interdependence of the Great Powers. Pp. 141–53 in Le tribut dans l’Empire perse: Actes de la table ronde de Paris, 12–13 décembre 1986, ed. P. Briant and C. Herrenschmidt. Travaux de l’Institut d’études iraniennes de l’Université de la Sorbonne nouvelle 13. Paris: Peeters.

Agrarian Economy through City-Elites’ Eyes: Reflections of Late Persian Period Yehud Economy in the Genealogies of Chronicles Louis Jonker

Stellenbosch University

Introduction The genealogies in Chronicles have been the focus of many studies in the distant and recent past. Not many studies have yet been devoted to the economic conditions reflected in these lists, however.1 Questions such as the following are not often investigated: What perspectives on demographic distribution, landownership, agrarian tribes and power relations (to name a few issues) are evident in 1 Chronicles 1–9? And how do these perspectives relate to themes such as temple economy, legal system, and the like in the rest of the book of Chronicles? In the investigation of these issues, one should of course remain aware of the fact that this literature does not necessarily reflect real-life conditions of the time that is discussed and, even more so, of the time during which the writing down of this literature took place. Like any other literature, the genealogical introduction to Chronicles wants to achieve some rhetorical aims. Thus far, scholars have mainly focused on socioreligious, sociopolitical, and social identity perspectives when Author’s note:  I acknowledge the financial support of the National Research Foundation of South Africa for attending the annual conference of the Canadian Society of Biblical Studies 2013, where I presented an oral version of this essay. Any opinion, finding, and conclusion or recommendation expressed in this material is mine, not the NRF ’s. 1.  Horsley (1991: 163) argues in favor of a political and economic interpretation of the Bible when he says: [B]iblical literature and history are not ‘religious’ and do not deal with ‘religious’ matters in a way separate for or separable from other dimensions of life, such as the ‘political’ or the ‘economic’. ‘Religion’ is embedded with kinship and/or local community life and/or ‘the state’ in virtually any traditional agrarian society, and hence is inseparable from political and economic matters. Our scholarly habit of pretending that we are dealing with primarily religious texts or institutions, which means basically imposing modern presuppositions and concepts on historical materials that had neither concepts nor terms for religion, blocks rather than enhances understanding.

77

78

Louis Jonker

discussing the rhetorical thrust of the genealogies. However, another perspective that might prove to be helpful for our understanding of this prominent part of the Chronicler’s work is a socioeconomic perspective. What did the Chronicler want to achieve economically through the inclusion of these genealogies? When dealing with this question, I should make three important preliminary remarks: First, the perspectives reflected in the Chronicler’s genealogical introduction are those of an author belonging to the cultic elite in Jerusalem in the Persian province of Yehud, most likely in a time toward the end of the Persian imperial dominion.2 Second, this study is undertaken as a biblical studies contribution in a research project, which of necessity should be multidisciplinary. Since my own field of specialization is biblical studies and not economy or archaeology, one should remain aware of the fact that the results of this investigation should be brought into interaction with, inter alia, specialized research on historical economy and with the insights which can be offered by archaeology on the material culture of the said time. And last, one should remain aware of the fact that the methodological angle taken in this investigation will codetermine the outcome. If one uses inappropriate economic models to read texts from ancient times, one’s results may prove to be worthless.3 This point deserves further attention, and the next section will reflect more fully on it.

Methodological Reflection In a recent article, Roland Boer presupposes that economic analysis is (and should be) central in historiographical study of the ancient Near East (Boer 2007: 30). He therefore argues that the Marxist economically informed ancient Near Eastern scholarship from the former Soviet Union can potentially be of much value for our research. He indicates that, due to the ideological bias of the Cold War era, Western scholarship of the ancient Near East did not take any notice of the economically informed scholarship in the former Soviet Union and that Western scholarship still remains oblivious to this vast scholarly resource.4 Applying some insights from this scholarship, he thereafter traces and analyzes what he identifies as “the key nodes of the sacred economy of Ancient Israel” (Boer 2007). He begins by tracing the emergence of 2.  See the various commentaries on Chronicles for arguments supporting this statement. 3.  See, for example, Bang, Ikeguchi, and Ziche 2006 for a discussion of the applicability of modern methodologies in the studying of ancient economies. 4.  Boer (2007) admits, however, that later Western scholarship has gone into directions that are similar to the former Soviet ancient Near Eastern scholarship.

Yehud Economy in the Genealogies of Chronicles

79

the state in response to the conflict between the village commune and the temple-city complex. He then proceeds from this insight to identify the constituent elements of what he calls the abiding system of theoeconomics and its constituent regimes of allocation and extraction. Boer is not the only voice in recent scholarship that has argued in favor of employing Marxist models of analysis in our ancient Near Eastern, and therefore, biblical scholarship (see Chaney 2006: 145–60; Gottwald 1993: 3–22; Horsley 2009; Jobling 2007: 184–210; West 2011: 511–32). And these scholars have made us aware of the fact that our models matter when it comes to economic and political issues. In the present research project on agrarian economies in ancient Persian times, this methodological point is certainly valid. The models we use will determine the outcome of our investigations. In my contribution, I therefore would like to take up Boer’s challenge to consider these economically informed models from Marxist scholarship in an experiment to read the Chronicler’s genealogies economically.5 The explanatory resources that I find particularly helpful in Roland Boer’s proposal are his understanding of a sacred economy in which a tension exists between the economic systems of the village commune and the temple-city complex, as well as his understanding of so-called theo-economics, which is built on regimes of allocation and extraction. Boer (2007: 34) defines sacred economy as “a system in which the economy operates and is understood in terms of the sacred rather than the political.” He sees the village commune “as both a kinship and economic unit, [which] persists through a series of economic and sociopolitical crises” (Boer 2007: 35).6 The term temple-city complex designates for him (2007: 36) the way the city (what we would call a town) is unthinkable without the structuring role of the temple and indeed the palace which is an add-on 5.  In doing so, my intention will be the same as West’s modest aim when he attempted a similar experiment in a recent article. He emphasized the necessity of keeping economic models on the exegetical table, particularly those economic models that have become unfashionable but that retain considerable explanatory power. Marxist analytical categories have fallen out of favour in much of biblical scholarship, even within postcolonial biblical criticism, and while economic analysis . . . might not be able to explain fully the particularities and complexities of Ancient Near Eastern . . . contexts, Marxist models offer explanatory and exegetical resources to biblical scholarship. (West 2011: 530)

6.  Boer (2007: 36) refers to Carter’s work when he indicates that “[a]t least until the Persian era . . . there was no substantial change in the structure and nature of the village commune, except for a significant reduction in the number of village communes in Persian Yehud.”

80

Louis Jonker to the temple. . . . In contrast to the village commune, the temple-city complex underwent a constant process of change from Sumer to the Hellenistic era. The two are, however, inextricably connected, much like a wheel: the village communes comprise the multiple points at the ends of the spokes, all of which have some connections with the axle.

Boer cautions, however, that the temple-city complex should not be equated with the state. He says (2007: 36): “Rather, the state arises in a tension between the village commune and the temple-city complex.” Within this context, Boer’s indications on the mediations between empire and village commune are informative (and this view will specifically be taken up in our analysis of the Chronicler’s genealogies below): What the studies thus far do not consider sufficiently are the various mediations of the relationship between the despotic empire and village commune. This is especially true of small vassal states such as Israel, and indeed cities like Jerusalem, that stand between empire and village commune. . . . Not only is the vassal state a crucial mediator between empire and village commune, but it may also act as a buffer [my emphasis]. (Boer 2007: 38)

A last aspect of Boer’s study that should be emphasized is his understanding of what he calls theo-economics. He proposes (2007: 39) “that we speak of the sacred economy of the Ancient Near East (including Israel) in terms of a tension between allocation and extraction, or between an allocative economics and an extractive economics.” According to Boer, extraction plays a minor role in comparison with allocation in the ancient Near East, and extraction can also be disguised as allocation. It is particularly in the area of tribute that this tension is manifest. Boer explains (2007: 41–42): Tribute . . . is not the determining feature of the sacred economy. . . . Indeed, the primary role of tribute was the maintenance of the temple-city complex and the larger imperial state. However, even though it is not the determining feature, tribute is the point at which the main tension of the sacred economy shows up. We can now see that the tension I noted earlier between the village commune and the temple-city complex manifests itself in the tribute extracted by the latter from the former. Yet, the fact that tribute was a problem within a sacred economy shows up in the way that it was justified in terms of an allocative ideology. Thus the deity was the one who sanctioned tribute, most notably in the notion of the temple tax or tithe, a reallocation of produce for the sake of the priests and the material structure of religion [my emphasis].

Particularly the allocative issues of land and family or kinship, and the extractive notion of tribute will prove to be quite helpful for our understanding of the Chronicler’s genealogies. However, before applying

Yehud Economy in the Genealogies of Chronicles

81

some of Boer’s methodological insights to our interpretation of Chronicles as a perspective of city-elites on the agrarian economy of the late Persian era, we should now review the textual data.

The Textual Data City/Land Possession/Occupation/Inhabitation in 1 Chronicles 1–9 When an economic perspective is taken on the genealogies of 1 Chronicles 1–9, the issue of allocated or possessed land emerges from the textual data as one of the most prominent features. Numerous indications are included in these chapters about towns or pasturelands in Palestine and neighboring regions that were inhabited by various tribal groups—all descendants of Israel. Apart from the very first list of early humanity from Adam to Abraham, where no place or regional names occur (1:1–27),7 all other parts of the genealogical introduction to Chronicles abound with these indications.8 One can therefore safely agree with Kartveit and others that the issue of land possession or occupation was integral to this extensive description of who Israel was in the postexilic period (Kartveit 1989). Scholars agree that the genealogies were supposed to provide the Chronicler’s definition of All-Israel that is so prominently featured in the rest of Chronicles. Integral to the understanding of who All-Israel comprised was a certain view of what land and which towns belonged to the constituent parts of this people. The following observations are important to our study: First, the sections dealing with the tribes of Judah (2:3–4:23) and Benjamin (8:1–9:1a), as well as with Jerusalem’s inhabitants (9:1b–34) and Saul’s (Benjaminite) descendants contain relatively few place or regional names when compared with the rest of the genealogies. Second, Jerusalem occurs in the parts mentioned in the previous point (in 3:4b–5 in connection with the Judahite, King David; 8:28, 32 in connection with Benjaminite heads of ancestral houses and chiefs; 9:3 in connection with people from Judah, Benjamin, Ephraim, and Manasseh; and 9:38 again in connection with some Benjaminites). The explicit mentioning of Jerusalem also occurs three times in the part where the focus is on Levi (5:27–6:66). The first two instances (5:36, 41) mention that the city was the location of Solomon’s temple, while the last (5:41) indicates that Jerusalem 7.  Many names in this first list are of course associated with ancient regions or powers, such as Cush, Egypt, Put, Canaan, and Sheba. However, in the genealogies these names are presented as individual ancestors. 8.  For a discussion of the land of Edom being treated as part of Abraham’s list of descendants in 1:28–2:2, see for example, Willi 2009: 43–46, 52–53; and Assis 2006: 287–302.

82

Louis Jonker

was taken into exile together with Judah. Jerusalem is not mentioned in the parts dealing with Simeon’s descendants, with the Transjordanian tribes (Reuben, Gad, and the half-tribe of Manasseh), or with the Northern tribes (Issachar, Naphtali, Manasseh, Ephraim, and Asher). Third, the verb yšb and the related plural noun yošebîm occur most frequently as an indication of inhabitation and/or possession. This term is descriptive and denotes a settled existence in a certain location. Fourth, in a few instances the transfer of possession through violent means of some sort is indicated (2:23 with lqḥ, 4:41 with nkh and ḥrm, 4:43 again with nkh, 5:10 and 19 with ʿsh milḥamah, and 7:28 with ʾḥz). Lastly, the term ntn occurs exclusively in the part where the focus is on Levi. No other term of possession occurs in this part of the genealogies. The term ntn was borrowed from the source text in Joshua 21—which will be discussed more fully in the next subsection. From these initial observations, it already becomes clear that the section where the focus is on Levi occupies a prominent role in the genealogies. The discussion will now focus on this text. Levite Settlements The long section in 1 Chr 5:27–6:66 (6:1–81 in most English translations) covers various aspects of the lineage of Levi (Levin 2004: 601–36). Levi, one of the 12 sons of Israel mentioned in 1  Chr 2:1–2, is credited with being the ancestor who gave Israel its priestly lineage. The very extensive genealogy therefore starts (5:27–41[1–15]) with the list of Kohathite priests who served in Jerusalem until Judah went into the Babylonian Exile. Then (6:1–15[16–31]) a more general genealogy of the Levite families is presented, followed by a list (6:16–34[32–50]) of those who were appointed as singers in the “House of the Lord.” A short passage (6:35–38[51–54]) returns to the priestly lineage, before the final section (6:39–66[54–81]) provides an elaborate description of the cities and land that were allotted to the priests and Levites. Apart from the occasional mentioning of Jerusalem (as described above), this last section is the only part of the Levite genealogy in which place or region names occur. However, the density of localities in this section makes it the most prominent part of the genealogies in terms of place-names. It is clear that the list in 6:39–66[54–81] derives from the narrative text in Joshua 21, which reports the allotment of dwelling places to the Levites during the conquest of the land.9 However, the Chronicler made 9.  Scholars agree that the narrative about the allocation by lot of Levite cities in Josh 21:5–40 is closely related to Priestly tradition. The introduction to this narrative states in vv. 1–3 that the Levite family heads came to Eleazar the priest, to Joshua, and to the tribal family heads of Israel and reminded them of the command of Yah-

Yehud Economy in the Genealogies of Chronicles

83

interesting changes in the order of the material in Joshua 21. After a general introduction to the list of Levitical cities in v. 39[54], the Chronicler first offers a detailed list of the places given to the descendants of Aaron who were from the Kohathite clan in vv. 40–45[55–61] (that is, the priestly line). Although this material is also present in Joshua 21 (in vv. 10–19), the Chronicler shifted it earlier in his presentation. By first presenting the Aaronite cities, the Chronicler has given prominence to this part of the Levitical lineage, emphasizing the status of the Aaronites within the priesthood, and highlighting their connection to the tribe of Judah (which is presented in 2:3–4:23, with King David being its most prominent member), but also of Benjamin.10 weh through Moses that they be given towns to live in. This introduction refers to the command that is stated in Num 35:2, a text that is considered by scholars to be of Priestly origin (see, for example, Ben Zvi 1992: 77–106). Although there are different views on the historicity and setting of the Levite city narrative in Joshua 21, scholars agree that this chapter is clearly associated with Priestly tradition. The same would then apply to the reflection of this narrative in the Chronicler’s Levite city list. There is more or less a consensus in scholarship about the priority of Joshua 21 in comparison to the Levite town list in 1 Chronicles 6. Although alternative views have been formulated in the past (e.g., by Auld who saw the relation between the two texts in the inverse direction), the majority of scholars now agree that the Chronicler made use of the Joshua text. For a good summary of the debate, as well as for additional arguments why Joshua 21 should be seen as the earlier text that served as source for the Chronicler, see Oeming 1990: 146–49. 10.  Sparks (2008: 154–55) explains the reorganizing of material from Joshua 21 in terms of a theological motif—namely, the contrast between faithfulness/possession and unfaithfulness/exile. He is of the opinion that this contrast is constructed in various ways by the genealogies in conjunction with some narratives in Chronicles. He comments on the position of the Aaronite towns in the list: This connection between exile and dwellings may further explain why the Chronicler brought the list of the towns of the sons of Aaron forward to a position prior to the summary statements. As the ‘exile’ had been explicitly mentioned in relation to the sons of Aaron, and not to the other Levitical clans, so also must a mention of the ‘dwelling places’ be explicitly made to the sons of Aaron. If the Chronicler had allowed the reference to the towns of the sons of Aaron to remain in its Joshua 21 position, the dwellings could be taken to refer to all of the Levitical clans, and possibly bring the legitimacy of the sons of Aaron, who had been exiled, into question. By drawing attention to the current dwelling places of the sons of Aaron, the Chronicler is indicating that those actions which caused the exile of the sons of Aaron have been restored not only to their lands, but also to their position within the cult. The sons of Aaron still retain the duty of offering sacrifice and making atonement for the people.

Although this interpretation in terms of the faithfulness of the Levites and Aaron­ ites is certainly valid in terms of the textual data, the present study rather takes an economic perspective to explain the textual data below.

84

Louis Jonker

After the extensive description of Aaronite cities, there are three short summaries of the towns given to Kohath’s descendants (v. 46[61]), to the descendants of Gershom (v.  47[62]), and to the descendants of Merari (v. 48[63]). Similar to Joshua 21, the Levitical tribe did not receive their own tribal land, but land and cities belonging to other tribes were allocated to them (described with the verb ntn). The summary section in vv. 46–48[61–63] concludes with the general remark in v. 49[64] that the Israelites gave (again with ntn) the Levites these towns and their pasturelands. Verse 50[65] is misplaced in the context of the Chronicler’s list. Its equivalent in Josh 21:9 introduces the information given in vv. 10–19, but with these verses being moved to an earlier position in Chronicles, the logical order is broken. Verse 50[65] would have made good sense if it had preceded v. 39[54]. In its new position, v. 50[65] is supposed to introduce the more detailed indications of land allocations in the following verses, but the indications of donor tribes do not fit the following verses.11 From 1 Chr 6:51[66], the order of presentation returns to the order of Joshua 21. Verses 51–55[66–70] contain a detailed description of the Kohathite cities, vv. 56–61[71–76] of the Gershomite cities, and vv. 62– 66[77–81] of the Merarite cities. Land Beneficiaries and Land Allocation in the Context of the Chronicler’s Genealogies Table 1 charts the land beneficiaries and allocations that emerge specifically from 1 Chr 6:39–66[54–81]. What would be the function of this picture within the context of the Chronicler’s genealogies? From various studies on the overall composition of 1 Chronicles 1–9, the consensus has emerged that the Levite genealogy in 1 Chr 5:27–6:66[6:1–81] occupies the central position in a ring structure, with the genealogies of Judah and Benjamin forming the other foci at the beginning and end of the composition.12 The other 11. See Dirksen (2005: 111): Like the parallel text Josh. 21:8, v. 49 concludes the preceding passage. V. 50, however, has a different function from the parallel text, Josh. 21:9. The latter verse opens a new passage in which the cities are mentioned by name; but v. 50 forms part of the conclusion of the preceding passage, in which the mention of Judah and Benjamin refers back to vv. 40–45. After v. 49, which has a general conclusory nature, v. 50 is actually superfluous.

See also Japhet 1993: 161: “Any view of the structure of the list must regard this verse as misplaced.” 12.  For a summary of views, see Levin 2004: 609–11.

Yehud Economy in the Genealogies of Chronicles

85

tribes feature in less prominent parts of the overall construction, with Reuben, Gad, and the half-tribe of Manasseh occupying the position before the Levite center, and the tribes of Issachar, Naphtali, Manasseh, Ephraim, and Asher thereafter. The overall construction is framed by the universalistic introduction from Adam to Abraham, the more particularistic summary of Israel’s sons at the beginning, and a list of Jerusalem’s inhabitants and cultic officials at the end, clearly reflecting the Chronicler’s own time. In terms of the geographical indications, it is noteworthy that Jerusalem is only mentioned in the focal points of the genealogies— namely, in the sections on Judah, Levi, and Benjamin, and then again in the frame at the end, where Jerusalem’s inhabitants are indicated. Kartveit’s dissertation (1989: 166), which traces the motifs and layers of a land theology in the Chronicler’s genealogies comes to the following conclusion: “Bei der Untersuchung des ersten Kapitels hat sich ein Weltgeographie ergeben, die die bekannte Welt in einem grossen Kreis beschreibt. Zentrum und Spitze ist Israel.”13 Kartveit explains how the Chronicler, taking his point of departure as Jerusalem, started with Judah, continued to the south (with Simeon), then to the Transjordan from south to north (with Reuben, Gad, and the halftribe of Manasseh in the south, then to Issachar and Naphtali in the north), then southward again on the western side of the Jordan (with Manasseh and Ephraim), ending with a short detour to the northwest (with Asher), and via Benjamin back to Jerusalem. Kartveit (1989: 167) remarks the following about this circular description: “Dieser grosse Kreis mit Jerusalem als Anfang und Ende . . . ist ein geschlossenes Ganzes, in der die Konzentration auf Judäa und Jerusalem . . . deut­ lich wird.”14 Oeming, in his published dissertation, takes this one step further when he interprets this “world geography” identified by Kartveit as a deliberate attempt to define the true Israel. He (Oeming 1990: 218) indicates: So dienen die Genealogien in der alttestamentlichen Spätzeit, in einem Israel ohne eigenen Staat und unter fremder Obrigkeit, der Selbst­definition und Selbstvergewisserung Israels. Sie bezeugen ein enormes Selbstbewusstsein der Jerusalemer Tempelgemeinde, besonders ihres Kultpersonals. [D]iese 13.  English translation: “In the investigation of the first chapters, a world geography emerged that describes the known world in big circles. At the center and climax is Israel.” 14.  English: “This big circle with Jerusalem as beginning and end forms a closed system in which the concentration on Judah and Jerusalem is clear.”

86

Louis Jonker “Partei” [beansprucht] den Titel “Israel” exclusiv für sich: Wir—und nur wir—, die wir von diesen Stammvätern herkommen, die wir uns fest zu Jerusalem, zu David, zum Tempel und seinen Institutionen halten, die wir dieses Land bewohnen, die wir diese Geschichte mit Gott und seine Verheissungen haben, wir allen beanspruchen den Titel ‘Israel’ mit Recht, denn wir sind “das wahre Israel” [his emphasis].15

Sparks, in his attempt to summarize the intention of the Chronicler’s genealogical construction not only takes into account the broad geographical and ethnographic description emerging from the genealogies but also gives emphasis to the central role of the Levite genealogy when he (Sparks 2008: 367) concludes: “This study has made clear that the purpose of the genealogies, indeed the purpose of the book of Chronicles as a whole, is to encourage and support the work of the proper cultic officials, performing the proper cultic duties, in the proper cultic place.” With reference to the Levite genealogy in 1  Chronicles 6, Sparks (2008: 157) comes to a similar conclusion: “Finally, the position of the cultic officials . . . and the Levitical lands . . . , surrounding the cultic place and functions, shows the centrality of the cult to the restored community. They are a sign of restoration, as well as the means of maintaining possession of the land.” However, when inquiring into the function of the Chronicler’s genealogical construction, is there perhaps something more that could be seen in this literature? No one would question Kartveit’s, Oeming’s, and Sparks’s conclusions. 1 Chronicles 1–9 certainly reflects an ideally constructed world, and it certainly fulfills the role of the self-identification of All-Israel in the postexilic period of reconstruction. It surely emphasizes the centrality of the Jerusalem cult within the self-understanding of the postexilic community. But, given the prominence of land or town indications in the genealogies, particularly in the part that focuses on the tribe of Levi, could one perhaps also see an economic reflection emerging from this genealogical introduction to Chronicles? It is at this point that we should put some methodological insights promoted by Roland Boer to the test. 15. English: The genealogies therefore served the purpose of Israel’s self-definition and selfprotection late in the Old Testament time, in a time when Israel had no state of its own and was under foreign rule. They attest the enormous self-confidence of the Jerusalem temple congregation, particularly of its cult personnel. This “party” claimed the title “Israel” exclusively for itself: We—and we alone—we who descend from these ancestors, we who remain loyal to Jerusalem, to David, to the temple and its institutions, we who live in this land, we who have this history with God and his promises, we alone lawfully claim the title “Israel,” because we are “the true Israel.”

Yehud Economy in the Genealogies of Chronicles

87

Agrarian Economy through City-Elites’ Eyes As already mentioned in the preliminary remarks in the introduction, there is a general consensus among scholars that the authorship of the genealogies and, in fact, of the book of Chronicles as a whole, belonged to the cultic elite in Jerusalem in the Persian province of Yehud in a time toward the end of the period of Persian imperial dominion. Levin (2003: 237) summarized the consensus view in the following: It is now generally agreed that the Chronicler, although writing about the monarchic era, lived in the province of Yehud, during the 4th century b.c.e. In the postexilic period, the monarchy was gone; the Judean province was ruled by Persian-appointed governors, some local and some not. The priesthood was becoming more and more powerful, assuming a role in political leadership.

Gerrie Snyman (2010: 805) indicates his frustration, however, that the consensus view does not take one step further to reflect on the power relations in that context. He therefore accuses Chronicles scholarship of the following: “It does not seem as if the Chronicles scholarship is concerned with the Chronicler’s relationship with Persian power, or with the influence that imperial power would have had on the composition of the book.” In his definition of the cultic elite group responsible for Chronicles, Snyman goes beyond the customary description in terms of education (being literate officials) or sociopolitical role (cultic officials’ filling the leadership vacuum). Taking his cues from post­ colonial theory, in which the issue of power features centrally, Snyman (2003: 41–42) defines an elite group as follows: [T]he word “elite” is understood in terms of a group of people that has the power to rule and exert power by means of a bureaucracy . . . within a socio-cultural context. . . . A ruling elite will be tolerated when the society considers its presence necessary. . . . An elite group is in constant need of legitimizing its privileged position. One of the techniques by which the group maintains its position of power is keeping the power to prescribe . . . the “public transcript.” The public transcript describes the open interaction between subordinates and those who keep them in submission.

Snyman’s views alert the reader of the Chronicler’s genealogies to the fact that these name and geographical lists were not merely descriptive but probably stood in the service of exerting power in a specific sociocultural context. By means of these genealogies—and their relationship to other themes in the rest of the book (which will be dealt with below)—the elites of Jerusalem were legitimizing their own position and tried to prescribe their relationship to subordinates (in the case of the genealogies, to the agrarian tribes).

88

Louis Jonker

However, one should also take into account that this elite group in Jerusalem existed within the Persian imperial context. Their role and attempts to exert power were not only played out on a local level but happened within the context of Persian imperial dominion. Lester Grabbe (2004: 191) indicates that [t]he ancient imperial power seems to have had two main interests: taxes and military services. Its “economic policy” was simply one of maximizing revenue from a short-term point of view. . . . Reasonable relations had to be maintained with the local administration which was often made up of the native elite. Certain institutions, such as cities and temples, were frequently honoured.

More specifically referring to the Persian Empire, Grabbe (2004: 195) indicates: As far as we can tell, the basic economic and social structures of the Persian empire were mainly a continuation of those extant under the Assyrian and Neo-Babylonian empires, though naturally these varied from area to area. There were also new developments over time. As with both previous and subsequent empires, the primary aim of the government in this sphere was to promote a stable status quo which allowed the collection of as much revenue as possible. In order to do this, the Persians seem generally to have tried to keep the various entities in submission by certain concessions to local custom (e.g., respecting local cults and deities). Nevertheless, they had no hesitation in gaining or maintaining territory by the use of military might.

The city-elites, who are normally assumed to be the authorship of the genealogies and royal narratives of Chronicles, therefore occupied a position between the Persian imperial power and the local community in the province of Yehud. There is general consensus that the economy of Yehud throughout the Persian period and, indeed, throughout its history in antiquity was agrarian. Although Grabbe does not subscribe to Hoglund’s thesis (1991) that there was a Persian policy of “ruralization,” Grabbe (2004: 204) concedes the following: “Despite there being no evidence for any Persian policy of ‘ruralization’, Hoglund is quite right that the population of Yehud lived mainly in small, unwalled villages and had a subsistence-level economy.”16 Most recent studies on 16.  See also Charles E. Carter (1999: 249), who indicates: Scholars have long maintained that the Persian period economy in Yehud was predominately village based. This assertion can now be based in part upon the large numbers of Very Small and Small settlements within the province and on the nature of the agricultural installations that both surveys and excavations have unearthed. These augment the seal impressions, coinage and Greek pottery al-

Yehud Economy in the Genealogies of Chronicles

89

the archaeology of Persian period Yehud, such as those of Oded Lipschits and others, confirm this point of view. Lipschits and Tal (2007: 47) indicate the following: When we compare the number of settlements in late Persian period Judah to the number in late Iron Age Judah, we note a sharp decrease not only in the total number of settlements, but also in the amount of administrative-oriented sites. In other words, late Persian period Judah as a political entity may be defined, according to data retrieved from both excavations and surveys, as a rural province with no more than half the number of settlements as the late Iron Age.17

Lipschits furthermore indicates that Jerusalem was one of a few functioning administrative centers during the second half of the Persian period. On account of the distribution of the so-called yhwd seal impressions, Lipschits (2005: 179) comes to the following conclusion: “Given the current state of the data, the usable geographical-historical information contained in the seal impressions is scanty and limited. It does reflect the administrative importance of Jerusalem and Ramat-Rahel.”18 At this point, Boer’s distinction (2007: 35) between the economies of the village commune and the temple-city complex comes to mind again. We have seen that he defines the village commune “as both a kinship and economic unit, [which] persists through a series of economic and socio-political crises.” The temple-city complex, on the other hand, acts like the axle of a wheel to which the different dependent village communes are connected like the end points of spokes. The temple-city complex, according to Boer, unlike the village commune underwent a constant process of change due to sociopolitical circumstances. The role of the city was unthinkable “without the structuring role of the temple and indeed the palace which is an add-on to the temple” (Boer 2007: 36). When we take this distinction between the village commune and the temple-city complex as our vantage point, an interesting perspective on the Chronicler’s genealogies emerges. A very interesting relationship is constructed in this literature between Jerusalem and the ruraltribal villages and areas. We have seen that Jerusalem does not feature ready discovered from excavations to provide an unprecedented opportunity to examine the socio-economic status of the province more carefully.

17.  See also John W. Betlyon (2005: 20), who says: “While some claim that Yehud was the political/economic successor to the semi-autonomous state of Judah from the seventh and sixth centuries b.c.e., in reality Yehud was a small dependency in a huge world empire.” 18.  See also Lipschits and Vanderhooft 2007: 75–94.

90

Louis Jonker

often in the genealogies. The first occurrence is in 1 Chr 3:4b–5 where it is explicitly indicated that King David ruled (mlk) from there after he had moved his capital from Hebron. Very early in the genealogies, already in the family list where the focus is on the tribe of Judah, the relationship between the idealized king of the past and Jerusalem is established. Jerusalem also features prominently in the last parts of the genealogies, where it is indicated that some Benjaminites lived (yšb) in Jerusalem (in 8:28, 32, but also repeated in 9:38). In the section where the inhabitants of postexilic Jerusalem are described (9:1b–34), it is interesting that the claim is made in 9:3 that some of the people of Judah, Benjamin, Ephraim, and Manasseh lived (yšb) in Jerusalem. The combination of these four tribal names that represent the power center of the former united kingdom under David and Solomon but also the former divided kingdoms of Judah and Israel makes the claim here that “the whole of Israel” was represented in the postexilic Jerusalem. This genealogical description of the postexilic inhabitants of Jerusalem also clearly indicates that the different groups of priestly and Levite cultic officials had their seat in the temple in Jerusalem (see 9:34). This fact relates closely to those references in the Levite genealogy where Jerusalem is also indicated as the place of the temple where the different priestly and Levite families carried out their functions. From these textual data, one may deduce that the broad geographical construction of the genealogies were intended to associate Jerusalem with the royal past and to establish it in the Chronicler’s present as the seat of the cult. Although there was no royal institution present in postexilic Jerusalem, and although the royal palace no longer played any prominent role in this period, it is clear that the overall geographical construction of the genealogies were intended to emphasize the royal connection of the cult in Jerusalem.19 The aspiration reflected in this literature fits well in the second half of the Persian period when, according to archaeological finds (particularly seal impressions), the administrative influence shifted southward from the Benjaminite region (with Mizpah and Gibeon) to Jerusalem again (and Ramat Raḥel; see Lipschits: 2005: 180). It was a time when it would have been important to show the Benjaminite connections to Jerusalem but also to reclaim the political and cultic leadership position for Jerusalem. In this respect, one could say that the genealogical construction betrays something of the aspiration to establish Jerusalem as a temple-city complex again. The city-elites (who were probably the 19.  See also 1 Chronicles 23–27, where a detailed description is found of how King David gave all the Levite cultic officials their positions. See also the discussion below.

Yehud Economy in the Genealogies of Chronicles

91

hands behind the Chronicler’s genealogies) used their position of power to dictate the public transcript in which the central role and influence of Jerusalem is advocated. Simultaneously, the relationship between the temple-city complex of Jerusalem and the rural areas forms another part of this public transcript dictated by the city-elites of Jerusalem. The beneficiaries of the tribal lands as described in the Levite genealogy in 1 Chr 6:39–66[54– 81] are the very same priestly and Levite cultic functionaries who inhabited Jerusalem and served in the temple in the Persian period (according to 9:1b–34). And the donors of towns and land in the Levite genealogy are the very same tribes whose areas are described in the rest of the Chronicler’s genealogical introduction. The use of the verb ntn in the Levite genealogy is therefore conspicuous. Although the verb itself abounds in the Hebrew Bible, mostly without economic connotations, the distribution of verbs in the Chronicler’s genealogies highlights the use of ntn. It occurs only in the Levite section. I have shown above that the geographical list given here was taken from Joshua 21, which could be an adequate explanation for the use of the verb. However, in Joshua 21 the list forms part of a narrative that tells the story of the conquest of the Promised Land, a narrative context in which the giving of the land holds deep theological meaning, because Yahweh is the subject of the verb. The list was taken out of its narrative context when it was employed by the Chronicler, and the priestly families are given more prominence in the latter’s arrangement of the materials (as illustrated above), but the theological nuance of Joshua 21 is transferred to 1 Chronicles 6 by means of the verb ntn (with the donor tribes now becoming the subjects of the verb, instead of Yahweh). In both cases, Joshua 21 and 1  Chronicles 6, the use of the term ntn may therefore be interpreted as a reflection of a regime of allocation and extraction that operated in the tension between village commune and temple-city complex but was motivated theologically. From the rest of the genealogical construction, the connection of the priestly and Levite groups with the temple in Jerusalem is also prominent, as mentioned above. The overall geographical construction of the genealogies reflects the economic aspiration of the temple-city complex, Jerusalem, to extract provisions from the rural areas indicated as belonging to the other tribes. As Boer has indicated, the regime of extraction is often disguised in the form of allocation and often calls upon kinship-structures as motivation. By invoking the conquest narrative at this point of the genealogies, the Chronicler reminded his audience that the tribal and rural areas actually “were given” to the Levites by Yahweh, the deity of Israel.

92

Louis Jonker

This certainly reflects the language of allocation, particularly of the so-called sacred economy or theo-economics that Boer is describing. Although it is nowhere indicated in the Levite genealogy that Yahweh actually gave the land to the priestly lineage through the different tribes (unlike Joshua 21, where the narrative context explicitly suggests this),20 an explicit linking of the sanctuary in Jerusalem with Yahweh occurs twice in the Levite genealogy (6:16–17[31–32]) and again in the description of the postexilic inhabitants of Jerusalem (9:19, 23). The allocation of land to the Levite families by the donor tribes takes place within the sacred economy of the temple-city complex, extracting provisions from these same lands in order to establish and maintain the power center in Jerusalem.21 This extraction takes place on account of kinship structures, which legitimizes the postexilic claims.22 20.  Oeming (1990: 147) is of the opinion, however, that there are indications (such as the use of the wayyiqtol-form of the verb ntn in vv.  40, 49, 50, 52[55, 64, 65, 67] without a specified subject) that the Chronicler presupposed knowledge of Joshua 21 by his readers: “Man sieht hier besonders klar, dass Chr nicht nur die Liste, sondern auch ihren situativen Kontext im Jos-Buch als bekannt voraussetzt, was sich auch daran beweisen lässt, dass V. 49 Zitat aus Jos 21,8, also aus der Rahmenerzählung ist” (“It is clear that the Chronicler presupposed not only knowledge of the list but also the situational context in the book of Joshua. This can be seen from the fact that v. 49 is a quotation from Josh 21:8, that is, from the frame of the narrative”). 21.  Horsley (1991: 163) remarks that the role of the temple in Jerusalem should be defined in not only religious terms but also political and economic: “The temple in Jerusalem was a political-economic institution. Along with the high priesthood based there, the temple was the particular form taken by what we moderns would call ‘the state,’ at least during Hasmonaean times. Prior to and following the Hasmonaeans, the temple and its priesthood were part of the Persian, then Hellenistic, then Roman imperial political-economic apparatus of domination.” 22.  If one takes into account that the Chronicler wrote in a time when there were probably other operational Jewish temples (such as on Mount Gerizim and at Elephantine), this literature might have been part of an attempt to secure the Judean religious hegemony in that time. In a recent study, West (2011: 530) attempted “to provide evidence of the durability of the tributary mode of production model (and its association with and location within the temple-state). Durability both in terms of the viability of Marxist notions of modes of production for our exegetical and interpretive work in our post-Marxist world, and in terms of the capacity of the model to increase our economic of [sic] understanding across a long period of biblical history, bridging the testaments.” With reference to the Persian period, West remarks the following (quoting from Sam Mathew’s work Temple-Criticism in Mark’s Gospel): Zerubbabel’s temple, completed in 515 b.c.e., performed a similar economic function to the earlier Jerusalem temple. The difference here was that, “[s]ince Zerubbabel’s Temple stood alone with no adjacent royal compound, it gave legitimacy to a new form of community that was not linked to monarchic rule,” but continued with the same old forms of economic extraction, amassing “immense wealth in

Yehud Economy in the Genealogies of Chronicles

93

The rhetorical thrust of the genealogies in terms of a sacred economy can also be detected in other parts of the Chronicler’s historiography. Without going into detailed studies here, the following examples can be mentioned: First, in 1  Chronicles 29 there is a section (which stems from the Chronicler’s own hand) that indicates that King David, after he had made abundant provisions for the building of the temple in Jerusalem, also urged the “whole assembly” to contribute toward this cause. Verses  6–9 then narrate that “the leaders of the ancestral houses” responded positively by following David’s example. This narrative forms part of the Chronicler’s version of David’s preparing the kingship for his son. It already begins in chap. 22 but is then interrupted in chaps. 23–27 before it continues to chap. 28 and the section in chap. 29 that we are referring to here. It should not escape our attention that the great insertion in chaps. 23–27 deals with the organization of David’s cultic officials, with the generic indication that they were all Levites but served in different capacities. The preparation of the temple and the request for donations from the leaders of the ancestral houses are again conspicuously connected to a section where the authority of the Levites in Jerusalem is confirmed. In light of my economic reading of the genealogies, this combination of material can also be interpreted as a reflection of a sacred economy that is envisaged here, with Jerusalem and the temple complex extracting contributions from the ancestral houses. A second example occurs in the various narratives where the restoration of the temple is a theme. The passage beginning with 2 Chr 24:4 indicates that King Jehoash wanted to restore the temple, and for that purpose he sent out some Levites to all the cities of Judah to gather money from All-Israel for this purpose. When the Levites did not respond quickly enough, the king implemented an alternative system of voluntary gifts, which were put in a money chest outside the temple. The people are said to have contributed joyfully and abundantly to this cause so that there was even money left after the restoration (the lastmentioned piece of information stems from the Chronicler’s own addition to the narrative). The theme of temple restoration is also prominent in Hezekiah’s and Josiah’s histories.23 In 2  Chronicles 31, the narrator its building and treasury.” Once again, however, that there were other temples outside the district of Judea during the post-exilic period, at Elephantine, at Leontopolis, and at Gerizim indicates that Judean religious hegemony was contested, even if it is not clear to what extent these sanctuaries represented a different form of political economy.

23. See Lipschits (2006: 239–54), in which an interesting comparison is made between the Jehoash and Josiah temple-restoration narratives.

94

Louis Jonker

says that King Hezekiah reinstituted the offerings in the temple and contributed abundantly toward this cause. He also called on the people of Jerusalem to give their tithes to the temple personnel, the priests and Levites, and soon the custom spread throughout the cities of Judah and Israel of people dedicating their tithes to sustain the temple in Jerusalem. The temple restoration in Josiah’s days is done with money that has been collected by the Levites “from Manasseh and Ephraim and from the entire remnant of Israel and from all Judah and Benjamin and from the inhabitants of Jerusalem” (2 Chr 34:9 nrsv). All these examples show that a close association is reflected in key texts in Chronicles between the temple economy in Jerusalem and the donating rural areas, with the Levites often facilitating this extraction of resources. Again, through the economic lens of the genealogies, these narratives reflect something of the sacred economy of Yehud. These two examples confirm that the Chronicler’s genealogical construction already provided the framework within which the later narratives should be understood. In terms of an economic perspective, the genealogies already shape the outlines of the sacred economy which is propagated by the narratives that follow. The influential position of the cultic center in Jerusalem and the relationship of the temple-city complex to the village communes—themes that are also discussed in some of the following narratives—are already set out in the genealogical introduction. However, the situation might be much more complex than what is suggested above. Jerusalem not only functioned as a city-complex in the sacred economy of Yehud (and All-Israel), it was simultaneously the reestablished capital of a rural province of the Persian Empire. Betlyon (2005: 21) reminds us of the following: [S]ome scholars argue that Jerusalem was the capital of an autonomous or semi-autonomous province in the Persian period, but such arguments, even if they refer to the latter part of the period, fail to deal with the tremendous power of Persia. Any attempt to define what an autonomous province may have been must reckon with overwhelming military force and centralized controls administered through various satrapal bureaucracies.24

24.  See also Fried (2004), who suggests that a bureaucratic model of imperial control (as proposed by Eisenstadt) should be assumed for the relationship between the imperial center and Yehud. Although many scholars would agree with this model of Persian administrators throughout the Empire exerting considerable control in the provinces, there are divergent views on the issue of internal autonomy in the provinces.

Yehud Economy in the Genealogies of Chronicles

95

Boer’s comment (2007: 38) about the “double position” that Jerusalem occupied comes to mind here. He indicates (as discussed above): What the studies thus far do not consider sufficiently are the various mediations of the relationship between the despotic empire and village commune. This is especially true of small vassal states such as Israel, and indeed cities like Jerusalem, that stand between empire and village commune. . . . Not only is the vassal state a crucial mediator between empire and village commune, but it may also act as a buffer [my emphasis].

Although Boer is referring here to earlier periods in Judah’s and Israel’s existence under the Assyrian or Neo-Babylonian yoke, a similar situation may be assumed for the Persian period. Due to their concessions to local customs and religions (e.g., by respecting the cults and deities of local populations), the Persians in general have a fairly benevolent reputation. However, this does not mean that they did not exercise their imperial power. Grabbe (2004: 196–97) describes the situation as follows: All the efforts of the Persian government appear to have gone into realizing revenue from a short-term perspective. . . . [M]ost of the imperial income came from taxation which was gathered and then dispersed to pay state expenses. . . . The Persepolis Fortification and Treasury Tablets give examples of how the tribute was used on state business. . . . The satraps no doubt passed down a good deal of the responsibility for collecting the tribute to the provincial governors, but there is little indication that they tried to micro-manage the local economy.

One may therefore assume that the local governors in Jerusalem were also tasked with putting in operation the extractive regime of the Persian imperial center. In this respect, Jerusalem with its provincialinfluence sphere functioned as a village commune vis-à-vis the imperial power center25 but also was tasked with the regime of extraction by the imperial power. In this respect, Jerusalem had a double role and occupied an in-between position. Various studies have suggested that the temple in Jerusalem was actually a part of the taxation system of the Persian Empire (see, e.g., Schaper 1995: 528–39). The temple had to act as tax collector and processor on behalf of the Persian imperial government. Unfortunately, since we do not have much evidence of the Persian taxation and tribute system, it remains uncertain how the tax and tribute burden was administered in rural areas such as Yehud. 25.  Grabbe (2004: 199) writes: “Judah was mainly made up of people living in unwalled farming villages, with Jerusalem the only urban area of any significant size. With about 3000 people at its largest Jerusalem had approximately 10 per cent of the province’s inhabitants living in it.”

96

Louis Jonker

The “in-between” position of the temple in Jerusalem (between the village communes of Yehud and the temple-city complex of the imperial regime) probably contributed to the Chronicler’s overall construction. Whereas the genealogical introduction to the book is witness to the regimes of extraction and allocation that upheld the sacred economy in Yehud, the conclusion to the book also corroborates this sacred economy. 2 Chr 36:22–23 indicates that Cyrus, the Persian emperor, actually acted on account of a Yahwistic prophecy and that he responded to the command of Yahweh when he ordered the return from Babylonia and the rebuilding of the temple in Jerusalem. This closing section of the book therefore motivates the position of the temple and the city of Jerusalem in the imperial context, but with reference to the regime of sacred allocation again. All this leads us to this question: to whom was the book of Chronicles addressed? If the genealogies (and the rest of the book of Chronicles) also wanted to serve an economic agenda within late Persian period Yehud, which audience did the writer have in mind? The majority view in scholarship is that the Chronicler simply wrote for the community in Jerusalem.26 That means the genealogies and the rest of the book served the purpose of self-legitimization. It therefore reflects the selfunderstanding of the cultic community and served the purpose of encouraging this community to be loyal to the Persian imperial economics on the one hand, but on the other hand also to serve its administrative function of extracting tribute and tax from the village communes. Levin (2004) has argued differently in his quest for the Chronicler’s audience, however. Although Levin concedes that the writer of Chronicles must have been part of the cultic elite in Jerusalem, he sees a shift in attitude among this elite from the early postexilic to the late Persian era (2004: 244): “So while the Chronicler was obviously a member of the Jerusalem elite, that elite in the late 4th century seems to have been very different from the ‘Golah-returnee’ elite of the mid-5th century, as represented by the author of Ezra–Nehemiah.” Levin argues that the Chronicler, despite being part of the cultic elite in Jerusalem, made use of oral tribal-genealogical traditions for the construction of the introduction. He maintains (2004: 245): This audience, living in a society that was still to a large extent “tribal,” could easily understand the Chronicler’s message of the basic unity of all Israel in all of its land, in the past and in the present. The Chronicler, as opposed to the separatist, maybe anti-Samaritan, Priestly author of Ezra–Nehemiah, is not telling his “history” from the perspective of the 26.  See discussions in all the major commentaries.

Yehud Economy in the Genealogies of Chronicles

97

urban elite of Jerusalem. When the Chronicler, in his genealogical “introduction,” lays out the ethnic and geographical framework of his “Israel,” his perspective is that of the tribal, village society, which was very much alive and functioning in his day. The villagers of the hill country of Judah and Benjamin, but also those of Ephraim and Manasseh, were both the Chronicler’s source of information and his audience.

Since one may assume that the levels of literacy in the rural areas of the tribal lands were very low, it is highly unlikely that these people would in fact have read the Chronicler’s materials or that the expectation would have been for them to read it with their own eyes. However, Levin may be right that the rural population in the tribal areas under Jerusalem’s jurisdiction might have been the intended audience of this literature. The reason that the Chronicler gave such a prominent place to his introductory genealogies, which use a typical literary form of tribal societies, may be that he wanted to strengthen the regime of extraction from these areas by disguising it in an allocative format and by casting it in the form of kinship relationships. He wanted to motivate the villagers by explaining to them why they needed to remain loyal to the cultic elite in Jerusalem by providing them with goods and how, in doing so, they would take part in the sacred economy of the land.

Conclusion I have set out to contribute to the multidisciplinary discussion on agrarian economies in depopulated areas of the Persian period by focusing on the Chronicler’s genealogies, which most likely originated near the end of this period. My intention was to investigate whether these genealogies, which provide numerous references to land possession or allocation, could shed any light on our reflection on the agrarian economy of its time of origin. I focused on the Levite genealogy in 1 Chr 5:27–6:66[6:1–81], in which numerous indications of land allocation occur, as well as on how this genealogy relates to the other parts of the Chronicler’s genealogical introduction. Boer’s models of economic description, which he derived from Marxist theory, provided the methodological perspective for this investigation. I remain well aware throughout this study of the fact that our methodological models at least co-determine the outcome of our investigations. Former studies that have been done on the Chronicler’s genealogies were therefore not neglected or ignored, but my contribution hopefully showed that Boer’s notion of “sacred economy” could potentially enrich our interpretation of the Chronicler’s genealogies. I have shown above that the Chronicler’s genealogies very clearly configure the relationship between Jerusalem and rural towns in the

98

Louis Jonker

tribal areas in terms of donors and beneficiaries. The beneficiaries are the Levite families who receive their land from all of the tribal areas. It is stated in the Levite genealogy that the various towns and their surrounding land “were given” (with the verb ntn) to the Levite families. It further becomes clear that Jerusalem is closely associated with the rule of David—but also with the temple, where the priests and Levites serve. Using Roland Boer’s distinction between village communes and temple-city complexes, as well as his understanding of the sacred economy in terms of regimes of allocation and extraction, I came to the conclusion that the genealogies, corroborated by various narratives in the Chronicler’s royal accounts, construct an economic vision that relies on the theological motivation that Yahweh allocated the land to the cityelites, and thus tension is created between Jerusalem as the temple-city complex vis-à-vis the village communes of the tribal areas. However, I also indicated that this economic vision was formulated under Persian imperial rule and that particularly the conclusion to Chronicles sets the genealogies in another framework. According to that framework, which also reflects a specific understanding of the sacred economy, Jerusalem and its dependent areas become the village communes that stand in tension with the Persian imperial center of power. The regimes of allocation and extraction functioned in terms of taxes and tribute, and the temple in Jerusalem most likely had the function of administering the imperial economy. The Chronicler’s genealogies within the greater composition of the book of Chronicles, therefore, reflect a double position in the economy for Jerusalem. Jerusalem functioned as a temple-city complex in contrast to the village communes of the tribal lands but was simultaneously under the imperial yoke, administering the sacred economy on behalf of the Persian emperor in the rural environment of Yehud. Boer cautions us not to confuse the temple-city complex with the state. He argues that the state instead emerges within the tension between the economies of the village communes, on the one hand, and the templecity complex, on the other hand. Although the concept of a state is too modern to use in connection with Jerusalem and Yehud, one acquires a new perspective on the emergence of Yehud as a Persian province when applying this perspective. With the return of administrative and cultic functions to Jerusalem from the Benjaminite towns in the middle of the Persian era, Jerusalem—particularly the cultic city-elites—began dictating the public transcript that embodied the aspirations of this emerging power. The book of Chronicles, particularly the genealogical introduction, formed part of this public transcript in which Jerusalem was configured in terms of the surrounding rural tribal areas, but also

Yehud Economy in the Genealogies of Chronicles

99

in terms of the Persian imperial center. Jerusalem occupied a double economic position, acting as a buffer between the village communes of the rural areas and the allocative regime of the imperial center.

Bibliography Assis, E. 2006 From Adam to Esau and Israel: An Anti-Edomite Ideology in 1 Chronicles 1. VT 56: 287–302. Bang, P. F.; Ikeguchi, M.; and Ziche, H. G., eds. 2006 Ancient Economies, Modern Methodologies: Archaeology, Comparative History, Models and Institutions. Bari: Edipuglia. Ben Zvi, E. 1992 The List of the Levitical Cities. JSOT 54: 77–106. Betlyon, J. W. 2005 A People Transformed: Palestine in the Persian Period. NEA 68: 4–58. Boer, R. 2007 The Sacred Economy of Ancient Israel. S JOT 21: 30. Carter, C. E. 1999 The Emergence of Yehud in the Persian Period: A Social and Demographic Study. JSOTSup 294. Sheffield: Sheffield Academic Press. Chaney, M. L. 2006 Micah—Models Matter: Political Economy and Micah 6:9–15. Pp. 145–60 in Ancient Israel: The Old Testament in Its Social Context, ed. P. F. Esler. Minneapolis: Fortress. Dirksen, P. B. 2005 1 Chronicles. HCOT. Leuven: Peeters. Fried, L. S. 2004 The Priest and the Great King: Temple-Palace Relations in the Persian Empire. Biblical and Judaic Studies 10. Winona Lake, IN: Eisenbrauns. Gottwald, N. K. 1993 Social Class as an Analytic and Hermeneutical Category in Biblical Studies. JBL 112: 3–22. Grabbe, L. L. 2004 A History of the Jews and Judaism in the Second Temple Period. London: T. & T. Clark. Hoglund, K. G. 1991 The Achaemenid Context. Pp. 54–72 in Second Temple Studies 1: The Persian Period, ed. P. R. Davies. Sheffield: JSOT Press. Horsley, R. A. 1991 Empire, Temple and Community—but No Bourgeoisie! A Response to Blenkinsopp and Peterson. Pp. 163–74 in Second Temple Studies 1: The Persian Period, ed. P. R. Davies. Sheffield: JSOT Press. 2009 Covenant Economics: A Biblical Vision of Justice for All. Louisville: Westminster John Knox.

100

Louis Jonker

Japhet, S. 1993 I and II Chronicles: A Commentary. OTL. Louisville: Westminster John Knox. Jobling, D. 2007 Very Limited Ideological Options: Marxism and Biblical Studies in Postcolonial Scenes. Pp. 184–210 in Postcolonial Biblical Criticism: Interdisciplinary Intersections, ed. F. F. Segovia and S. D. Moore. New York: Continuum. Kartveit, M. 1989 Motive und Schichten der Landtheologie in I Chronik 1–9. ConBOT 28. Stockholm: Almquist & Wiksell. Levin, Y. 2004 From Lists to History: Chronological Aspects of the Chronicler’s Genealogies. JBL 123: 601–36. Lipschits, O. 2005 The Fall and Rise of Jerusalem: Judah under Babylonian Rule. Winona Lake, IN: Eisenbrauns. 2006 On Cash-Boxes and Finding or Not Finding Books: Jehoash’s and Josiah’s Decisions to Repair the Temple. Pp. 239–54 in Essays on Ancient Israel in Its Near Eastern Context: A Tribute to Nadav Naʾaman, ed. Y. Amit, E. Ben Zvi, I. Finkelstein, and O. Lipschits. Winona Lake, IN: Eisenbrauns. Lipschits, O., and Tal, O. 2007 The Settlement Archaeology of the Province of Judah: A Case Study. Pp. 33–52 in Judah and the Judeans in the Fourth Century b.c.e., ed. G. N. Knoppers, O. Lipschits, and R. Albertz. Winona Lake, IN: Eisenbrauns. Lipschits, O., and Vanderhooft, D. 2007 Yehud Stamp Impressions in the Fourth Century b.c.e.: A Time of Administrative Consolidation? Pp. 75–94 in Judah and the Judeans in the Fourth Century b.c.e., ed. G. N. Knoppers, O. Lipschits, and R. Albertz. Winona Lake, IN: Eisenbrauns. Mathew, S. 1999 Temple-Criticism in Mark’s Gospel: The Economic Role of the Jerusalem Temple during the First Century c.e. Delhi: ISPCK. Oeming, M. 1990 Das Wahre Israel: Die “Genealogische Vorhalle” 1 Chronik 1–9. Stuttgart: Kohlhammer. Schaper, J. 1995 The Jerusalem Temple as an Instrument of the Achaemenid Administration. VT 45: 528–39. Snyman, G. F. 2003 A Possible World of Text Production for the Genealogy in 1 Chron­ icles 2.3–4.23. Pp. 32–60 in The Chronicler as Theologian: Essays in Honor of Ralph W. Klein, ed. M. P. Graham, S. L. McKenzie, and G. N. Knoppers. London: T. & T. Clark.

Yehud Economy in the Genealogies of Chronicles 2010

101

The Ethics of Reading and the Quest for the Audience in the Book of Chronicles. OTE 23: 804–21. Sparks, J. T. 2008 The Chronicler’s Genealogies: Towards an Understanding of 1 Chronicles 1–9. Atlanta: Society of Biblical Literature. West, G. O. 2011 Tracking an Ancient Near Eastern Economic System: The Tributary Mode of Production and the Temple-State. OTE 24: 511–32. Willi, T. 2009 Chronik (1 Chr 1,1–10, 14). BKAT 24/1. Neukirchen-Vluyn: Neukirchener Verlag.

Ancient Comparisons, Modern Models, and Ezra–Nehemiah: Triangulating the Sources for Insights on the Economy of Persian Period Yehud Peter Altmann Universität Zürich

Introduction A well-traveled volume on the topic of the ancient Greek economy opens with the following statement: “The ancient economy is an academic battleground” (Hopkins 1983: ix–xxv, ix). Although this statement arose amidst the heated debates of the Finley era, the same is true for current scholarship of the ancient southern Levant during the Persian period: Asian or tributary mode of production? Substantivism? Patronage? Formalism? In order to address this complex of problems, my discussion will begin by (1) considering the goal of addressing the economy in the Persian Levant. Then, (2) I will move to a short engagement with the classic Polanyian and formalist economic models. These preliminary investigations provide the basis for (3) an attempt to lay out some developments forward, through the work of Douglass North and “New Institutional Economics.” (4) An additional step will be to cast a sidelong glance at a populated region of the Persian Empire—Babylonia. My final destination (5) will be the application of these comparisons and models in a case study of Neh 13:15–22.1

Defining the Goal The term economies is quite elusive. What is an economy and how does one approach the history of an economy? I take the definition of “economic history” from Douglass North (1981: 3) as a starting point: [T]he task of economic history [is] to explain the structure and performance of economies through time. By “performance” I have in mind the typical concerns of economists—for example, how much is produced, 1.  One limitation of my study is my generally undifferentiated use of the “Persian period” as a single entity of about 200 years. The more-specific period I have in view is roughly from the mid-5th to the mid-4th century, when there was surely something of an independent Yehud subprovince. However, for the purposes of this essay, I am generally sidestepping implications of Persia’s loss of Egypt.

103

104

Peter Altmann

the distribution of costs and benefits, or the stability of production. The primary emphasis in explaining performance is on total output, output per capita, and the distribution of income of the society. By “structure” I mean those characteristics of a society which we believe to be the basic determinants of performance. Here I include the political and economic institutions, technology, demography, and ideology of a society.

I am, then, interested in two aspects: structure and performance. Performance requires identification of what is produced, who produces it, how “costs” of production are measured, and who benefits from the production. This might include speculation on per capita output and on distribution of income. And, underlying performance are the questions of structure, such as the approach of the society to property rights, and for North, especially the cultural and political institutions. Yet many of the factors listed in North’s definition are out of reach for the Persian-period southern Levant, or, the focus of this essay, Yehud. It is possible, however, to establish a general baseline of what was produced in Persian period Yehud. Most of what was produced consisted of the items necessary for families’ basic needs: grain, traditional pottery, etc. Perhaps the more interesting side of this question relates to what was produced over and above what was required for families to sustain themselves—the surplus. What can be said about this? First, archaeological investigations of the material culture provide evidence for large grain storage facilities—mostly outside Yehud and closer to the Egyptian border, however, pointing to military attention to Egypt (Edelman 2011: 133–44, 140). Second, Neh 13:15–16 lists the trading of various goods in Jerusalem. The exports consist of grain, wine, grapes, and figs; and the imports of fish, and the nebulous designation “merchandise” (meker). Third, the biblical texts such as Ezra 1–8 report “cultic” production in the form of temple religious services. Conceived in economic terms, the cultic services offered in Jerusalem brought in outside investment— silver, gold, temple vessels, clothing, and state property in the form of the tax receipts of the satrapy that the Persian kings designate for the rebuilding and maintenance of the Jerusalem temple. Neh 10:33–40[32– 39] also names the products that a community of “Israelites” (Neh 8:1, 9:1–2) commit themselves to give to the temple: 1⁄3 shekel to buy bread, flour, and animals for regular offerings, as well as wood for fuel. These are augmented by yearly contributions of first fruits of all produce, firstlings, and tithes. Fourth, Nehemiah 5 indicates royal tax payments of silver, implying the trade of agricultural products for silver, as well as payments in

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

105

kind to the peḥâ.2 With these—primarily textual—data in hand, what theories are available to establish a working theoretical structure?3

Classic Approaches The most ready-at-hand framework from current Western experience, that of the “formalist” position is most prominently represented in scholarship by Morris Silver. He brings a modern economics background to his studies, arguing for the use of modern Western (and capitalist) economic categories in the analysis of the ancient societies. For example, Silver proposes, “[E]ven rudimentary knowledge of modern economic analysis is capable of enhancing our understanding of ancient societies” (Silver 2004: 65–87, 65). This approach understands ancient actors as making decisions according to principles of economic efficiency or rational choice. Furthermore, Morris interprets (Mesopotamian) texts with price data as showing that prices and actual interest rates fluctuated in accordance with modern market mechanisms.4 Philippe Guillaume’s monograph Land, Credit and Crisis (2012: 121–22) also incorporates formalist tendencies, stating: “ ‘permanent over-indebtedness’ seen as characteristic of the situation of biblical farmers . . . has in fact always been the norm, even more so today, and it constitutes the mark of a healthy economy rather than the sign of a structural crisis.”5 In Guillaume’s analysis, debt is an important sign of a well-functioning economy, not of a crisis. I find it important to recognize the underlying tone of these positions—a tone also on display in Jursa’s analysis of Babylonia in the 6th and 5th centuries b.c.e.: the focus is more on how the machine works than on the exploitation of certain groups. Or, similarly, the focus is on the maximizing behavior of individuals within an established state structure rather than on the predatory nature of the state (North 1981: 22). Nonetheless, while there is room for a good number of insights from classical economic theory, there are significant obstacles to its application as a dominant model in the form of formalism to biblical studies. One problem is the difference in mentalities argued by Polanyi in terms of the lack of a substantial economic profit motive (Polanyi 2001; before him, Mauss 1984: 173). This critique, while perhaps overblown, is also 2.  If one follows the translation in the jps, then 10:38[37], “in all our towns subject to royal service,” provides a second reference to the royal extraction. 3.  Van de Mieerop (2004: 54–64) provides orientation to scholarly discussions of ancient Near Eastern economies, which he divides into three general camps—Marxist, substantivist (or the related primitivist), and modernist (or the related formalist). 4.  Note the summary of underlying principles in Nam 2012: 33–34. 5.  Guillaume refers to Kessler 2006: 91–121, esp. p. 120.

106

Peter Altmann

aimed at the tautological nature of the formalist contention of “rational maximization.” To say that ancient actors made decisions based on what they thought was best or most profitable for them sidelines many important questions and subjects of what was important to them and why—that is, culture, religion, and politics.6 A second significant hurdle is the lack of data. How could one calculate the per capita GDP for the Persian period Levant?7 Finally, as argued by North, the transaction costs were so high in the ancient world that the things modern analysis can explain bracket out many interesting and important variables in the ancient economy. The substantivist position, generally identified with Karl Polanyi, developed around the middle of the 20th century as a critique of the use of modern categories and modern economic positivism (Narotzky 1997: 2–3). Specifically, substantivist analysis posits the lack of widespread depersonalized trade, especially on a local level. The fundamental goal of this scholarly approach was to separate noncapitalist economies from the modern capitalist economy, thereby suggesting the uniqueness of the developments associated with the Industrial Revolution. Polanyi and his followers instead view ancient economics as embedded within the larger matrix of social relations (Fusfeld 1957: 342–56, 343). Polanyi (2001: 46) himself sums up as follows: The outstanding discovery of recent historical and anthropological research is that man’s economy, as a rule, is submerged in his social relationships. He does not act so as to safeguard his individual interest in the possession of material goods; he acts so as to safeguard his social standing, his social claims, his social assets. He values material goods only in so far as they serve this end.

Polanyi and other substantivists provide a significant service to modern understandings of ancient economic systems in reminding modern readers that a certain paradigm shift has taken place in modern Western and capitalist-oriented societies: economics was typically seen as 6.  It also fails to explain so-called free-riders—that is, why the majority so often choose not to act in their own self-interest but instead tow the party line. North (1981: 53) comments: [T]he costs of maintenance of an existing order are inversely related to the perceived legitimacy of the existing system. To the extent that the participants believe the system fair, the costs of enforcing the rules and property rights are enormously reduced by the simple fact that the individuals will not disobey the rules or violate property rights even when a private cost/benefit calculus would make such action worthwhile.

7.  Bedford posits some calculations for Persian period Mesopotamia but not for Syria–Palestine (Bedford 2007: 302–29, 327).

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

107

part of the larger social and political systems, rather than functioning as a prime mover underpinning these systems, operating at the level of the individual. Value was not only—or primarily—determined by economic standards. A further service from the substantivists comes in highlighting the categories of reciprocal, redistributional, and market-exchange structures for the economic movement of goods. While Polanyi went too far in denying market exchange in the ancient Near East, the suggestion of these different basic societal structures for economies provides insight: it is helpful to view the palace-temple centered economies of second millennium and early first millennium Mesopotamia and Egypt in terms of primarily redistributional structures, without denying the presence of significant reciprocal exchange and some market exchange. Perhaps the Phoenician city-states are one exception to this schema in their consistent push into markets on the limits of societies. However, patron-client type exchange, the importance of which Walter Houston (2008: 44–50) has expounded for Persian period Yehud, is better subsumed under Polanyi’s “reciprocal” category then in a formalist framework.

New Institutional Economics The “New Institutional Economics” or NIE of Douglass North, developed over the past several decades, capitalizes on several strengths of both the schools of thought mentioned above. I am not the first to suggest the helpfulness of his position for the ancient Mediterranean: several recent works in the study of economics in classical studies draw on his thought (Bang 2009: 194–206; Scheidel, Morris, and Saller 2007). As an economist, North exudes a certain trust in the philosophically “conservative” neo-classical economics. This trust is balanced, however, by his frustration with the inability of classical economics to explain key issues in economic history (North 2005: viii).8 He thus posits—note the similarity with the substantivists whom he also criticizes9—“Economic 8.  North’s primary interest is in the question “Why do economies change?” 9.  North (1981: 42) states: Polanyi made a market synonymous with a price-making market. It should be readily apparent, however, that any form of voluntary contractual exchange involves a market and that its form will be dictated by the considerations advanced above. Polanyi made a basic error in thinking that any deviation from the Agora-type market implied non-economizing behavior: even the era, which in The Great Transformation (1957), he regarded as the epitome of the market mentality was characterized by an enormous variety of contractual arrangements that were not price-making markets. Two considerations militated against the existence

108

Peter Altmann

change, therefore, is for the most part a deliberate process shaped by the perceptions of the actors about the consequences of their actions” (North 2005: viii; italics original). Ideas and culture play a profound role. Second, as the name suggests, NIE is interested in the institutions that set the perimeters for human economic action: “Institutions are a set of rules, compliance procedures, and moral and ethical behavioral norms designed to constrain the behavior of individuals in the interests of maximizing the wealth or utility of principals” (North 1981: 201–2). This premise addresses some of the problems for neo-classical economics, such as the “free-rider problem”: why people often choose to follow customs and laws when it is not in their personal best interest (North 1981: 10–11) and factors that keep economies from growing, such as states working against economic growth. His key insight here is bound up in the notion of “transaction costs.” These costs go far beyond transportation, also including a complex notion of the state, which can be seen as both an asset in providing stability and a liability in its extraction of rents. NIE as a model brings formalist insights into a somewhat “substantivist” or, perhaps better, “Weberian” framework where all economic action is embedded (Bang 2009: 196). NIE offers categories for the broader institutional—both political and cultural—perimeters that also include space for considering active individual, family, or group responses within these structures that work toward economic prosperity. This structure appears promising even for the limited economic data at hand for the Persian period southern Levant. The promise lies in its dual focus. The economic roles of imperial and regional government need not cancel out individual active responses.

Comparison with Babylonia I would like to cast a short glance at Babylonia of the 6th and early 5th centuries for its positive and negative comparisons with Yehud during the Persian period. Both the considerable literary and thematic borrowing10 and the shear number of economic texts suggest Babylonia as a fruitful comparison. Jursa depicts the Babylonian economy from Nebuchadnezzar to Xerxes as a highly developed—for antiquity— of price-making markets before the 6th century b.c.e. One was the transaction costs considerations that have been the subject of this chapter; the second was the wealth-maximizing objectives of the rulers of the state.

10.  The clearest examples include the interaction with the Enuma Elish in the Priestly Primeval History, or Mesopotamian city laments in Lamentations, DeuteroIsaiah, Jeremiah, and Ezekiel.

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

109

market economy.11 Both temples and private individuals were quite involved in exchange practices for meeting their daily needs, especially in the cities.12 The temples were emerging out of a structurally redistributive economy, relying increasingly on market exchange for many essential needs—a solid 75–85% of all external transactions were paid in silver by the temples (Jursa 2010: 569). In fact, clear economic maximization took place: for example, the temple Ebabbar in Sippar increased date plantations to the detriment of barley to maximize sales for silver (Jursa 2004: 115–36, esp. p.  119).13 Yet, unlike Jerusalem, the Babylonian temples played significant roles as landholders and slaveholders.14 How­ever, given the interaction with external vendors and labor, there was nothing like clear-cut Polanyian redistributive economies in the Babylonian temple systems by this time. Nonetheless, Jursa does detect divergent motivations in different sectors of the economy, and this is where the NIE becomes especially helpful. For the wealthy temple adherents, prestige played a central role in economic choices, while decisions by non-temple entrepreneurs appear more economically based (Jursa 2010: 294–95). The former, in 11.  Yehud does develop coinage eventually—in the 4th century—but there was at least the Athenian tetradrachm in the mid-5th century, found at Beth-Zur, though whether these were viewed as coins, per se, rather than bullion before the 4th century is unclear, given the cut marks found in early coins. There is little evidence for coinage in Babylonia during the Persian period, though qalû and ginnu silver seems to imply some kind of centrally certified silver purity, and Jursa argues that erbu ša aranni indicates a coin (Jursa 2010: 489). However, regardless of the frequency of use of the sigloi or drachma in Babylonia, there are ample indications of trade for both luxury and everyday low-value items; in fact, rents on temple lands in the 5th century were increasingly paid in silver (Jursa 2008: 621). 12.  One connection often made is in the centrality of the temples. Schaper (1995: 528–39) argues that the collection of taxes by Babylonian temples on behalf of the imperial treasuries was mimicked in the Jerusalem temple collection. I instead imagine that there were significant divergences. First, in thinking of the temple comparisons, the Esagila in Babylon and other Babylonian temples had long-standing relationships with the political rulers, and these temples did not undergo destruction in the same way that the Jerusalem temple did (Altmann 2014). 13.  Jursa notes that the amount was ca. 60 tons of dates for silver, according to CT 57 22, which documents the sale of just under 5% of the entire date harvest of the temple in a single transaction (2004: 199 n. 22). 14. Leviticus 27, if it implies land that was eventually dedicated to the temple in payment of temple loans, does not seem to have worked its way into reality so that its impact is seen in any other texts of the Persian period. Second, a more general difference between Babylonia and Yehud is the geographical and climactic conditions for agriculture. These led to a difference in the nature of agricultural production and products, especially during this period, when date gardens were becoming ever more prominent in Babylonia. This is also noted by Bedford 2007: 308.

110

Peter Altmann

addition to their prebends, were far more likely to be rentiers, while the latter gained their income from various kinds of trade. The media of transactions tended to be different in these spheres: even in the late 5th century, most agricultural rents were paid in kind, while traders purchased and sold their goods for silver—also, however, when involved in advanced purchases of agricultural goods in rural regions (Jursa 2010: 627–29). One can conclude from this that institutional structures set the boundaries on economic action. A final important development in the Babylonian economy is that the economy grew in real terms.15 There are even signs of growth per capita, if the rising payments to temple oblates and hired workers can be trusted to provide insight on this question.16 The implications of this could be extensive: if real incomes grew in Babylonia, then the Persian elites did not skim all the profits off the top: some of the in­creased ben­ efits of productive growth went to the workers.17 There is, then, considerable movement in Babylonia during this period toward market exchange. Jursa calls it commercialization, but it is—as one would expect—neither complete nor even. How much spillover effect was there in the small, out-of-the-way Yehud highlands? Given the origins of the benê-hagôlâ in greater Babylonia, the Yehud communities’ links with Babylonia suggest some experience with these developments.18 What evidence of this appears in the biblical texts? 19 15.  The relative peace allowed for considerable development of new irrigation systems, reclamation of land, urbanization, and a shift in agricultural regimes from arable farming to date gardens (contra the depiction by Weber quoted in North 1981: 102). 16.  With regard to grain, Jursa (2010: 811–12) notes: Overall, the majority of wage levels analysed by Scheidel falls within a core range from 3.5 to 6.5 litres per day. This is for instance true for Nuzi and Ur III data, but also for Ptolemaic Egypt and for early Islamic Iraq . . . Athens in the late fourth century bc: 13–15.6 litres, and in the late fifth century: 8.7 litres, Babylonia, around 12 litres in the mid-sixth century.

17.  This could be difficult for a static Marxist Asian production model to handle. 18.  One conclusion that can be drawn is that, by the late 6th or early 5th century, those immigrating to Yehud—thus, not necessarily the first travelers with Shesh­ bazzar, but likely even they—experienced such exchange. 19. Furthermore, I would follow Blenkinsopp’s favorable comparison of the ḫatrus of Babylonia with Yehud (2009: 118): “The internal affairs of self-governing collectivities formed by ethnic minorities in the Nippur region appear to have been regulated by an assembly (puḫru), always of course under the watchful eyes of the imperial authorities, to which the qĕhal haggôlâ would be roughly parallel.” I find this comparison especially helpful for the quasi resettlement of agricultural installations and establishments connected to the Jerusalem temple in some way through

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

111

Nehemiah 13:15–22: A Case Study The final section of this essay focuses on Neh 13:15–22, presenting insights from previous investigations of New Institutional Economics and contemporary Babylonia. The reason for choosing this text is that it provides a brief (if stylized) glimpse into market exchange in Yehud during the Persian era. The section is set off from the previous one through the general introduction “in those days” (bayyāmîm hāhēmmâ). Furthermore, the location changes in v. 15 from the temple in Jerusalem to the Judean winepresses. Similar to the previous section, however, here also the connection between the rural periphery and central place arises.20 While an earlier compositional layer may have focused more on Judean nobles pressing peasants into agricultural work on the Sabbath (Neh 13:15a, 17–18*, 22*),21 the MT Sabbath legislation focuses on religious contributions. Where this becomes difficult is the manner in which this relates to the collection of “secular” taxes. It seems to me that the consistent push in Ezra–Nehemiah for Yehud’s residents to identify themselves economically with Jerusalem and its temple, however, demonstrates the Jerusalem-affiliated Judeans’ need repeatedly to motivate farmers to ally themselves with Jerusalem instead of some alternative religious establishment, such as Gerizim. This is where some version of a model concerned with a separate religious, in addition to a separate political association within Yehud is quite plausible, and it is compatible with Kessler’s “charter group” idea of multiple and various internal and external motivations for “returning,” rebuilding, and resettling Jerusalem and its temple. He states: “Thus the imperial administration’s most foundational goal would have been the basic stability of the region. For the Babylonian Golah as a whole, I venture to suggest that the purpose of the re-establishment of a temple-centered community in Yehud was in large measure to provide a tangible anchor for the exilic communities in the east” (Kessler 2006: 104). For example, the boundaries between the interests of the Jerusalem Yahweh temple were neither shared nor protected by the political structure in a way that alleviated its financial pressure (according to Neh 13:10–13). 20.  Some argue that ṣāyid is an addition. It is rendered by the LXX as πράσεως (activity) and otherwise only represents hunted game or provisions for a journey. Schunck (2009: 381) also notices philological difficulty; he goes on to propose a complex solution. He proposes that meker bāʿîr was erroneously changed to mikrām ṣāyid. 21. In terms of the compositional history of the section, several issues arise: (1) the divergent geographical locations, (2) the discrepancy between working on the Sabbath and selling/buying on the Sabbath, and (3) the omission of “the Tyrians” from the LXX. First, in line with the move in v. 15 outside Jerusalem to the vineyards, fields, and winepresses in the countryside, Wright (2004: 240) suggests the repeatedly redactional nature of references to Jerusalem in the section (vv. 15aβ, 16b, 18aβ, 19–20). He thus identifies the earliest layer as a problem of the Yehud nobles pressing others into service on the Sabbath (vv. 15a, 17–18*, 22*), without any exchange addressed.

112

Peter Altmann

transport and exchange, quite different from what one encounters in Deuteronomy 5/Exodus 20.22 The agricultural products mentioned in Neh 13:15 bring aspects of exchange into focus—grain (lit., “heaps,” hāʿărēmôt); then in the second half of the verse, wine, grapes, and figs. These could be exchanged at a market or delivered as tax payments.23 The mention of grain heaps could link with the large granaries identified in excavations,24 perhaps for provisioning Persian armies against Egypt. Of these actions, only the treading of grapes is related to a specific season. Treading a winepress, however, need not take place on a particular day, implying that none of the work needed to occur on the Sabbath.25 Whether it is state or patron extraction or private profit for the agricultural laborers that motivates the work, the text itself provides no answer.26 The final clause of the verse sums them up as “provisions” (ṣāyid), the apt translation in the jps, which notes their status as everyday items, rather than luxury products, which is an important contrast to v. 16.27 Thus the focus of v. 15 lies on community domestic production that could take place in any agricultural, rural community, regardless of the redistributive, market, or reciprocal nature of the economy. This conclusion would be true, however, only in a form of the text excluding mikrām (“their sale”), which points to market type exchange, parallel to Babylonia. Such a text would maintain the perspective of the Ten Commandments somewhat more clearly, where doing agricultural work is most clearly in focus for the Sabbath commandment (for different reasons in the Exodus 20 and Deuteronomy 5 versions, of course). Note that vv. 15b, 18, 22 are also seen as additions by Gunneweg (1987: 170). 22.  Reinmuth (2002: 288) states, “Der Nehemia-Bericht scheint eher an den Konkreta des Handelsverbotes interessiert.” Compare the similar tendency in Amos 8:5, as pointed out by Williamson (1985: 395). From Wright’s perspective, the purchase of goods—foreign goods brought by the Tyrians—appears next in the section, but still prior to the spatial move to Jerusalem, which appears later (only subsequent to the addition of the pact in Neh 10:1, 32–40[31–39]). 23.  It is surprising, however, that olive oil is missing from the list. 24.  Faust and Weiss (2005: 73) posit the supplying of grain for the coast (Ashkelon) in the preexilic period. 25.  This observation could support Wright’s contention of an inner-Judean issue. 26.  Note as well that v. 15 does not cite a formal prohibition, pointing instead to a less-official custom, which will culminate in v. 18. 27.  DCH 7.113 renders the verbal form “take along (as) provision(s)”; the noun is separated into two different entries (7.114): the first as game, catch, hunting (Gen 27:5, 33; Lev 17:13) and the second as provision (of food; Neh 13:15; Ps 132:15; Josh 9:5, 14; Job 38:41).

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

113

It comes as quite a surprise in Neh 13:16 when the spotlight moves from the Judean countryside and domestic issues to Tyrian merchants (in the MT) bringing fish and all kinds of wares to sell on the Sabbath. Important here is also the fact that they are selling them “to Judeans and in Jerusalem.”28 Trade is no longer within the domestic community, and Nehemiah thus formulates community boundaries in terms of economic transactions at the proper time. Both cultural and institutional forces appear to be at work. Edelman (2006: 207–46, esp. p.  212), following other interpreters, comments: “Thus, it seems that [Nehemiah] wanted to send an important message to both the native and the gēr in Jerusalem and Yehud concerning the sanctity of Shabbat.”29 The extremely attenuated state of Jerusalem during the whole Persian period makes the plausibility of Tyrian traders in Jerusalem during the middle of the 5th century questionable, and continuity with preexilic fish imports should not be overstated.30 However, regardless of when this verse became part of the text, it highlights the fact that any attempt to regulate market transactions on the Sabbath becomes quite problematic when not all the residents of the region adhere to the same religio­ethnic conviction about Sabbath regulations. Cultural and political groups and institutions thus surface here as important regulators of economic performance, much as one would expect with NIE. 28.  The absence of the Tyrians from the LXX, however, is quite puzzling: they do not appear in the rest of the section, nor do the issues focus on their action, thus suggesting that the MT may have added them, perhaps in connection with hārōkĕlîm (v.  20), in light of Tyre’s frequent association with traders using various forms of rōkēl in Ezekiel 27 (vv. 14, 15, 17, 20, 22, 24). Batten (1913: 295) notes the omission of “Tyrians” from the LXX and considers the LXX the better text, though he has difficulties constructing a smooth reading of the text. Yet it is in a curious support for Batten’s position that Reinmuth (2002: 290) identifies the “traders” in v. 20 with the Tyrians of MT v. 16 on the basis of Ezekiel 26–28. He never notes the omission from the LXX. Wright (2004: 225–38) argues similarly and makes the presence of the Tyrians a linchpin for his identification of vv. 16, 20–21 as later; however, he never provides text-critical arguments for including them in the text. Myers (1965: 213) argues for the MT because no good reading of a text without the Tyrians has been found, especially in light of the contrast with the bĕnê yĕhûdâ, to whom they were selling. 29.  Perhaps they were awarded trading rights with Yehud as a reward to Tyre for its shipbuilding on behalf of the Persian overlords? 30. Contra Noonan (2011: 281–98). The logic of the passage would be clearer without all of v. 16: One very speculative redaction-historical possibility is that the interruption of the Tyrian merchants was added in light of separation between the Tyrian and “Israelite” workers of Ezra 3 at a later stage when the various parts of the book of Ezra–Nehemiah were put together.

114

Peter Altmann

The selling of fish, as I mentioned above, should be set in contrast to goods collected, transported, and sold as “provisions” in v. 15 (Lernau and Lernau 1992: 131–48). As foreign goods, fish are probably designated products of elite consumption; unnecessary for survival, but functioning as a symbolic carrier of social status (MacDonald 2008: 38). Nehemiah turns his critique in vv. 17–18 against the “nobles of Judah” (ḥōrê yĕhûdâ), which is unexpected after v. 16 but may fit better with v.  15. The “nobles” come under fire from Nehemiah repeatedly (for example, 6:17–19), indicating that the book views them as the ones who could and therefore should steer the community in the direction of keeping the Sabbath.31 The content of vv. 17–18 (especially v. 18: “Indeed your ancestors did this and [as a logical consequence] our God brought all this evil upon this city”) postulates a connection between these elites and the preexilic elite responsible for the Babylonian destruction. This connection suggests, therefore, that these nobles were not just located in Yehud but also were identified as Judeans.32 This point is important because it means that Nehemiah attempts to regulate this economic issue through an in-group appeal, not force, and not directed against an outsider group.33 31.  Fried (2004: 201) notes six places in Nehemiah were “nobles” and “prefects” are mentioned together. Cf. J. Blenkinsopp (1988: 252): Nobles (ḥôrîm) and officers (segānîm) are often mentioned together in the book (Neh 2:16; 4:8, 13[14, 19]; 5:7; 7:5). The former were the hereditary Judean nobility (e.g., Jer 27:20, 39:6; Isa 34:12; see H. C. M. Vogt 1966: 107–11). The function of the latter never emerges clearly. In the prophetic texts from the 6th century, they are almost invariably linked with provincial governors, which would suggest regional administration (cf. Jer 51:23, 28, 57; Ezek 23:6, 12, 23). Both classes receive considerably censure in Ezra–Nehemiah. The segānîm were among the worst offenders in the matter of foreign marriages (Ezra 9:2). Both were charged by Nehemiah with economic exploitation (5:7). Several of the nobles maintained close contacts with his enemies (6:17). Neither class seemed particularly zealous in the specifics of religious observance (13:7, 11).

32.  It is open to question whether these nobles would acknowledge their connection to preexilic Judah/Israel, but the book of Nehemiah indicates that such a connection was at least plausible. 33.  The actual extent of “all this evil” is not developed, however, and this may suggest that the present subjugation of Jerusalem to Persia is included. Included could be subjugation to Persian rule, but the focus is even more likely to be the lowly, half-empty, and half-destroyed nature of Jerusalem in the Persian period. These verses, however, do not amount to more than an admonition, appealing to the individual motivations of the Judean elites. Cf. North (1981: 19), who states: “Compliance is so costly that the enforcement of any body of rules in the absence of some degree of individual restrain from maximizing behavior would render the political

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

115

The fact that the focus changes to the use of gubernatorial power to close the gates in vv. 19–22 indicates that Nehemiah’s appeal in vv. 17– 18 concludes the issue of agricultural work and selling on the Sabbath raised in the previous verses with an appeal to conscience. It is also difficult to conceive how Nehemiah would have enforced this prohibition through other means: even tiny Yehud was too large, and Nehemiah likely had too few resources to make sure it was followed. Such factors make cultural customs paramount for influencing a group’s actions. To borrow from NIE, ideology can be far more cost effective than enforcement. The measures taken by Nehemiah in vv. 19–22 do not appeal to the sense of group belonging and the religiously inspired ethic of the nobles. These verses instead portray Nehemiah as an authority who is able to put his wishes into practice through force in Jerusalem. As others have noted, the closing of Jerusalem’s gates for the Sabbath would only be enough to keep peddling from taking place in the city, but the traders could stay outside the gates in hopes that buyers would come out to them through small passageways. This problem gives rise to Nehemiah’s threats to the traders to stay away from Jerusalem on the Sabbath.34 Important here is the interaction between individual attempts to profit or economic institution non-viable—hence the enormous investment that is made to convince individuals of the legitimacy of these institutions.” 34.  Comparison with Jer 17:21–22, which forbids carrying burdens out of one’s house or into Jerusalem on the Sabbath is enlightening. The prohibition concerning Jerusalem of course shows thematic connections with Neh 13:15b, though Jer 17:21 specifically mentions “bringing into the gates of Jerusalem,” which arises as a major point of conflict in Neh 13:19–22. Wright rejects the opinion of many interpreters that this transport of goods into Jerusalem implies trade, which brings up the question of just what people would be bringing the goods into Jerusalem for. He can only make this argument by seeing v. 15b as a later addition (2004: 223). His position is supported by Jeremiah 17, which does not provide any conclusive evidence for the gates as a place of trade but instead contrasts lifting burdens through Jerusalem’s gates with bringing cultic offerings through the gates (Jer 17:26). The connection in Jer 17:20, 25 between Sabbath obedience and the security of the Davidic throne may also lead to understanding the goods as tributary burdens, again possibly similar to the denunciation of the Judean nobles in Neh 13:17. So there is some question as to whether the people would be bringing goods into Jerusalem on the Sabbath for trade in Jeremiah 21—tax would also be an option. The fact that Jeremiah’s critique is directed toward the people as well as the kings, taken as a whole, again tips the scales in the direction of a critique of trade. For what other reason would the people be criticized for bringing agricultural goods into Jerusalem on the Sabbath? Perhaps instead of bringing offerings—the desired result—they were bringing goods for trade or tribute. As a result, the questions are intrinsically economic—are the goods to enrich the cult, or are they to enrich individual or royal coffers?

116

Peter Altmann

(both on the side of the sellers and the buyers) and Nehemiah’s institutional mandate, which he relates in v. 22 with his appeal for status before God. Finally, attention to the “the day of their selling” (v.  15b) and the Tyrian traders (v.  16) necessitates a focus on trade. It is striking that the groups reprimanded are the Judean nobles, specifically, and an ethnically unidentified group of traders (v.  20), rather than “the peoples of the land” with whom one should avoid trading on the Sabbath in 10:32[31]. The Tyrians, who also import cedar for the temple along with Sidonians in Ezra 3:7, appear in a positive light in Ezra–Nehemiah. Here, as well, their presence as dwellers in Jerusalem does not seem to raise any issues in and of itself. The problem is instead that the Judeans purchase goods from them on the Sabbath. Taking lāhem in 13:17 to refer to the Judeans (and those in Jerusalem) as the nearest referents, it is the community of Yehud that bears the responsibility for observing the Sabbath.35 These Sabbath buyers, especially those purchasing the Tyrian fish, likely belong to the elite. As such, they are also those most likely to be patrons. When we think of the structure of the Yehud economy, this group appears time and again to play key roles: for example, they are those with the capital to lend to small landholders (5:1–13, esp. v. 7, which names the nobles and officials).36

Conclusion How do the dealings in Nehemiah 13 compare with those in Babylonia? In spite of coming from a slightly earlier period, the Babylonian 35.  This regulation extends the Sabbath in a new manner, generally unknown in earlier biblical texts. Whereas, there is the one possibility (Amos 8:5) of an earlier prohibition of selling on the Sabbath—or, more specifically, the greedy desire for the Sabbath to end so that one could return to selling—Neh 13:16–17 lays the onus on the buyers regarding such practice on the Sabbath. 36.  The logic implied in Nehemiah’s prohibition of buying is the protection of Judean merchants: the Tyrian traders would have an unfair advantage if they were the only ones selling on the Sabbath. According to this view, there is no intrinsic problem with buying from non-Judeans or even a preference for in-group merchants but, rather, a desire to create a “fair” playing field for economic profit. This explanation extends both to the prohibitions in vv.  16–17 and also to the closing of the gates and dispersal of merchants from outside them in vv.  19–22. No later than with the sanctification of the gates by the Levites in v. 22 (often viewed as a late insertion), the inclusion of commercial trade into the dynamics of the sacred obtains. As Eskenazi (1988: 113) argues, this action brings about the sacralization of the entire (though tiny!) Persian period city of Jerusalem. The Levites’ presence as cleansing guards of the wall to sanctify the Sabbath day indicates the inclusion of the gates in the holy space.

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

117

data display a far more developed market trade system, even in the once “redistributional” temple economies.37 In Yehud, we have no primary sources testifying to significant market trade. Neh 13:15b alone portrays some daily “provisions” as being on sale in Jerusalem in the Persian period.38 Appropriating insights from NIE, however, one can view the external traders in Jerusalem who provide products for elite consumption—likely in exchange for regional agricultural harvests—as suggesting somewhat freer markets (the buyers and sellers do not appear predetermined). These traders’ presence points to some kind of interregional trade that was profitable enough to be worth the risks involved in undertaking it. For such trade to take place, certain structures, governmental and otherwise, had to be in place to maintain an adequate level of transaction costs, such as relative security and the ability for sellers and buyers to find one another. The imports influence only the domestic market exchange indirectly. They likely impinged on the economic system similarly to the trade exchanges for agricultural products in the hinterlands of Babylonia, especially if one considers the mention of grain imports from Judah bound for Tyre in Ezek 27:17 in this context. These agricultural exports brought some market exchange, perhaps providing the silver necessary for the royal tax mentioned in Neh 5:4. The introduction of New Institutional Economics has revealed how the long-estranged individualistic formalist economic approach can overlap with substantivist concerns for cultural-political curbs on economic structures and performance. As both Babylonia and Yehud provide evidence of both market thinking and concern for social status, further exploration and appropriation of this approach could prove helpful for the evaluation and modeling of their economies in the Persian period. 37.  One disclaimer is that they are from slightly different eras—Jursa’s analysis generally ends in the early 5th century, while Nehemiah 13 could certainly only represent a time post-450 b.c.e. 38. This statement can be viewed as an addition (Wright 2004: 222–23; Gunneweg 1987: 170).

118

Peter Altmann

Bibliography Altmann, P. 2014 Tithes for the Clergy and Taxes for the King: State and Temple Contributions in Nehemiah. CBQ 76: 215–29. Bang, P. F. 2009 The Ancient Economy and New Institutional Economics: Review of The Cambridge Economic History of the Greco-Roman World, by W. Schei­del, I. Morris, and R. Saller. Journal of Roman Studies 99: 194– 206. http://dx.doi.org/10.3815/007543509789744783. Batten, L. W. 1913 A Critical and Exegetical Commentary on the Books of Ezra and Nehemiah. ICC 11. Edinburgh: T. & T. Clark. Bedford, P. R. 2007 The Persian Near East. Pp. 302–29 in The Cambridge Economic History of the Greco-Roman World, ed. W. Scheidel, I. Morris, and R.  Saller. Cambridge: Cambridge University Press. Blenkinsopp, J. 1988 Ezra–Nehemiah: A Commentary. OTL. Philadelphia: Westminster. 2009 Judaism, the First Phase: The Place of Ezra and Nehemiah in the Origins of Judaism. Grand Rapids, MI: Eerdmans. Edelman, D. V. 2006 Tyrian Trade in Yehud under Artaxerxes I: Real or Fictional? Independent or Crown Endorsed? Pp. 207–46 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. 2011 Apples and Oranges: Textual and Archaeological Evidence for Reconstructing the History of Yehud in the Persian Period. Pp. 133–44 in Congress Volume: Helsinki 2010, ed. M. Nissinen. VTSup 148. Leiden: Brill. Eskenazi, T. C. 1988 In an Age of Prose: A Literary Approach to Ezra–Nehemiah. SBLMS 36. Atlanta: Scholars Press. Faust, A., and Weiss, E. 2005 Judah, Philistia, and the Mediterranean World: Reconstructing the Economic System of the Seventh Century b.c.e. BASOR 338: 71–92. Fried, L. S. 2004 The Priest and the Great King: Temple-Palace Relations in the Persian Empire. Biblical and Judaic Studies from the University of California, San Diego 10. Winona Lake, IN: Eisenbrauns. Fusfeld, D. B. 1957 Economic Theory Misplaced: Livelihood in Primitive Society. Pp. 342–56 in Trade and Market in the Early Empires Economies in History and Theory. New York: Free Press. Guillaume, P. 2012 Land, Credit and Crisis: Agrarian Finance in the Hebrew Bible. Bible World. London: Equinox.

Ancient Comparisons, Modern Models, and Ezra–Nehemiah

119

Gunneweg, A. H. J. 1987 Nehemia. KAT 19/2. Gütersloh: Gütersloher Verlag. Hopkins, K. 1983 Introduction. Pp. ix–xxv in Trade in the Ancient Economy, ed. P. Garnsey, K. Hopkins, and C. R. Whittaker. London: Chatto and Windus. Houston, W. J. 2008 Contending for Justice: Ideologies and Theologies of Social Justice in the Old Testament. 2nd ed. LHBOTS 428. London: T. & T. Clark. Jursa, M. 2004 Grundzüge der Wirtschaftsformen Babyloniens im ersten Jahr­ tausend v. Chr. Pp.  115–36 in Commerce and Monetary Systems in the Ancient World: Means of Transmission and Cultural Interaction— Proceedings of the Fifth Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project, Held in Innsbruck, Austria, October 3rd–8th, 2002, ed. R. Rollinger and C. Ulf. Oriens et Occidens 6. Stuttgart: Franz Steiner. 2008 Economic Change and Legal Innovation: On Aspects of Commercial Interaction and Land Tenure in Babylonia in the First Millennium bc. Pp.  601–28 in I diritti del mondo cuneiforme, ed. M. Liverani and C. Mora. Pavia: Iuss. 2010 Aspects of the Economic History of Babylonia in the First Millennium bc: Economic Geography, Economic Mentalities, Agriculture, the Use of Money and the Problem of Economic Growth. AOAT 377. Münster: Ugarit-Verlag. Kessler, J. 2006 Persia’s Loyal Yahwists: Power Identity and Ethnicity in Achaemenid Yehud. Pp. 91–121 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. Lernau, H., and Lernau, O. 1992 Fish Remains. Pp. 131–48 in Excavations at the City of David, 1978– 1985, Directed by Yigal Shiloh, vol. 3, ed. A. De Groot and D. T. Ariel. Qedem 33. Jerusalem: Institute of Archaeology, Hebrew University of Jerusalem. MacDonald, N. 2008 What Did the Ancient Israelites Eat? Diet in Biblical Times. Grand Rapids, MI: Eerdmans. Mauss, M. 1984 Die Gabe: Form und Funktion des Austauschs in archaischen Gesellschaften. 3rd ed. Frankfurt: Suhrkamp. Myers, Jacob M. 1965 Ezra, Nehemiah. AB 14. Garden City, NY : Doubleday. Nam, R. S. 2012 Portrayals of Economic Exchange in the Book of Kings. BibInt. Leiden: Brill. Narotzky, S. 1997 New Directions in Economic Anthropology. London: Pluto.

120

Peter Altmann

Noonan, B. J. 2011 Did Nehemiah Own Tyrian Goods? Trade between Judea and Phoenicia during the Achaemenid Period. JBL 130: 281–98. North, D. C. 1981 Structure and Change in Economic History. New York: Norton. 2005 Understanding the Process of Economic Change. Princeton Economic History of the Western World. Princeton: Princeton University Press. Polanyi, K. 2001 The Great Transformation: The Political and Economic Origins of Our Time. 2nd ed. Boston: Beacon. Reinmuth, T. 2002 Der Bericht Nehemias: Zur literarischen Eigenart, traditionsgeschicht­ lichen Prägung und innerbiblischen Rezeption des Ich-Berichts Nehemias. OBO 183. Freiburg: Universitätsverlag. Schaper, J. 1995 The Jerusalem Temple as an Instrument of the Achaemenid Fiscal Administration. VT 45: 528–39. Scheidel, W.; Morris, I.; and Saller, R., eds. 2007 The Cambridge Economic History of the Greco-Roman World. Cambridge: Cambridge University Press. Schunck, K.-D. 2009 Nehemia. BKAT 23/2. Neukirchen-Vluyn: Neukirchener Verlag. Silver, M. 2004 Modern Ancients. Pp.  65–87 in Commerce and Monetary Systems in the Ancient World: Means of Transmission and Cultural Interaction— Proceedings of the Fifth Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project, Held in Innsbruck, Austria, October 3rd–8th, 2002, ed. R. Rollinger and C. Ulf. Oriens et Occidens 6. Stuttgart: Franz Steiner. Van de Mieerop, M. 2004 Economic Theories and the Ancient Near East. Pp. 54–64 in Commerce and Monetary Systems in the Ancient World—Means of Transmission and Cultural Interaction: Proceedings of the Fifth Annual Symposium of the Assyrian and Babylonian Intellectual Heritage Project, Held in Innsbruck, Austria, October 3rd–8th 2002, ed. R. Rollinger and C. Ulf. Oriens et Occidens 6. Stuttgart: Franz Steiner. Williamson, H. G. M. 1985 Ezra, Nehemiah. WBC. Waco, TX: Word. Wright, J. L. 2004 Rebuilding Identity: The Nehemiah-Memoir and Its Earliest Readers. BZAW 348. Berlin: de Gruyter.

‫ּורבּו‬ ְ ‫ ְפרּו‬and the Seventh Year: Complementary Strategies for the Economic Recovery of Depopulated Yehud Philippe Guillaume Universität Bern

Living in a fast-changing and hectic world, today’s biblical exegetes tend to imagine the ancient world as inherently static, where change occurred only slowly—in particular, in the domains of economy and demography. In fact, studies of Iron Age Greece reveal fast-growing cities and a buoyant economy that weathered crises thanks to a thriving rural demography and to the favorable climatic conditions that prevailed in the Mediterranean world throughout the first millennium b.c.e. (Morris 2009: 64–80). This contribution applies this dynamic view of Mediterranean economy to the revival of the economy of Yehud to supply the needs of the military operations of the Achaemenids in and against Egypt.

Preliminary Remarks I take it for granted that what qualifies today as economics existed in some archaic forms in the ancient world and in the Levant, even if Polanyi’s claim (1944) that independent institutional economic structures did not rise before the late 17th century c.e. is correct. Polanyi (1998: 64–97; see also Lowry 1998: 1–10) did broaden his view of economic activities as merely embedded in other social and political institutions and eventually recognized the presence of economic thought as early as in Aristotle’s writings. I do not consider it necessary to enter the old formalist-versussubstantivist debate (Lowry 1979: 65–86). For the present purposes, it suffices to note that the ancient world in general—and more specifically, the Levant—did possess an emerging awareness of the interaction between production, consumption, and transfer of wealth and how best to benefit from the effects of these basic economic principles. I am not convinced that biblical texts outside the Torah are helpful to reconstruct the agrarian economy of depopulated areas in Yehud before 123

124

Philippe Guillaume

the Achaemenid administration revived the economy in the Palestinian highlands. The few verses reporting Gedaliah’s term in 2 Kings 25 and Jeremiah 40 reflect the first stage after the destruction, the initial period of euphoria before the effects of depopulation began to be felt. As for Chronicles, Ezra, and Nehemiah, they mostly reflect conditions prevailing during the Hellenistic era (for a different position, see, e.g., Leuchter 2010: 345–65). Therefore, the input from archaeological surveys is essential, but some limiting factors must be taken into consideration to avoid overinterpretation. The publication of the results of systematic surveys and of salvage excavations enables quantitative studies with population estimates. There is a consensus that the destruction of Jerusalem had disastrous effects on the immediate periphery of the city. In Benjamin, the population gathered around large tells of the area: Mizpah/Tell en-Nasbeh, Gibeon/el-Jîb, and Gibeah/Tell el-Fûl. The case of Bethel/Beitin is disputed. Israel Finkelstein and Lily Singer-Avitz (2009: 33–48) claim that Bethel was probably uninhabited or almost deserted in the Babylonian and Persian periods. Their claim that most of the tell was excavated is exaggerated, and it diminishes the significance of the pottery repertoire they examined. The possibility remains that ancient Bethel was situated elsewhere, at el-Bireh rather than at Beitin, and that Beitin was Beth-Aven (Wood 2008: 205–40). Whether or not Bethel was inhabited, the Benjamin region was populated, and a second demographic cluster spanned the ridge south of Jerusalem, between Beth Zur and Bethlehem. Archaeological surveys suggest that only a third of Late Iron Age sites were occupied during the Persian period (216 sites out of 586). Moving from this rough estimate to the calculation of actual population levels is, however, wrought with difficulties, as the results hinge upon the size of each site and the density of its occupation (Faust 2007: 33). Hypotheses about the population in any settlement can vary as much as 66% (Zevit 2007: 440–41). Moreover, estimates of the number of sites inhabited at any period are rough, and inferring the population of such numbers of sites is meaningless.1 Nevertheless, I accept the present consensus that during the 6th century b.c.e. the highlands lost 80–90% of 1.  Despite a detailed analysis of surveys, Edelman (2005: 316) goes even further by stating that none of the percentages for the new settlements established during the Persian period “can be relied upon to reflect the actual settlement trends accurately. None should be cited in arguments concerning the changes in settled sites or demographics between the Neo-Babylonian and Persian periods.”

Strategies for the Economic Recovery of Depopulated Yehud

125

their population and that it took the two Persian centuries to achieve a modest recovery that merely doubled the Babylonian low (Carter 1999: 247; Faust 2007: 42; Lipschits 2004: 99–108). It is clear that the land was not empty after the destruction of Jerusalem and that the restoration of Jerusalem did not bring a return to the demographic and economic levels of the 7th century b.c.e. Although the impact of the shivat Zion on the demography of Yehud has certainly been exaggerated in biblical studies (Lipschits 2005: 372), the revived strategic importance of Yehud entails some immigration for the people to have revived the economy of the depopulated southern Levant. From this perspective, whether or not they were “returnees” is of little importance compared with their urgent need to increase the available manpower quickly if food surpluses were to be stockpiled in the Shephelah in preparation for military campaigns in Egypt.

Depopulation as Dearth of Workers Depopulation means that there was no pressure on arable land, which before modern times had no value in itself (Guillaume 2012: 271–330). Workers were the crucial resource. This was true for earlier periods and remained true throughout the Achaemenid era, as the development of military fiefs (hand, bow, horse, and chariot fiefs, and ḫaṭru estates) in Mesopotamia and the 12 arourae of tax-free land granted to the Egyptian machimoi2 shows. People serving in the army received no salary to defray the cost of service. Instead, they obtained revenues by cultivating the land they were granted. The large-scale outsourcing of labor on military fiefs revealed by the Murašu archive confirms that workers, rather than land, remained in short supply in the Persian Empire (Stolper 2001: 83–132). In depopulated Yehud, most of the arable land around Jerusalem was fallow. After half a century, two-thirds of the previously cultivated fields were either in advanced stages of reforestation or at least covered with shrubs and dense Mediterranean maccia evergreens (Liphschitz 2008: 48–55), of the kind that claimed more victims than the sword because it can easily catch on fire (Isa 9:17; 10:17). Absalom was reportedly caught in it (2 Sam 18:8–10). The absence of pressure on land together with the abundance of wild fauna and flora discouraged intensive agriculture. Beyond this general picture, archaeology can contribute little to the understanding of this era marked by depopulation and by small scattered settlements, which apart from the administrative centers of 2.  Herodotus 2.164, 168; Ruzicka 2012: 21–22.

126

Philippe Guillaume

Ramat Raḥel and Lachish left scant remains.3 This archaeological deficit brings back to the fore the literary sources, in particular the Torah as the main source for that period, in contrast to the common reliance on the books of Nehemiah, Ezra, or Chronicles for the historical reconstruction of the Persian era (Lipschits 2011: 197 n. 63). The very existence of the Torah signals the presence of a group of highly literate personnel with adequate resources to produce such a corpus of texts (Ben Zvi 2011: 102). Since it is unlikely that resources were allocated primarily for this purpose, I take it for granted that the writers of the Torah were involved in the running of the administrative apparatus, keeping ledgers, recording taxes and the delivery of goods, and calculating interest rates (Jursa 2004: 145–98). The production and reproduction of literary texts were secondary activities for the literati, who did it as part of the training of other scribes (Ben Zvi 2011: 102). For provincial scribes, as much as for their superiors in the central administration, managing the workforce was a constant headache. Mobile and mortal, workers were also in scarce supply. The surviving documentation reveals the competition between administrators, contractors, institutions, and governors to obtain a workforce. Working parties were chronically undermanned. Work assignments were excessive, rations and tools inadequate, workers famished or dead, and fugitives were brought back in chains when they were caught. The available workforce was conscripted by competing overseers, and legal texts deal with the settlement of the conflicts generated by the difficulty of meeting the expected workload with ever-failing resources (Stolper 2003: 265–87). Given the limited size of Yehud and its modest agricultural potential, the number of active literati must have been small, but the amount of texts they produced and handled suggests that their daily chores left them enough spare time and resources to devote themselves to literary work (Ben Zvi 2011: 141–42). Rooted as they were in social realities that they actively contributed to, the Yehud scribes produced writings that are an abundant source of information about the consequences of depopulation in Yehud and the measures that were taken to remedy it.4 3.  The storage centers at Tel Halif, Tell el-Hesi, and Tel Haror belonged to the Gaza network rather than the Yehud (Edelman 2005: 319). For other potential administrative centers within Yehud located between the northernmost #265 (Rujm Abu Hashabe) and the southernmost #276 (Beth-Zur), see Edelman 2005: 309, fig. 51. On Beth-Zur, see Reich 1992: 247–52. 4.  If, as Ben Zvi (2011: 103) argues, it is legitimate to piece together the social memory of the literati out of the texts they wrote, since social memory is not a dis-

Strategies for the Economic Recovery of Depopulated Yehud

127

‫ּורבּו‬ ְ ‫פרּו‬: ְ A Reflection of the Achaemenid Economic Revival In the absence of machinery, the output of ancient agrarian economies depended heavily on the number of hands that could be sent to the fields at the crucial moments of crop growth and maturation. Hence, it is significant that demography stands out as one of the main issues in the first chapters of the book of Genesis. In Gen 1:22, Elohim blesses the sea monsters, fish, and birds, telling them, ‫ּורבּו‬ ְ ‫“ ְפרּו‬fructify, multiply, and fill the sea and the land.” In v. 28, Elohim blesses humankind with the same words: ‫ּורבּו‬ ְ ‫פרּו‬. ְ If these words were meant as a commandment, no divine command has been more faithfully obeyed than this one. If they were given as a blessing, then since the days of Robert Malthus the side effects of the blessing have become a cause of much worry. Against the notion that human populations are capable of almost limitless growth, Malthus prophesied that eventually population levels would be checked by famine and disease. Malthus (1999: 61) worried that the power of reproduction is greater than the land’s potential to produce food. Disastrous overpopulation is bound to occur because the availability of land is limited. In contrast to Babylonian flood stories, the biblical flood does not function as a Malthusian demographic check. Instead of overpopulation leading to famine and disease, Genesis attributes humanity’s predicament to violence and sin (Gen 6:1–8). After the biblical flood, the Hebrew writers repeat the divine order to multiply. The only difference with Genesis 1 is that the animals are no longer called to multiply. The writers obviously lived in a world teeming with wildlife to the point that wild animals threatened the scarce human population. Hence, the call to master the animals in Genesis 1 turns into a call to terrorize them and eat them in Gen 9:2. Given the huge demographic increase since Malthus, these passages in Genesis have become very unpopular, and theologians tend to water down the harshness entailed by the verbs kābaš in Gen 1:28 and rādâ in Gen 9:2 (Wittenberg 2010: 427–53). Even though there is no doubt that it has now become urgent to transform the call to dominate and subdue the animals into a call to protect creation, from the point of view of a scribe in the Palestinian mountains over 2,000 years ago nothing short of a demographic explosion was necessary before humanity would be in a position to fulfill the renewed call to fill the land (Walton 2011: 177). cursive system and thus cannot be identified with any book. It is at least as legitimate to search for echoes of actual practices in their writings.

128

Philippe Guillaume

Filling the land depended on humanity’s ability to subdue the animals so that they would stop eating people.5 In such a world, heroes proved themselves by killing bears and lions. In Genesis, animals are so threatening that Elohim guarantees the survival of humans by stating that he will personally charge blood money from murderers of humans, whether human or animal. How Elohim would force lions and bears to pay compensation is not mentioned, but this weird image reinforces the urgency of the call to multiply. No human for whatever reason should be killed because doing so would put demography and the survival of humanity in jeopardy. The present level of human demography can certainly not be attributed entirely to the ‫ּורבּו‬ ְ ‫ ְפרּו‬formula.6 There are many other factors involved. Had the Genesis writer been a prophet, had he foreseen the present size of human demography, would he have toned down his calls to multiply? No one can answer that question, but I believe that Genesis had good reasons to maintain the call to multiply after the flood, which leads us to considering the contrast with the Babylonian flood. The Contrast with Other Attitudes toward Human Reproduction The postdiluvian ‫ּורבּו‬ ְ ‫ ְפרּו‬formula marks a sharp contrast with Atrahasis, the supersage who listed population checks at the conclusion of the Babylonian flood: sterility, celibacy, and child mortality (Hallo 1997: 1.452). Whether or not the Genesis writers had a copy of Atrahasis in their library is moot. What is clear is that the attitude toward reproducing in Iron Age Judah is the opposite of that in Bronze Age Mesopotamia. Childless goddesses and childless female cult functionaries in Mesopotamia show that the terms fertility and fertility magic used to describe Mesopotamian rituals are tainted by the biblical ‫ּורבּו‬ ְ ‫ ְפרּו‬slogan. The nadītus enforced childlessness inasmuch as the conclusions of the Babylonian flood stories indicate that Mesopotamian cultures worried more about overpopulation than about underpopulation (Assante 2003: 24–26).The same can be said for earlier periods. Paleontologists claim that our prehistoric ancestors used magic as much to control reproduction as to spur it, and that they were generally more interested in lim-

5.  I doubt that the absence of rdh in Gen 9:7 shows that God has assumed the care of the animals: pace Schellenberg 2009: 97–115. Verse 7 is a repetition of v. 1, framing the crucial words. 6.  On the supposed role of Gen 1:28 in the present ecological crises, see White 1967: 1203–7.

Strategies for the Economic Recovery of Depopulated Yehud

129

iting population growth than in increasing it (White 2003: 58; Luquet 1926; Ucko and Rosenfeld 1967). Given this millennia-long tradition of efforts to curb population growth, some exegetes suggest that the Genesis writers were reacting to Mesopotamian models (Kilmer 1972: 160–77). This is possible since the biblical writers were also depicting Yhwh in opposition to vegetation gods such as Baal, who dozed throughout the summer and required the mournful wails of women at autumn festivals to waken. If Yhwh never sleeps because he is never tired, phonic pollution caused by overpopulation is of no concern to him, even though the Genesis writers could imagine cities that were loud enough. Writing in a world where towns were villages and where travelers were attacked by bears or lions (Judg 14:5; 1 Sam 17:36; 2 Kgs 2:24), the Genesis scribes saw no need to have Elohim bless wildlife with more reproductive power or to have him strike humanity with sterility.

Demography as a Crucial Factor of Economic Growth Today, the dread that Noah’s descendants were supposed to inspire in the animals has turned into the dread of seeing the entire creation collapse under the weight of humanity. Overpopulation seems far more threatening than it did in the days of Malthus, but this does not prove that Malthus was right. Although famines continue to strike in parts of the world, overpopulation is not their primary cause. Economists disagree on whether population is a check to human development, but the legitimacy of the very concept of overpopulation is far from certain. In antiquity, demography in the form of the ideal size of a family was already a debated matter. Hesiod, a contemporary of the first major phase of the production of Hebrew literature, weighs out the disadvantages of a large number of offspring. He advises his brother in this way: “Let there be a single-born son (μουνογενὴς δὲ πάις εἴη) to nourish the father’s house: in this way wealth is increased in the halls” (Hesiod, Works and Days, line 376). The first impression is that Hesiod saw large families as a threat to farmers and thus tied prosperity to strict population control: each farmer ought to father only one child. But Hesiod adds: “and may he die an old man (γηραιὸς, masculine) leaving behind one son in his turn.” Fathering a single son is risky because the child may die before reaching maturity and having time to produce a son of his own. Therefore, Hesiod reverses his initial advice: “And yet Zeus could easily bestow immense wealth upon more people: more hands, more work, and the surplus is bigger. If the spirit in your breast longs

130

Philippe Guillaume

for wealth, then act in this way, and work at work upon work” (Hesiod, Works and Days 376–81).7 Contrary to his initial claim, Hesiod insists that the limiting factor to economic growth is the workforce. The equation is simple: more hands equal more work done, which at the end of the season means more surplus. Hesiod is no Malthusian, and his argument presupposes two important factors: (a) political stability, without which farmers are prevented from accomplishing their tasks and are robbed of the fruit of their labor; and (b) unlimited availability of arable land. When these two conditions are met, the only way to increase production is by increasing the area of cultivated land, and the only way to increase land under cultivation is to increase the number of workers in the field. Works and Days is not an agricultural treatise per se (Nelson 1996: 45–53), but there is little doubt that Works and Days reflects more or less faithfully the general situation in Beotia between the 7th and 5th centuries b.c.e. Hesiod’s entire argument would fail if his audience could imagine his brother Perses having good reasons to reply that land shortage forced him to leave the village of Ascra and seek a living in the city of Thespiae. Hesiod saw work as the key to success. If you want to become rich, you must produce many children and work, work, and work some more. Hesiod is convinced that Zeus increases the wealth of the man who has more than one child for the simple reason that he puts more hands to work in the fields. Elsewhere in Works and Days, Hesiod mentions slaves, but he is aware that they cannot replace children. Slaves are costly, they may run away, and there is no guarantee that they will be more healthy and hardworking than children. Although he repeats the myth of Pandora, which attributes all evils to women, Hesiod advises his contemporaries to marry a local girl and father children if they want to see their farms prosper (e.g., Works and Days 380, 698–700; cf. 228). Hesiod’s view on the subject accords with Ps 128:1–3, where happiness is described as the ability to eat the fruit of one’s labor thanks to the portrayal of the farmer’s wife as a fruitful vine and the couple’s sons as olive shoots around the patriarch’s table. Like Hesiod, the psalmist ignores daughters—modern translations often replace the word “sons” by the more gender-neutral “children.” Likewise, the death of Job’s 7 boys and 3 daughters, which is often understood as being “compen7.  ῥειῖα δέ κεν πλεόνεσσι πόροι ζεὺς ἄσπετος ὄλβον πλeίων μὲν πλεόνων μελέτη, μείζων δ’ ἐπιθήκη (Most 2006: 119 [latest Loeb edition]).

Strategies for the Economic Recovery of Depopulated Yehud

131

sated” by the birth of an equal number of sons and daughters, may well have been “compensated” in a very different manner. The Hebrew ‫( שׁבענה‬Job 42:13) is a dual form of 7. According to the writer of this text, the compensation entailed Job’s recouping twice the amount of his losses—that is, 14 sons but only 3 daughters, just as he originally had. The birth of 6 daughters would have been a curse rather than a blessing. But whereas Hesiod, Job, and Psalm 138 (cf. Sir 22:3) view daughters as a burden, Genesis sees them as assets. Contrary to Hesiod, Job, and Psalm 138, the ‫ּורבּו‬ ְ ‫ ְפרּו‬formula in Genesis makes no distinction between sons and daughters.8 Lev 26:9–10 establishes a clear correlation between food abundance and large families: “I will look with favor upon you and make you fruitful and multiply you; and I will maintain my covenant with you. You shall eat old grain long stored, and you shall have to clear out the old to make way for the new” (Lev 26:9–10). Instead of viewing an increase in the number of mouths to be fed as a likely cause of food shortage, Leviticus 26 agrees with Hesiod that more people equal more grain. The prerequisite for more surplus is the farmer’s ability to multiply, and the masculine-plural pronoun ‫ אתכם‬designates that the addressees can include daughters. That large families entail more mouths to feed was not viewed as a problem for the simple fact that children were not idle; they worked on the farm as did the adults. Jeremiah mentions daughters alongside sons in his call to the exiles to prosper in Babylonia (Jer 29:6). Despite divergent attitudes regarding the value of daughters, none of these ancient writers saw land as a limiting factor. They all considered land as an unlimited quantity available to anyone who had the ability and the willingness to work it. I belabor the point because Malthusian views are spreading in biblical exegesis, particularly among so-called social-scientific exegetes, who often assume that land was in short supply already in the 8th century b.c.e. (Guillaume 2012: 90–115). As the story goes, the biblical oracles against the rich reflect the appearance of a proletariat of landless farmers because of the rise of the Israelite state. Contrary to the ancient authors mentioned above, social-scientific exegetes consider populations as players in a zero-sum game in which arable land is a constant that can only produce a limited amount of food. 8.  Gen 17:2, 20; 22:17; 26:4, 24; 28:3; 35:11; Exod 32:13. See also Deut 6:3; 7:13; 13:17; Jer 23:3; Ezek 36:10; 37:26; Ps 107:38. But Jer 30:20 has sons, and Ezek 36:11 has ‫ָדם‬ ָ ‫ א‬and ‫ב ֵהמָה‬. ְ

132

Philippe Guillaume

In a zero-sum game concept, the more mouths to feed, the less on each plate because the amount of available food is limited by the amount of land available for cultivation rather than by the number of hands in the field. This is the opposite of what we find in the Hebrew Bible and in Hesiod. Agricultural land in the biblical world was never an invariable quantity. Agricultural land was and still is a highly variable factor. It increases and decreases with demographic levels. At least until the last century, wealth was proportional to population levels. In this case, Hesiod and the Torah were correct in tying prosperity to a vibrant demography.9 Until modern times and the arrival of tractors, the problem of farmers the world over was not scarcity of land but scarcity of hands. That was true in Ottoman times and all the more so in the Iron Age. The notion of land scarcity is absent from Works and Days, although Hesiod was a contemporary of the great population explosion discussed below. However, Hesiod never complains about land shortage, nor does he tie poverty to an insufficient holding (Edwards 2004: 37, 109). For Hesiod, poverty is caused by laziness or a lack of workers. Exod 23:23 neatly illustrates what happens when the number of farmers (and children) drops. Yhwh promises to drive out the Canaanites by stages only, in order to avoid the proliferation of wild animals that would ensue if the small Israelite flock exterminated the Canaanites all at once. Despite Exod 23:32–33, which claims that a progressive elimination of the Canaanites increases the risk of having Israel contaminated by Canaanite practices, v.  23 insists that the challenge caused by wild animals is greater than “Baalism.” It was better to keep the Canaanites alive than to have the land infested with lions. This perspective is similar to Genesis 9: the greatest danger that faced humanity in those days was wild animals. 9.  It is not certain that the point at which high demographic levels turn into overpopulation has been reached even today; surely it was not in the first millennium b.c.e. Ilan (2011: 152) nevertheless interprets the size of houses at Tell Dan during the Iron Age I as a sign of overpopulation: “When land is scarce it is most often transferred to one son as an impartible package. Family members who cannot inherit land either receive something else instead (a house or some other form of capital), find a new means of making a living (e.g., craft specialization), or leave the household to seek their fortune elsewhere.” The implicit notion is that, if the father of a large family divided his farm between all his sons, the resulting land-shares would be too small for each son to support his own family. Surrounded by a lush plain with abundant water, Dan is one of the most blessed biblical regions, while the Iron Age I had some of the lowest demographic levels of all the biblical ages. How could the size of houses at Dan reflect overpopulation caused by land scarcity?

Strategies for the Economic Recovery of Depopulated Yehud

133

Dynamic Iron Age Demography Infant mortality, the death of parturient mothers, short life span, and recurrent famines were far more prevalent in the ancient world than they are today. Yet, the negative impact of these factors on demographic growth in the Mediterranean world of the Iron Age is surprisingly limited. A few centuries after Hesiod, Greek philosophers were more worried by over-population than by the lack of people.10 In the 4th century b.c.e., Aristotle saw a large increase in population as a threat to the ideal city: [I]t is more necessary than ever to regulate property, to take care that the increase of the people should not exceed a certain number; and in determining that, to take into consideration those children who will die, and also those women who will be barren; and to neglect this, as is done in several cities, is to bring certain poverty on the citizens; and poverty is the cause of sedition and evil. (Politics 2.6)

Aristotle’s fears are in sharp contrast to what we read in Hesiod and in the Torah. The change of attitude in ancient Greece reflects the demographic explosion that occurred between 900 and 300 b.c.e., when the population increased tenfold, with a population doubling every 20 years in Athens and Argos (Morris 2006: 27–43).11 Yet, these two cities which experienced the most spectacular demographic explosion founded no colonies (Tandy and Neale 1996: 13). This apparent paradox is easily explained. As Greek population increased, standards of living increased even faster, which proves that Hesiod and Genesis were right (Morris 2004: 160). More hands mean more surplus. Demographic increase did not lead to land shortage. Instead of causing new problems, population growth resulted in general economic progress, as revealed by skeletal stature, increased lifespan, size and quality of housing, drainage, and the use of roof tiles. Hence, Aristotle’s fears were unjustified in real as opposed to ideal cities. The 8th century b.c.e. marked the beginning of “one of the most sustained and rapid improvements in aggregate and per capita consumption known from the premodern world” (Morris 2009: 6; see also Scheidel 2004: 743–57). There is no reason to believe that this phenomenon applied to Greece only. 10.  Plato, Laws 5.740d; 6.784 b–e; Coale 1977: 131–54. 11.  Morris’s cases of fast growth raise a substantial additional question to the hypothesis that explains Jerusalem’s growth during the 7th century b.c.e. by postulating a flood of Israelite refugees coming from the north. For other questions about this model, see Naʾaman 2014.

134

Philippe Guillaume

Given political stability and the favorable climatic conditions that prevailed in the Mediterranean world throughout the Persian period,12 all it took to induce a population explosion was the shift toward increased childbearing reflected in practice in the reduction of girls’ marriageable age and the inclusion of all (or almost all) mature women in the social category of “married wife” by means of polygamy. The diffusion of the ‫ּורבּו‬ ְ ‫ ְפרּו‬among the Yehudite population fostered a shift toward early and lifelong marriage. Its actual efficacy was enhanced by the favorable climatic and sanitary conditions around the Mediterranean of the first millennium b.c.e., which were far better than the conditions prevailing in Northern Europe up to the modern period. By themselves, the climatic conditions increased the average lifespan of women. Since most women died before menopause, each additional year of married life carried an impact on demography. Therefore, the Mediterranean world had a real potential for demographic explosion and, after several centuries, of fast increase; Aristotle recommended population control, and Ben Sira, Job, and some psalmists viewed the increased numbers of daughters to be a dangerous development. Population growth blinded Aristotle and his contemporaries to the consequent increase in wealth since the growth in economic level can only be perceived with the hindsight and comparative data that can be obtained from archaeological studies. I suspect that urbanization was one if not the main cause for the change in attitude toward daughters and large families. Given the size of Jerusalem throughout the Achaemenid era, Genesis, like Hesiod, reflects a rural point of view, while Aristotle wrote about an ideal city rather than actual cities. Ben Sira reflects an urbanized, Ptolemaic Jerusalem or crowded Alexandria (McKechnie 2000: 3–26).13 On the farm, girls as much as boys were necessary hands (see Neh 5:5), but in crowded urban settings, the situation was different and, especially, for girls who needed to be fed but also confined, so they might retain their value on the matrimonial market and spare their fathers shame (Sir 42:11–12). As a late text, Psalm 128 reflects a Jerusalem turned into an urban center for the first time. From that time 12.  Issar and Zahor (2004: 195) state, “[A]fter the spell of cold and humid climate around 1000 b.c.e. came a warm and dry period with a low peak of humidity around 850 b.c.e., then it steadily improved, reaching favorable conditions from ca. 300 b.c.e. until the third century c.e.” 13.  This is the most straightforward explanation for the very close parallels between Sir 38:24–39:11 and the Instruction of Duauf (or Dua-Cheti). English translation is by Parkinson 1997: 273–83. See the discussion in Jäger 2004: 305–17; Sanders 1983: 62–63; and Williams 1981: 1–20, esp. p. 10.

Strategies for the Economic Recovery of Depopulated Yehud

135

on, a sharper distinction is made between the sexes as a consequence of urbanization. To recapitulate, Bronze Age Atrahasis displays a clear worry about overpopulation. By contrast, in the Iron Age, Hesiod thinks that having many sons is the way to improve one’s lot, but after centuries of fast demographic growth, Aristotle sees overpopulation as a threat to his ideal city. Written in a much depleted Palestine, Genesis urges people to multiply.

Child Labor as Motivation for the ‫ּורבּו‬ ְ ‫ ְפרּו‬Slogan Studies of Neolithic skeletons in central Europe show a correlation between the intensification of agriculture and higher fertility rates (Piontek and Vančata 2012: 23–42, esp. p. 34). Neolithic farmers had more children than their ancestors although they ate less meat, and they were subject to more infectious diseases due to a more settled lifestyle. What ensured larger families, despite a poorer diet in proteins and a greater incidence of infectious diseases, was child labor. Palaeolithic hunter-gatherers engaged in less labor-intensive agriculture and thus did not need large families. Consequently, their few children did not toil in the fields. By contrast, the more numerous children of Neolithic farmers worked hard in their parents’ fields from an early age, with consequences that are visible on skeletons: reduced bone growth, reduced size, more bulky shape since child labor stimulates muscle growth while bones are still growing (Piontek and Vančata 2012: 37). The general trend observed during the Neolithic agricultural intensification is relevant to the Palestinian situation because both were times of transition. Between the 6th and the 4th centuries b.c.e., Palestine evolved from the very low demographic and economic levels following the Neo-Babylonian destructions toward a slow but steady demographic recovery and agricultural intensification. After the destruction of the region of Jerusalem and of the Shephelah, Benjaminite farmers turned to more extensive agriculture. With many abandoned villages and most cities destroyed except in Benjamin, the workforce was rarer than ever. Labor-intensive tasks were reduced, and the per capita acreage of plowed land in the 6th century was lower than it was in the 7th century b.c.e. At first, life was easier. The deportation of thousands of Judeans left great tracts of good land empty. The only advice we have from Gedaliah reveals that land was a free-for-all resource: “[G]ather wine and summer fruits and oil, and store them in your vessels, and live in the towns that you have taken over.” And the text adds: “[T]hey gathered wine and summer fruits in great abundance” (Jer 40:10–12).

136

Philippe Guillaume

Abundance in the first years, certainly—but afterwards, the lack of a workforce to maintain existing structures reduced the abundance of the first years. More extensive modes of production afforded survival, thanks to reforestation. Fuel was easier to obtain and the nearby forests provided mushrooms, berries, nuts, snails, and much game. However, bears and lions multiplied as well. Overall, 6th-century b.c.e. farmers were in no better situation than in the previous centuries, except that they sweated less in the fields. All of the sudden, the Achaemenid administration turned its attention to the Levant in the wake of military operations in and against Egypt. Food supply is the first condition of military success, and Achaemenid troops were no exception.14 The movement of troops to and from Egypt required large amounts of food and fodder stored along the routes traveled by the troops. As the costs of transport of basic staples were prohibitive, grain and fodder had to be produced as close as possible to the royal stores established along the Via Maris. Any sudden rise in agricultural demand is a major challenge, more so in the ancient world where general scarcity made supply the only active factor of the market (Ohrenstein 1998: 231). The restoration of temples in Jerusalem and Gerizim was a direct consequence of the Empire’s need to stockpile large amounts of food in the Shephelah. Lachish provided a middle point between Gaza and the next relay farther north. Mizpah and Jerusalem were responsible for the supply of Lachish and Ashdod, which are situated downhill from the Judean highland.15 To feed Persian armies, land left unplowed for a century had to be put back under cultivation. The production of barley, an easy staple to transport and store in the Shephelah, needed to increase dramatically. With plow teams being in greater demand, more bullocks were castrated and trained instead of being eaten as fattened calves. Once plowed and sown, fields needed to be weeded by hand and, when the early summer came, everyone, children and women included, was busy harvesting, threshing, or winnowing. In the fields, young girls are as good as young boys. In fact, they are even better. Besides their work in the fields, girls had the ability to put the ‫ּורבּו‬ ְ ‫ ְפרּו‬slogan in practice if they were married off soon after the appearance of their first periods. The new natalist mood frowned upon spinsters and non-remarried widows. Every woman was expected to 14.  Briant (2002: 372) with a quotation from Xenophon, Anab. 1.5.7, mentions the strategic problems entailed by the depletion of food reserves in the regions that Persian or enemy troops had crossed. 15.  The mention of an Ashdodite dialect in Neh 13:24 presupposes the existence of close relations between Jerusalem and its closest active Philistine neighbor.

Strategies for the Economic Recovery of Depopulated Yehud

137

bear as many children as possible (remember Leah and Rebecca), while their husbands slept at home every night instead of leading their flocks to distant grazing grounds (compare Esau and Jacob). Thanks to polygamy, nonmenopausal widows were married off (compare Naomi and Ruth). The number of children able to work in the fields increased steadily as long as no armies passed by to ravage fields, cut down trees, deplete provisions or deport people. Ten-year-old children can do nearly as much weeding as adults—girls can do as much weeding as boys. Ten-year-olds, girls and boys, can do nearly as much harvesting as adults. Accordingly, agricultural production multiplied, and Persian Yehud moved from a situation where food production just covered the needs of the local population to a position where food surpluses were increasingly available. After the harvest, the tithes had to be delivered to stores in the Shephelah and the rest brought home. To do so, more donkeys had to be bred and trained. As pack animals, donkeys were as precious as cows. The Persepolis Fortification Archive shows that the balance in foodstuffs was usually devoted to acquiring donkeys or mules, besides cows (Henkelman 2005: 149, 156). The special status of donkeys is confirmed by the exchange rate of one sheep for a newborn donkey in Exod 13:13; 34:20.16 If Yehud was as depopulated as surveys suggest, the surplus that could be expected from local farmers was tied to the rate of natural demographic increase. The ‫ּורבּו‬ ְ ‫ ְפרּו‬slogan was thus motivated by economic considerations.

The Seventh-Year Ban on Sowing as Complementary Strategy The potential of the ‫ּורבּו‬ ְ ‫ ְפרּו‬slogan to change mentalities and encourage the early marriage of women and the remarriage of widows was real, but its demographic effects clearly became insufficient once the Persians lost Egypt. At least one decade was necessary before the natural population increase had a noticeable effect on the amount of grain surplus. To support the military campaigns to regain control over Egypt, the Achaemenid administration had to organize immigration to complement the natural demographic increase. Even more hands had to be put in the fields of Yehud. 16.  This does not necessarily confirm Henkelman’s notion that asses and mules were durable goods for which storehouse keepers strove to dispose the remaining grain, wine, or fruit that was their responsibility. Although donkeys certainly have a long life-span (30 years), they are liable to sickness, accident, and theft. Precious metals remain the only commodity that does not spoil, liable only to theft.

138

Philippe Guillaume

Military campaigns were planned years in advance since it was imperative that the extraction of agricultural surpluses did not starve the precarious pool of Yehudite farmers and push them in the arms of the enemy. Darius I had stock-piled supplies several years in advance in preparation for the invasion of Greece (Herodotus 7.1).17 Darius died and the campaign was led by his successor, Xerxes, who first had to quench a rebellion in Egypt (484 b.c.e.). Xerxes’ Egyptian campaign taxed Persian resources to the point that another four years of preparation were necessary before Xerxes could finally set out against Greece, despite the resources Darius had prepared for the European campaign (Briant 2002: 520). A good millennium later, the Mongol Khan Hulagu took five years to assemble his army in Persia and it was not until 1256 c.e. that Hulagu was ready to proceed west to confront the Mamlukes. Starting with Cambyses’ invasion of Egypt (525 b.c.e.), the Southern Levant saw four large Persian armies travel across the Shephelah within a seven-year period: May 525 (Cambyses); Spring 522 (Cambyses or Darius); Autumn 519 (Darius); and Spring 518 (Darius). After this period of heavy military traffic, no large contingent was likely to have traveled via the Shephelah until 484 b.c.e., when Xerxes had to secure his throne by leading an Egyptian campaign early in his reign. That Xerxes’ Egyptian campaign is mentioned as a mere formality in the sources may be due to the natural focus of Greek historians on Persian operations on Greek soil. Yet, Xerxes’ successful campaign in Egypt at a time when he had to affirm his authority can also be taken as the consequence of the improvement of the storage facilities along the Via Maris thanks to the three decades of quiet between Darius’s return from Egypt and Xerxes’ passage in the Shephelah on his way to Egypt. These three decades afforded the necessary time to correct the inadequacies in the supply line that the troops encountered in the initial seven years of intense activity between Cambyses and Darius, whose military endeavors in Egypt were constrained by the lack of adequate storage facilities in the southern Levant. While Herodotus claims that some of Cambyses’ troops were transported by sea, a large contingent passed through the Shephelah around the time of the barley harvest, before it got too hot to travel with heavy equipment and before the Nile started its yearly inundation, which prevented significant deployment of troops between June and October (Ruzicka 2012: 16). After that, Cambyses waited three years before returning home. If the need to set up new Persian-friendly structures in 17.  Herodotus (in Godley 1982: 300–301).

Strategies for the Economic Recovery of Depopulated Yehud

139

Egypt can partly explain why Cambyses stayed so long there, his campaign in Nubia was most likely motivated by logistic reasons. He had to feed his army on fresh ground. An early return through the Via Maris was seemingly out of the question as reserves there had been depleted in the spring of 525 b.c.e. Since too long of a stay in Egypt would have taxed the local reserves and aggravated the Egyptians, going farther south into Nubia was Cambyses’ only option. The return home was envisaged three years later, presumably when enough supplies had been stored along the Via Maris. Or the fact that the return occurred again in the spring, this time in 522 b.c.e., suggests that, despite the three years between the two passages, the Persian Supply Corps relied again mainly on that year’s harvest which was in the process of being collected (at the time when grains were most easily obtained). The puzzle over the fate of Cambyses, who never reached home in that year, adds weight to the idea that things were still not running smoothly in 522.18 Darius’s passage in the autumn of 519 is unique as the other troops traveled in the spring. Darius’s autumn journey presupposes the existence of adequate stores that allowed him to mount a surprise attack in the autumn, when the Egyptians expected him to wait until the next spring. Darius’s return from Egypt a mere six months after his previous passage through the Shephelah can be interpreted as being motivated by the need to travel back no earlier than the time of the next harvest but before the heat of the summer. Additionally, it can be understood that the presence of supply centers made a return possible a mere six months after his Egypt-bound passage. The differences in the timing of the movement of troops between Cambyses and Darius reveal, in part, an improved chain of supply centers along the Via Maris. Since no large expeditionary corps is known to have gone to Egypt until Xerxes’ campaign in 484 b.c.e., the produce extracted from the Shephelah sufficed to cover the needs of the Persian administration without having to organize more costly extraction and transport of agricultural products from the Central Range. The Central Range was of little concern until the first Persian domination over Egypt ended. The Persians left Gaza and its hinterland of the Beer-Sheba valley in Qedarite hands, while the rest of the Philistine coast, from Acco to Ashkelon, were possessions of Tyre or Sidon (Ruzicka 2012: 22–23). Darius’s canal opened a sea route between Mesopotamia and Egypt via the Rea Sea and the Persian Gulf. This direct maritime connection was doubled by the development of a North Sinai route (Graf 1994: 167–89). The amalgamation of the eastern and 18. See Herodotus 3.63–66. See also Balcer (1987) and Walser (1983: 8–18).

140

Philippe Guillaume

western parts of the empire “represented the culmination of a struggle going back 1,000 years and was an achievement unprecedented in the whole of that long time” (Ruzicka 2012: 25). These new routes left the Palestinian highlands in the fringes. The political structure set up by the Neo-Babylonians at Mizpah was sufficient for the needs of the first Persian kings, and it was only after Persia’s first loss of Egypt that the need was felt to invest in Yehud and Samerina. The first loss of Egypt marked a major turning point. The Southern Levant suddenly became a strategic frontier. After three centuries of routine servicing of the relay stations along the southern part of Via Maris, the narrow land bridge had to be widened and turned into a fortress guarding the border. To obtain a substantial increase of foodstuffs from Yehud to stock up the supply center at Lachish, new settlers, whether “returnees” or not, would have to be brought in since natural growth was no longer adequate to the urgency of the new situation. As the Hesiodic principle discussed above postulates that more hands equals more food, the number of new settlers could be calibrated to meet the amount of food required to provision the expected size of the force that would have to be deployed for the upcoming Egyptian campaigns. The arrival of people by the thousands is mentioned in the books of Ezra and Nehemiah, which ignore the logistics involved. Given the low yields of ancient agriculture, even bumper harvests could not absorb the consumption shock resulting from the arrival of large groups of immigrants if they were to be fed out of the stores of the locals. Although Neh 2:7 has the travel expenses of the newcomers drawn from royal stores en route, the biblical texts ignore the larger question of how they would be fed once in Yehud while waiting to gather their first harvest. The book of Ruth depicts the crisis provoked by the arrival of only two empty-handed widows, a crisis that required the attention of the entire village community and involved delicate negotiations before deciding how and who would shoulder the extra burden. A passage in Leviticus, often viewed as utopian, could in fact reflect a practical response to the logistic challenge of organizing the arrival and settlement of a significant number of new farmers in Yehud within a single year. Somewhat paradoxically, one of the most impressive measures implemented to revive the agrarian economy in depopulated Yehud would have been the ban on sowing on the seventh year (Lev 25:3–7). Such a ban is not equivalent to a fallow cycle because it involves a general and province-wide measure imposed on every field at the same time, rather than in some rotation. Freeing all farmers from labor-intensive harvesting was meant to deploy them into work gangs

Strategies for the Economic Recovery of Depopulated Yehud

141

for major land reclamation projects over an entire year rather than for a few months at a time between harvest and sowing times (Guillaume 2012: 234–39). A total ban on sowing for the entire population of a province makes little sense if the aim is to stockpile food in anticipation of a major military campaign in the coming years, unless the ban resulted in a major increase of arable land ready to be sown in the following year. The arrival of new settlers could be coordinated with the prescriptions of the so-called sabbatical year. As is clear from Lev 25:4, the Sabbath of the seventh year was a Sabbath for the land, not for the farmers. The seventhyear ban on sowing would have been a drastic and efficient measure to increase the surface of land ready for cultivation. Having lain fallow for a century, the terra rossa was very fertile and would produce high yields once cleared from the bushes that took over when ploughing and intensive grazing were discontinued.19 The corvée to which farmers were regularly submitted would have supplied some workdays toward the clearance of fresh land, but corvée duties had to occur during times of low agricultural activity. Freeing the workforce available locally from the yearly routine of sowing and harvesting would have doubled or tripled the output within a few years, depending on the number of immigrants, a number that had to be calculated according to the amount of land the locals could clear during the seventh year. The surface gained during an entire sabbatical year exceeded the capacity of the local population to cultivate it profitably the following years. The arrival of a wave of new settlers had to coincide with the end of the seventh year, when the freshly cleared tracts of land were ready to be allotted to the new comers and sown immediately in order to reduce as much as possible the interval between their arrival and their first harvest. Whether or not the date given in Ezra 7:7 for the arrival of Ezra on the seventh year of Artaxerxes is historically accurate or not, that it occurs on a seventh year may not be a coincidence. I take it as a hint that groups of new settlers were not sent off from other parts of the empire in a haphazard way. Their arrival was coordinated with the completion of vast land clearing operations in Yehud when gangs of local farmers were about to go back to their own fields to sow the harvest of the eighth year (see table 1). How the local population that missed a harvest could survive and feed the new immigrants during the five to six months between the 19.  Yields from virgin steppe land reach 20 to 30 per unit sown (compared to 3–5 in the Syrian western plains) without any fertilizers. Given good rain levels (3 years out of 10 in the Syrian Jazira, more in central Palestine), they can even reach the impressive figure of 60 or even 100 per unit (Métral, 2000: 130 n. 14).

142

Philippe Guillaume Table 1 Year

VI

VII

Month

May–June

December

May–June

Summer

December

Winter

Locals

harvest

sow

harvest

clear land

clear land no sowing

clear land

Immigrants

Year (cont.)

VIII

IX

Month

May–June

Summer

December

May–June

Locals

harvest aftergrowth

plough old and new fields

sow old fields

harvest old fields

Immigrants

Early arrivals

Late arrivals

sow new fields

harvest new fields

sowing time and the harvest of the ninth year must be answered in a more realistic way than Lev 25:21–22 does. That Yhwh would bless the sixth year so much that its surplus would be enough until the ninth year is a theological device indicating that no taxes were raised on the produce of the sixth year in order not to deplete the reserves that had to last two years and to ensure that all surpluses were stored locally. As the seventh-year ban in Lev 25:4 only applies to the most labor intensive activities (harvesting grain and grapes), foodstuffs other than grain and wine were produced as usual. Contrary to Lev 25:5, which forbids reaping the grain that grew on its own and gathering the grapes that grew on the unpruned vines, vv. 6–7 indicate that humans and animals would feed themselves on the after-growths and wild products. These were plentiful given the low human population levels. Moreover, women tended their gardens as usual and thus continued to produce vegetables, herbs, and fruits. Contrary to Exod 23:11, Lev 25:4 does not include olive trees in the banned harvesting activities. Hence, the production of olives, almonds, pomegranates, pistachios, figs, carobs, vegetables, and honey continued as usual (Schenker 2000: 100). Today, olive trees are thoroughly trimmed every three to five years, with minor yearly pruning of dead wood and suckers. Therefore, the yield of olives and almonds would hardly be affected during the sabbatical year if the trees were tended the previous year in anticipation. People could survive without grapes, while the stored grain surplus of previous years made up for some or all of the missed harvest, depending on the yield of the previous year. If needed, additional supplies

Strategies for the Economic Recovery of Depopulated Yehud

143

could be brought from neighboring provinces in exchange for the meat, leather, wool, and cheese produced in greater quantities, since the ban led farmers to rely even more on their livestock. Reduced culling in anticipation of the ban left larger herds, which were pastured farther away from the villages than was the case in normal years. Greater numbers of goats contributed actively to the clearance of fresh land during the year of the ban. Behind the flocks, the workforce freed from labor-intensive plowing and harvesting cleared stones, repaired terraces, walls, wells, and irrigation canals. Seen in this way, the ban on sowing in the seventh year would have been a most pragmatic measure to organize the arrival of thousands of immigrants, ensuring a speedy installation and the capacity to deliver surpluses already from their first harvest. It is not possible to know whether waves of immigrants arrived every seven years, or whether this ban was a one-off operation that was turned into a “law” in the biblical text. But we do know from administrative archives that the center of the Empire had the kind of bureaucracy that was necessary to implement such a vast operation (Briant 2002: 425–71). Before the 7th century b.c.e., the Neo-Assyrians had mastered the logistics for the transfer of large populations (Oded 2000: 91–103). The Neo-Babylonians continued the “land-for-service” scheme by which groups of individuals, often non-Babylonians, were settled on recently reclaimed land or in underdeveloped areas considered strategic by the central administration and earmarked as priority zones. In this context, groups of Judeans found themselves farming southeast of Babylon and north of Uruk, an area that was devastated by the Neo-Assyrian campaigns against Elam during the 8th and 7th centuries b.c.e. The Persians did not alter the situation, since the region was crucial to them, as it serviced main routes linking Babylonia and Persia (Wunsch 2013). Hence, the descendants of the Judean deportees did not all return to Yehud, and the late Achaemenid Murāšû archive shows that Judeans were still active in the area. Unless the demographic level of Judean settlements in Babylonia was sufficient to send the required number of families to Yehud without endangering the viability of their colonies in Babylonia, immigrants of non-Judean descent had to be brought in from other recently conquered regions. In any case, the costs involved in the transfer of large numbers of peasants rendered any measure likely to bolster the natural increase of the depleted Yehudite population highly desirable from the point of view of the Empire as well as the Yehudites themselves. It was only in the face of mortal danger such as the involvement of Greek cities in Egypt that the Empire reverted to costly population transfers.

144

Philippe Guillaume

The local farmers resented any land-grant of the fields they had just cleared, especially if the beneficiaries were foreigners who were not even distant cousins. Hence, tax pressure in itself would have encouraged farmers to raise larger families, but in the biblical text, the motivation for demographic increase is turned into faithfulness to the divine order to multiply, as discussed above. Deeper resentment would have been felt if the land cleared during the sabbatical year was declared royal land (hanšû, literally, “fifties”) and thus benefited from reduced tithe on the produce of the new fields, from the standard one-tenth to one-thirtieth, as attested by contemporary sources from Babylonia (Jursa 1998: 11; Bongenaar 1997: 106). The newly reclaimed land would have been situated primarily near Jerusalem. If the Mesopotamian system was applied in Yehud, each unit was attributed to ten households, which formed a tax unit collectively responsible for the payment of the relevant taxes in kind and labor obligations.20 Despite the resentment arising from the new situation, the arrival of newcomers represented fresh openings for the most enterprising people by increasing the critical mass of tradable commodities within Yehud. The newcomers shared the tax burden, in particular the service obligations, because the primary aim of the tax system was the mobilization of the workforce (Jursa 2011: 443). As service obligations were often provided through substitutes, any increase in the overall population presented new opportunities to large families with a surplus of workers. Some of their spare hands could be hired out to tax units for those who wished to pay a substitute for the service they were expected to provide. Despite the unpopularity of any kind of taxation, the combined implementation of a natalist theology and of the settlement of families from elsewhere fulfilled the expectations of the Achaemenid administration, or it is unlikely that the Hebrew Bible would have become the founding document of the various forms of Judaism. Both strategies secured greater levels of safety and prosperity. More hands meant more work done. More work translated into more cleared land and more food for everyone, for Persian soldiers as much as for the peasants who fed them. 20. On hanšû land, see Jursa 2011: 440; and Kleber 2008: 115.

Strategies for the Economic Recovery of Depopulated Yehud

145

Bibliography Assante, J. 2003 From Whores to Hierodules: The Historiographic Invention of Mesopotamian Female Sex Professionals. Pp. 13–47 in Ancient Art and Its Historiography, ed. A. A. Donohue and M. D. Fullerton. Cambridge: Cambridge University Press. Balcer, J. M. 1987 Herodotus and Bisitun: Problems in Ancient Persian Historiography. Stuttgart: Franz Steiner. Ben Zvi, E. 2011 Social Memory and Identity Formation. Pp. 95–148 in Texts, Contexts and Readings in Postexilic Literature, ed. L. Jonker. Tübingen: Mohr Siebeck. Briant, P. 2002 From Cyrus to Alexander: A History of the Persian Empire. Winona Lake, IN; Eisenbrauns. Bongenaar, A. C. V. M. 1997 The Neo-Babylonian Ebabbar Temple at Sippar. Leiden: Brill. Carter, C. E. 1999 The Emergence of Yehud in the Persian Period. Sheffield: Sheffield Academic Press. Coale, A. 1977 The Development of New Models of Nuptiality and Fertility. Population 32: 131–54. Edelman, D. V. 2005 The Origins of the ‘Second’ Temple: Persian Imperial Policy and the Rebuilding of Jerusalem. London: Equinox. Edwards, A. T. 2004 Hesoid’s Ascra. Berkeley: University of California Press. Evelyn-White, H. G. 1964 Hesiod. LCL 57. London: Heinemann. Faust, A. 2007 Settlement Dynamics and Demographic Fluctuations in Judah. Pp. 23–51 in A Time of Change, ed. Y. Levin. London: T. & T. Clark. Finkelstein, I., and Singer-Avitz, L. 2009 Reevaluating Bethel. ZDPV 125: 33–48. Godley, A. D. 1982 Herodotus. LCL 119. Cambridge: Harvard University Press / London: Heinemann. Graf, D. F. 1994 The Persian Royal Road System. Achaemenid History 8: 167–89. Guillaume, P. 2012 Land, Credit and Crisis: Agrarian Finance in the Hebrew Bible. Sheffield: Equinox.

146

Philippe Guillaume

Hallo, W. W., and Younger, K. L., Jr., eds. 1997 The Context of Scripture: Canonical Compositions from the Biblical World. 3 vols. Leiden: Brill. Henkelman, W. F. M. 2005 Animal Sacrifice and “External” Exchange in the Persepolis Fortification Tablets. Pp. 137–65 in Approaching the Babylonian Economy, ed. H. D. Baker and M. Jursa. Münster: Ugarit-Verlag. Ilan, D. 2011 Household Gleanings from Iron I Tel Dan. Pp.  133–54 in Household Archaeology in Ancient Israel and Beyond. Leiden: Brill. Issar, A. S., and Zahor, M. 2004 Climate Change: Environment and Civilization in the Middle East. Berlin: Springer. Jäger, S. 2004 Altägyptische Berufstypologien. Göttingen: Seminar für Ägyptologie und Koptologie. Jursa, M. 1998 Der Tempelzehnt in Babylonien. Münster: Ugarit-Verlag. 2004 Accounting in Neo-Babylonian Institutional Archives: Structure, Usage and Implications. Pp.  145–98 in Creating Economic Order, ed. M. Hudson and C. Wunsch. Besthesda, MD: CDL. 2011 Taxation and Service Obligations in Babylonia from Nebuchadnezzar to Darius and the Evidence for Darius’ Tax Reform. Pp. 431–48 in Herodot und das Persiche Weltreich, ed., R. Rollinger, B. Truschnegg, and R. Bichler. Wiesbaden: Harrassowitz. Kleber, K. 2008 Tempel und Palast: Die Beziehungen zwischen dem König und dem EannaTempel im spätbabylonischen Uruk. Münster: Ugarit-Verlag. Kilmer, A. D. 1972 The Mesopotamian Concept of Overpopulation and Its Solution as Reflected in the Mythology. Orientalia 41: 160–77. Leuchter, M. 2010 The Politics of Ritual Rhetoric: A Proposed Sociopolitical Context for the Redaction of Leviticus 1–16. VT 60: 345–65. Liphschitz, N. 2008 Reconstruction of the Arboreal Vegetation of the Judean and Samarian Mountains during Antiquity as Evident from Archaeological Wood Remains. Pp. 48–55 in In the Hill-Country, and in the Shephelah, and in the Arabah, ed. S. Bar. Jerusalem: Ariel. Lipschits, O. 2004 The Rural Settlement in Judah in the Sixth Century bce: A Rejoinder. PEQ 136: 99–108. 2005 The Fall and Rise of Jerusalem. Winona Lake, IN: Eisenbrauns. 2011 Persian-Period Yehud. Pp. 187–211 in Texts, Contexts and Readings in Postexilic Literature, ed. L. Jonker. Tübingen: Mohr Siebeck.

Strategies for the Economic Recovery of Depopulated Yehud

147

Lowry, S. T. 1979 Recent Literature on Ancient Greek Economic Thought. Journal of Economic Literature 17: 65–86. 1998 Introduction. Pp.  1–10 in Ancient and Medieval Economic Ideas and Concepts of Social Justice, ed. S. T. Lowry and B. Gordon. Leiden: Brill. Luquet, G. H. 1926 L’art néo-calédonien: Documents recueillis par Marius Archambault. Paris: Institut d’ethnologie. Malthus, T. R. 1999 An Essay on the Principle of Population. Oxford: Oxford University Press. McKechnie, P. 2000 The Career of Joshua Ben Sira. JTS 51: 3–26. Métral, F. 2000 Managing Risk: Sheep-Rearing and Agriculture. Pp.  123–44 in The Transformation of Nomadic Society in the Arab East, ed. M. Mundy and B. Musallam. Cambridge: Cambridge University Press. Morris, I. 2004 Economic Growth in Ancient Greece. Journal of Institutional and Theoretical Economics 160: 709–42. 2006 The Growth of Greek Cities in the First Millennium bc. Pp. 27–43 in Urbanism in the Pre-industrial World, ed. G. R. Storey. Tuscaloosa, AL: Alabama University Press. 2009 The Eighth Century Revolution. Pp.  64–80 in A Companion to Archaic Greece, ed. K. A. Raaflaub and H. van Wees. Malden, MA: Wiley-Blackwell. Most, G. W. 2006 Hesiod: Theogony, Works and Days, Testimonia. LCL 57. London: Heinemann / Cambridge: Harvard University Press. Naʾaman, N. 2014 Dismissing the Myth of a Flood of Israelite Refugees in the Late Eighth Century bce. ZAW 126: 1–14. Nelson, S. 1996 The Drama of Hesiod’s Farm. Classical Philology 91: 45–53. Oded, B. 2000 The Settlements of the Israelite and Judean Exiles in Mesopotamia in the 8th–6th Centuries bce. Pp. 91–103 in Studies in Historical Geography and Biblical Historiography: Presented to Zechariah Kallai, ed. G. Galil and M. Weinfeld. Leiden: Brill. Ohrenstein, R. A. 1998 Talmud and Talmudic Tradition: A Socio-Economic Perspective. Pp. 209–68 in Ancient and Medieval Economic Ideas and Concepts of Social Justice, ed. S. T. Lowry and B. Gordon. Leiden: Brill. Parkinson, R. B. 1997 The Tale of Sinuhe and Other Ancient Egyptian Poems 1940–1640 bc. Oxford: Clarendon.

148

Philippe Guillaume

Piontek, J., and Vančata, V. 2012 Transition to Agriculture in Central Europe: Body Size and Body Shape amongst the First Farmers. Interdisciplinaria archaeologica 3: 23–42. Polanyi, K. 1944 The Great Transformation. New York: Farrar and Rinehart. 1957 Aristotle Discovers the Economy. Pp. 64–97 in Trade and Market in the Early Empires: Economies in History and Theory, ed. K. Polanyi, C. M. Arensberg, and H. W. Pearson. New York: Free Press. Reich, R. 1992 The Beth-Zur Citadel II: A Persian Residency? ErIsr 23: 247–52. [Modern Hebrew] Ruzicka, S. 2012 Trouble in the West: Egypt and the Persian Empire, 525–332 bce. Oxford: Oxford University Press. Sanders, J. T. 1983 Ben Sira and Demotic Wisdom. Chico, CA: Scholars Press. Scheidel, W. 2004 Demographic and Economic Development in the Ancient Mediterranean World. Journal of Institutional and Theoretical Economics 160: 743–57. Schellenberg, A. 2009 Humankind as the “Image” of God. TZ 65: 97–115. Schenker, A. 2000 Der Boden und seine Produktivitat im Sabbat und Jubeljahr. Pp. 123– 33 in Recht und Kult im Alten Testament, ed. A. Schenker. Fribourg: University Press / Gottingen: Vandenhoeck & Ruprecht. Stolper, M. W. 2001 Fifth Century Nippur: Texts of the Murašus and from Their Surroundings. JCS 53: 83–132. 2003 “No-One Has Exact Information except You”: Communication between Babylon and Uruk in the First Achaemenid Reigns. Pp. 265–87 in A Persian Perspective, ed. W. Henkelmann and A. Kuhrt. Leiden: Brill. Tandy, D. W., and Neale, W. C. 1996 Hesiod’s Works and Days. Berkeley: University of California Press. Ucko, P., and Rosenfeld, A. 1967 Paleolithic Cave Art. World University Library. New York: McGrawHill. Walser, G. 1983 Der Tod des Kambyses. Historia-Einzelschriften 40: 8–18. Walton, J. H. 2009 Genesis 1 as Ancient Cosmology. Winona Lake, IN: Eisenbrauns. Wittenberg, G. 2010 In Search of the Right Metaphor: A Response to Peet van Dyk’s “Challenges in the Search for an Ecotheology.’’ OTE 23: 427–53.

Strategies for the Economic Recovery of Depopulated Yehud

149

White, L., Jr. 1967 The Historical Roots of Our Ecological Crisis. Science 155: 1203–7. White, R. 2003 Prehistoric Art: The Symbolic Journey of Humankind. New York: Abrams. Williams, R. T. 1981 The Sages of Ancient Egypt in the Light of Recent Scholarship. JAOS 101: 1–20. Wood, B. G. 2008 The Search for Joshua’s Ai. Pp. 205–40 in Critical Issues in Early Israelite History, ed. R. S. Hess, G. A. Klingbeil, and P. J. Ray Jr. Winona Lake, IN: Eisenbrauns. Wunsch, C. 2013 Glimpses on the Lives of Deportees in Rural Babylonia. Pp 247–60 in Arameans, Chaldeans, and Arabs in Babylonia and Palestine in the First Millennium b.c., ed. A. Berlejung and M. P. Streck. Leipziger Altorientalische Studien 3. Wiesbaden: Harrassowitz. Zevit, Z. 2007 Text Traditions, Archaeology, and Anthropology: Uncertainties in Determining the Populations of Judah and Yehud from ca. 734 to ca. 400 bce. Pp. 436–43 in Up to the Gates of Ekron, ed. S. W. Crawford. Jerusalem: Israel Exploration Society.

Exploitation of Depopulated Land in Achaemenid Judah Lisbeth S. Fried University of Michigan

The use and exploitation of depopulated land by the Achaemenids have been studied for over a century now, and the topic is well known. The Greek historians accurately described the effect of Achaemenid rule. Xenophon, for example, records how Cyrus distributed land and residences to his satraps and governors who were stationed abroad as well as to those retainers whom he sent abroad only occasionally. According to Xenophon, Cyrus wanted them to “have lands and houses there [in their provinces]; tribute paid to them here [in Persia from those lands] and, whenever they go there [to their provinces], to be able to lodge in residences of their own” (Cyr. 8.6.4). Xenophon reports further that Cyrus gave his friends “houses and servants in the various states which he had subdued. Even to this day (Xenophon continues) those properties, some in one land, some in another, continue in the possession of the descendants of those who received them then” (Cyr. 8.6.5). Xenophon tells us, for example, of the vast amounts of land that a provincial governor had in the years around 400 b.c.e. The personal wealth of Zenis of Dardanus, a provincial governor of the satrap Phar­ nabazus, was enough to provide nearly a year’s pay to an army of 8,000 men! (Hell. 3.1.27–28) But we do not need to accept the word of the Greek writers. The Murašu archive yields tangible evidence of manor estates in Babylon that had been taken over by the Persian king, the crown prince, and other members of the Achaemenid royal family and their retainers. Aramaic documents from Egypt and Bactria also reveal the estates there that were owned by members of the royal family and from which they received tribute. I need only quote the words of Dandamaev (1967: 37–42) written almost half a century ago: [A]fter the conquest of Babylon by the Persians, great changes took place in this area in agricultural relations. Part of the land, and to be sure, the most fruitful part, was transferred into the actual ownership of the king. Furthermore, the Achaemenids distributed to the members of the royal house, their friends, agents of the Persian nobility, high officials and to

151

152

Lisbeth S. Fried

all those persons who had shown great service to the Persian king, land that had been removed [i.e., confiscated] from their subject populations.1

Dandamaev (1974: 123–27) writes that, as a result of Darius’ administrative and fiscal reforms after 518 bc, the people of Egypt, Syria, Babylonia, Asia Minor and probably other countries of the Achaemenid Empire were deprived of their own lands. These lands were distributed in large estates as full and hereditary property of the members of the royal family, representatives of the Persian nobility, and high-ranking officials.

This would have taken place in Judah as well. Besides these manor estates, Stolper and others before him distinguish an additional type of land exploitation scheme called the ḫadru. Although this type of land-for-service agreement goes back at least to the Ur III period in Babylon, the word ḫadru was apparently not used outside the Murašu archive and the area around Nippur. These ḫadrus were blocks of land granted not to a member of the royal family or one of its retainers but to a corporate group of feudatories, usually agnatic relatives (Cardasçia 1951; Ries 1976; Stolper 1985). Each ḫadru was headed by a foreman who was responsible for the allocation of fiefs or shares among members, for assuring that the land was productive, for the collection of taxes, and for the feudatories’ performance of the military or service obligations incumbent on them—that is, the ilku (Stolper 1985: 70). These ḫadrus were thus a means for ensuring cultivation of land and for providing both a military reserve and cadres of state-controlled workers. The large use of foreigners on these lands also enabled the integration of deportees into the economy (van Driel 2002: 227–28). Jursa (2010: 406) suggests that land around Nippur had been “singularly depopulated” and that in the 6th and early 5th centuries Nippur was “rather isolated and self-contained and detached from the main axis of communication along the Euphrates River,” uniquely enabling foreign workers to be settled in the area and assigned land in exchange for service. This characteristic of Nippur is similar to 6th-century Judah, which was also depopulated, also enabling large number of newcomers—in this case, Judean exiles—to be settled in Judah in a similar land-for-service scheme. In fact, archaeological evidence suggests that those settl­ ing in rural sites in Persian period Judah did not settle in places that had been inhabited before (Faust 2012: 56). Continuity of rural settlements between the Iron Age and the Persian period was negligible. Of 45 excavated late Iron Age farmsteads in Judah and Benjamin, only 7 1.  See also Briant 1985: 53–71; and Stolper 1985.

Exploitation of Depopulated Land in Achaemenid Judah

153

showed limited habitation in the Persian period, and this was probably due simply to coincidence. Most Persian period rural sites did not exist in the Iron Age. The unavoidable conclusion is that farmers in Persian period Judah did not go back to homesteads from which their families had been deported (Hoglund 1991: 54–72). This suggests either a random return to locations suitable for farming or that Persian officials had assigned suitable land to the returnees irrespective of their place of origin. With this background in mind, let us turn to Nehemiah 5, which may provide a key to land use in Judah under the Achaemenids.

Was There a Debt Crisis in 5th-Century Judah (Nehemiah 5:1–5)? Neh 5:1–5 describes a dire situation during the governorship of Nehemiah: ‫ֲׁשר א ְֹמִרים ָּבנֵינּו‬ ֶ ‫ְוי ֵׁש א‬2 ‫ְהּודים׃‬ ִ ‫אחֵיהֶם ַהּי‬ ֲ ‫ו ְַּת ִהי ַצעֲקַת ָהעָם ּו ְנׁשֵיהֶם ְּגדֹולָה אֶל־‬1 3 ‫ּוכָרמֵינּו‬ ְ ‫ֲׁשר א ְֹמִרים ְׂשדֹתֵ ינּו‬ ֶ ‫ִחיֶה׃ ְוי ֵׁש א‬ ְ ‫אכלָה ְונ‬ ְ ֹ ‫ִקחָה ָדגָן ְונ‬ ְ ‫אנ ְַחנּו רַ ִּבים ְונ‬ ֲ ‫ּובנֹתֵ ינּו‬ ְ ְ ‫ָוינּו ֶכסֶף ְל ִמּדַ ת ַה ֶּמל‬ ‫ֶך‬ ִ‫ֲׁשר א ְֹמִרים ל‬ ֶ ‫ְוי ֵׁש א‬4 ‫ָרעָב׃‬ ָ‫ִקחָה ָדגָן ּב‬ ְ ‫אנ ְַחנּו ע ְֹר ִבים ְונ‬ ֲ ‫ּובָּתֵ ינּו‬ ‫ׁשים‬ ִ ‫אנ ְַחנּו כ ְֹב‬ ֲ ‫ׂשרֵ נּו ִּכ ְבנֵיהֶם ָּבנֵינּו ְו ִהּנֵה‬ ָ ‫ְוע ַָּתה ִּכ ְבׂשַ ר ַאחֵינּו ְּב‬5 ‫ּוכָרמֵינּו׃‬ ְ ‫ְׂשדֹתֵ ינּו‬ ‫ּוׂשדֹתֵ ינּו‬ ְ ‫ִכּבָׁשֹות ְואֵין ְלאֵל יָדֵ נּו‬ ְ ‫ָדים ְוי ֵׁש ִמ ְּבנֹתֵ ינּו נ‬ ִ ‫ֶת־ּבנֹתֵ ינּו ַלעֲב‬ ְ ‫אֶת־ ָּבנֵינּו ְוא‬ ‫אחִֵרים׃‬ ֲ ‫ּוכָרמֵינּו ַל‬ ְ Now there was a great outcry of the people and of their wives against their Jewish kin. 2For there were those who said, “With our sons and our daughters, we are many; we must get grain, so that we may eat and stay alive.” 3There were also those who said, “We are having to pledge our fields, our vineyards, and our houses in order to get grain during the famine.” 4And there were those who said, “We are having to borrow money on our fields and vineyards to pay the king’s tax. 5Now our flesh is the same as that of our kindred; our children are the same as their children; and yet we are forcing our sons and daughters to be slaves, and some of our daughters have been ravished; we are powerless, and our fields and vineyards now belong to others.” (Neh 5:1–5 nrsv) 1

The passage seems to describe a debt crisis, but Philippe Guillaume (2010) asserts that no such debt crisis ever existed historically and that it exists only on paper, both in the Hebrew text and in the various studies of the text. According to Guillaume, since all three groups cited in Neh 5:1–5 have fields, vineyards, and houses, they are neither poor nor destitute; they simply lack credit. Further, he asserts that since they have assets with which to secure loans, and since there is no indication that these loans were not forthcoming, and since there is no suggestion of foreclosures or of their credit drying up, there was no crisis.

154

Lisbeth S. Fried

Translation Difficulties Before discussing the possible existence of a debt crisis and the use of the land that is revealed in the passage, we need to clear up numerous translation difficulties. In v. 1, for example, the preposition ‫ אל‬following ‫ ַצעֲקַת‬should be translated “a cry to [our brothers, the Judeans],” as it is translated everywhere else in the Hebrew Bible, not “against [our brothers]” (Wright 2004: 181–82 n. 49). For example, Yhwh says, “[T]he cry of the people Israel has come to me,” ‫ִׂשָראֵל ָּבאָה ֵאלָי‬ ְ ‫( ַצעֲקַת ְּבנֵי־י‬Exod 3:9), not “against me.” More difficult is v. 2, where the word “we,” ‫אנ ְַחנּו‬ ֲ , appears strangely out of place since it is not connected to a verb. ‫ִחיֶה‬ ְ ‫אכלָה ְונ‬ ְ ֹ ‫ִקחָה ָדגָן ְונ‬ ְ ‫אנ ְַחנּו רַ ִּבים ְונ‬ ֲ ‫ּובנֹתֵ ינּו‬ ְ ‫ֲׁשר א ְֹמִרים ָּבנֵינּו‬ ֶ ‫ְוי ֵׁש א‬ The Greek reads, “With our sons and daughters we are many,” and that is how it is translated by some commentators and also in the nrsv (Myers 1965: 126, but see the note on v. 2; Williamson 1985: 231–32 note a; Wright 2004: 188). Most commentators emend the word ‫“ רַ ִּבים‬many,” to ‫ע ְֹר ִבים‬, “mortgaging” with the Vulgate and so read “we are mortgaging our sons and daughters so that we may get grain and eat and live” (Rudolph 1949: 125; Clines 1984: 166; Gunneweg 1987: 84; Blenkinsopp 1988: 252–53; Albertz 1994: 494; Reinmuth 2002: 116–17). However, this emendation is devoid of textual support. Commentators who emend the text see v. 3 as an expansion and repetition of v. 2. As will be clarified below, the emendation is not necessary. Verse 3 does not seem to present problems to exegetes. It contains the verb ‫עָרַ ב‬, “to mortgage or pledge”: ‫ָרעָב׃‬ ָ‫ִקחָה ָדגָן ּב‬ ְ ‫אנ ְַחנּו ע ְֹר ִבים ְונ‬ ֲ ‫ּוכָרמֵינּו ּובָּתֵ ינּו‬ ְ ‫ֲׁשר א ְֹמִרים ְׂשדֹתֵ ינּו‬ ֶ ‫ְוי ֵׁש א‬ There are also those who say, “Our fields, our vineyards, and our houses we are pledging so that we may get grain in the famine.” The verb ‫ עָרַ ב‬occurs 17 times in the Hebrew Bible, and each time, it refers to something offered as a guarantee. So, as Guillaume (2010: 4) points out, the people are not selling their fields but mortgaging them. As Cardasçia (1951: 37) emphasizes in his discussion of the Murašu documents, these are always antichretic loans, since property itself was considered inalienable. He points to the stipulations in Hammurabi’s Code, e.g., LH §§49–52: If a man borrows silver from a merchant and gives the merchant a field prepared for planting with either grain or sesame and declares to him, “You cultivate the field and collect and take away as much grain or sesame as will be grown”—if the cultivator should produce either grain or sesame in the field, at the harvest it is only the [original] owner of the

Exploitation of Depopulated Land in Achaemenid Judah

155

field who shall take the grain or sesame that is grown in the field, and he shall give to the merchant the grain equivalent to his silver which he borrowed from the merchant and the interest on it and also the expenses of the cultivation (Roth 1997).

Even though the text reads “if a man gives the merchant a field,” the field itself is not actually turned over to the merchant but only the portion of the produce equal to the amount borrowed. The emphasis on the original owner of the field counting out the grain or sesame and delivering it to the merchant clarifies that the land itself was not handed over but was held as a guarantee. The primary difficulty for exegetes appears in v. 4. ְ ‫ָוינּו ֶכסֶף ְל ִמּדַ ת ַה ֶּמל‬ ‫ּוכָרמֵינּו׃‬ ְ ‫ֶך ְׂשדֹתֵ ינּו‬ ִ‫ֲׁשר א ְֹמִרים ל‬ ֶ ‫ְוי ֵׁש א‬ Nearly every commentator translates it as the nrsv does: “We have had to borrow money on our fields and vineyards to pay the king’s tax,” inserting an ‫על‬, “on,” before “our fields and vineyards” (Rudolph 1949: 128; Clines 1984: 167; Williamson 1985: 131–32; Blenkinsopp 1988: 253; Albertz 1994: 495; Reinmuth 2002: 116; Wright 2004: 188). Gunneweg (187: 84), followed by Guillaume (2010: 4), translate it differently, however. They maintain that the phrase “fields and vineyards” is an addition that has no connection to the rest of the verse. Both of them leave the phrase off and translate simply: “We must borrow money for the king’s tax.” Taxes or Rents? ְ ‫מידַ ּת ַה ֶּמל‬. Neh 5:4 employs the phrase ‫ֶך‬ ִ All commentators translate this as “the king’s tax.” Is this the correct translation? The word ‫ ִמידַ ּת‬or actually ‫ְּדה‬ ָ‫ ִמנ‬appears in Aramaic contexts in Ezra 4:13 and 7:24, both ְ ‫ ַהל‬. times in connection with ‫ ְבלֹו‬and ‫ָך‬ ‫ה־בלֹו‬ ְ ‫ְּד‬ ָ‫ִׁשּתַ ְכ ְללּון ִמנ‬ ְ ‫ְתא ָד ְך ִּת ְת ְּבנֵא ְוׁשּורַ ּיָה י‬ ָ ‫ְדי ַע ֶלהֱוֵא ְלמ ְַלּכָא ִּדי הֵן ִק ְרי‬ ִ ‫ּכעַן י‬ ְ ‫הל‬ ‫ְּתנּון ְוא ְַּפתֹם מ ְַל ִכים ְּת ַה ְנזִק׃‬ ְ ‫ָך לָא ִינ‬ ֲ ‫ַו‬ Now may it be known to the king that if this city is rebuilt and the walls finished, they will not pay tribute, custom, or toll, and the royal revenue will be reduced. (Ezra 4:13 nrsv) ‫א ָלהָא ְדנָה‬ ֱ ‫ָלחֵי ּבֵית‬ ְ ‫ְתינַּיָא ּופ‬ ִ ‫הנַּיָא ְו ֵלוָי ֵא זַּמָרַ ּיָא ָתָר ַעּיָא נ‬ ֲ ‫הֹוד ִעין ִּדי כָל־ ָּכ‬ ְ ‫ּולכֹם ְמ‬ ְ ְ ‫הל‬ ‫ָך לָא ׁשַ ּלִיט ְל ִמ ְרמֵא עֲלֵיהֹם׃‬ ֲ ‫ְּדה ְבלֹו ַו‬ ָ‫ִמנ‬ We also notify you that it shall not be lawful to impose tribute, custom, or toll on any of the priests, the Levites, the singers, the doorkeepers, the temple servants, or other servants of this house of God. (Ezra 7:24 nrsv)

156

Lisbeth S. Fried

The nrsv translates these three words “tribute, custom, and toll,” indicating no real idea of what the terms mean beyond something that must be paid to the king. ְ ‫ = ַהל‬Ilku ‫ָך‬ It is clear from Ezra 7:24 that, before Ezra’s arrival, temple personnel had to pay these three taxes, but at his arrival he exempted them. It is also clear from Ezra 4:3 that, at least prior to Nehemiah’s building Jerusalem’s city wall, and undoubtedly after as well, the people of Judah were responsible for all of these burdens. ְ ‫ ַהל‬. We know what this is. It is well known that The first of these is ‫ָך‬ the term is derived from the Akkadian ilku and that ilku refers to service, military or, otherwise, that is owed on land given as a fief by a king or governor (Hoftijzer and Jongeling 1995: 283; Rosenthal 1995: #188; van Driel 2002: 254–59; Naveh, and Shaked 2012: 30; CAD I–J 73). Often the ilku is remunerated in kind or in silver so that someone else can be paid to perform the ilku service, instead of the fief-holder. A good example of ilku service appears in a text drawn from the Murašu archive. The text concerns an agreement in which Gedaliah promises to fulfill the ilku service that his father owes on the horse-land (bît sîsî) that his father had held, and Rîmût-Ninurta, son of Murašu, agrees to provide all of the equipment necessary. It is clear that the ilku service here is military. Agreement regarding the Ilku-Service2   In the joy of his heart, Gedaliah son of Rahim-ili speaks thus to RimutNinurta son of Murašu:   You hold the land both planted and in stubble, the horse-land (bît-sisî) of Rahim-ili, the whole part of Barik-ili, because in adopting Rahim-ili, your uncle Ellil-šum-iddin has received it.   Give me one horse with his ḫušuku and the harness, a DI of leather, an iron caparison, an iron helmet, a bodysuit of ḫattu leather, a shield for my torso, 120 (heavy)-impact and flying (light) arrows, an iron rebû for the shield, two iron swords, and one mina of silver for my supplies on the order of the king in view of my mission to Uruk, and I will fulfill the ilku service incumbent upon the horse-land (bît-sisî)—all of it.   Then Rimut-Ninurta agreed to it and gave me one horse and all the accessories of combat, conforming to what is written above, plus one mina of silver for my supplies on the order of the king in view of the mission to Uruk which is incumbent (as the ilku service) upon the said horse-land (bît-sisî). Gedaliah carries the responsibility if he does not present what 2.  I provide here my English translation of the French of Cardasçia, Les archives des Murašû, 179–82.

Exploitation of Depopulated Land in Achaemenid Judah

157

has been entrusted to him. Gedaliah will draw up the receipt coming from Sabin, head of the army paymasters, and will give it to Rimut-Ninurta.   Names of nine witnesses and of the scribe, Nippur, 18th day of the ninth month, the second year of Darius II.

According to this text, Ellil-šum-iddin of the house of Murašu “adopted” Rahim-ili and in so doing received control of property known as “horse-land” (bît-sisî). The land was a fief that was encumbered by the ilku service, in this case military service. In fact, Rahim-ili was adopted only in order to enable the Murašu firm to take control of the land, or rather, the produce from it, because the land itself was inalienable. The adoption was thus a necessary legal fiction. The ilku service owed by Rahim-ili was then fulfilled by his son. Another example of the ilku occurs in a letter drawn from the Ar­ sames archive (TAD A 6.11). This letter shows unequivocally that the satrap Arsames, a member of the Persian royal family, had estates that he controlled and parceled out in Egypt (in addition to the ones that the Murašu documents show he had in Babylon). The letter reveals that Arsames could give these estates to anyone he pleased, or he could keep them for himself. The holding that he had assigned to Pamun and then given over to Peṭosiri, Pamun’s son, was a fief on which ilku service was owed and for which Peṭosiri, Pamun’s son, was now obligated. Transfer and Registration of Lease, Late Fifth Century (TAD A 6.11) Inside   From Arsames to Nakhṭhor, Kenzasirma and his colleagues.   And now, a certain Peṭosiri, plenipotentiary (‫)ורשבר‬, my servant, sent to me. He says thus: There is a certain Pamun, my father. When there was unrest in Egypt, that one perished. And his domain (‫)בגה‬, of 30-a(rdab) seed capacity, which the one named Pamun, my father, had been holdingas-heir (‫)מהחסן‬, that was therein abandoned since all our household personnel per[rished]. The domain of Pamun, my father, [was not given] to me. Let one take thought of me. Let them give (it) to me. Let me-hold(it)-as-heir.   Now, Arsames says thus: If it is so according to these words which Peṭosiri sent [to me that] (one) named [Pamun], his father, when there was the unrest in Egypt, perished with [his household] personnel [and] the domain of that Pamun his father, of 30 a(rdabs) seed capacity was abandoned and not made over [to my estate] and not given by me to another servant, then I do give the domain of that Pamun to Peṭosiri. You, notify him. Let him hold-it-as-heir and let him pay the hlk (ilku, ‫ )הלכא‬to my estate just as formerly his father Pamun had been paying.   Artavahya knows this order. Rashta is scribe.

158

Lisbeth S. Fried

Outside   From Arsames to Nakhṭhor the official, (and to) [Kenza]sirma and his colleagues the tax-accountants/registrars who are in Egypt.   (Demotic:) About the fields of Pamun, which I have given to Peṭosiri.

It is clear from this letter that estates throughout the Empire were confiscated by the satrap and then rented out by him, sometimes back to the original owner or his heir. Besides the rent, these were additionally encumbered with the ilku-service. This service-tax was then inherited along with the land and obligated the son. A Letter (A 1) from Akhvamazda, the Satrap, to Bagavant, His Governor, in the Sixth Year of Artaxerxes III The ilku is also referred to in a letter from Akhvamazda, the satrap of Bactria, to one of his governors, Bagavant, in the sixth year of Artaxerxes III (Naveh and Shaked 2012: A 1). According to that letter, Bagavant, the Persian governor of a province in northern Bactria, had arrested a group of camel drivers who were guarding the camels of the king because they had not paid the ilku. The master of the camel drivers complained to Akhvamazda, the satrap, that the camel drivers were his servants, his apprentices, and that they were not obligated for the ilku tax. We are not told in the letter just why they were not obligated nor why the governor had thought that they were, but we may suppose that the act of guarding the king’s camels fulfilled the ilku-service, so they were not obliged to make an additional monetary payment (Shaked, personal communication, July 12, 2013). Thus, the ilku-service was not always military. Thus, texts from the Murašu archive, the Arsames archive, and from the newly translated archive from Bactria confirm what was already clear from the Greek historians—that land throughout the Empire was confiscated and held by the Persian king and members of his royal family and allotted to ordinary people in return for service, military or othְ ‫( ַהל‬ilku) erwise, or for a fee in exchange for service. References to the ‫ָך‬ in the Aramaic letters of Ezra show that the ilku-service tax also existed in Judah. Mandattu With this in mind, we may now examine the word mandattu as it appears in Neh 5:4. ְ ‫ָוינּו ֶכסֶף ְל ִמּדַ ת ַה ֶּמל‬ ‫ּוכָרמֵינּו׃‬ ְ ‫ֶך ְׂשדֹתֵ ינּו‬ ִ‫ֲׁשר א ְֹמִרים ל‬ ֶ ‫ְוי ֵׁש א‬ As stated above, most commentators translate this verse: “There are those who say, ‘We had to borrow money on our fields and vineyards in order to pay the king’s tax’,” inserting the preposition ‫ על‬before “fields

Exploitation of Depopulated Land in Achaemenid Judah

159

and vineyards.” In contrast, as noted above, Gunneweg and Guillaume assert that the phrase “fields and vineyards” is an addition unrelated to the rest of the verse and should be omitted. They translate the text simply: “We had to borrow money to pay the king’s tax.” Examination of the occurrence of the word middat, or mandattu in extrabiblical literature, however, sheds light on this term and on the role in our verse of the two words that follow it. A construction that is identical to our verse in Nehemiah appears in a letter from the Bactrian archive. Letter (A 8) from Akhvamazda, Satrap of Bactria, to His Governor, Bagavant The letter is fragmentary, but we are able to read mindat malkaʾ, exactly as in Neh 5:4 (the translation is that of Naveh and Shaked):   . . . you (plural) worked in his grain (field)   . . . Bactria, in order to collect the king’s rent (mindat malkaʾ)   . . . You (plural) bring [it] to me, to the fortress Zariaspi (i.e., the fortress in the capital city of Bactria)   . . . [as ] I said, and thus do (plural), as one   . . .[and if you (plural)] do not act completely as [you were told?]. . . . (Naveh and Shaked 2012: A 8)

Although the sender and addressee are missing, we may assume that the letter is from the satrap Akhvamazda to his provincial governor, Bagavant, as are all the other letters in this archive. Since the addressee is addressed in the plural, it may also include soldiers from the local garrison. In the letter, Akhvamazda complains that he has not received the mindat malkaʾ, the king’s mandattu, due on the property. Naveh and Shaked translate mindat malkaʾ as the rent due on property that is owned by the king—that is, rent paid to the king to work on and live off a royal grain field. Either Bagavant is renting the property himself or, more likely, he is charged with the task of collecting the rent from the holders of the property and bringing it to the satrap. The satrap was apparently responsible for collecting the rents on the king’s property and forwarding it to the king. The word also occurs several times in the Elephantine papyri, most often in the customs account to indicate payment for the temporary use of the Nile River paid by ships passing through it. It also occurs in the Arsames archive (TAD A 6.13): Letter from Arsames to Nakhtḥor, Kenzasirma, and His Colleagues, Late Fifth Century (TAD A 6.13). From Arsames to Nakhtḥor, Kenzasirma, and his colleagues   And now, Varuvahya, “son of the house” (bar baitaʾ, member of the Persian royal family) says to me here thus, saying:

160

Lisbeth S. Fried

  “The domain (bagaʾ) which was given to me by my lord in Egypt— that (one), they are not bringing to me anything from there. If, then, it thus please my lord, let a letter be sent from my lord to Nakhtḥor, the official (pakidaʾ), and [to] the accountants (hammarakarriaʾ, tax accounts/ registrars) that they issue instruction to one named Ḥatubasti, my official, to the effect that he release the rent of those domains (mandat bagayaʾ) and bring it to me with the rent (mandattaʾ) which Nakhtḥor is bringing.”

According to this text, Arsames had assigned land in Egypt to another member of his royal family, Varuvahya, but Varuvahya complains that he has not received the mandat due on it. The letter uses the same construction as in our verse in Nehemiah 5. It refers to the mandat bagayaʾ, that is, the mandat of the domains. The term bagayaʾ thus serves the same role in this phrase as does ‫ּוכָרמֵינּו‬ ְ ‫ ְׂשדֹתֵ ינּו‬in Neh 5:4. They are domains on which the mandattu rent must be paid. The phrase is in construct relationship. These texts from both Egypt and Bactria enable us to understand the mandattu as rent owed on land belonging to the royal family. The term mandattu also appears in the Murašu archive. Here it also refers to rent on land, but it seems to refer to a surcharge, over and above the normal rent (the sūtu) that was agreed to in advance; and whereas the sūtu rent was always in barley, the mandattu was usually in sheep, cattle, or flax (Stolper 1985: 140–41; Cardasçia 1951: 75, 135; Ries 1976: 76–78). We must conclude, therefore, that the maddat hammelek in Nehemiah 5 refers to a rent on fields and vineyards that belonged to the Persian king, in this case, Artaxerxes I, and that have been rented out to the Judeans. They are not borrowing against their own fields and vineyards to pay the king’s tax. They are borrowing money in order to pay rent to the king for the fields and vineyards that they are renting from him. We learn from this word, moreover, that they do not own their land, but it is the king’s, and they are simply renters. We Have Forced Our Sons and Daughters to Be Slaves We are now in a position to understand v. 5a of our text in Nehemiah. We read ‫ׁשים אֶת־ ָּבנֵינּו ְואֶת־‬ ִ ‫אנ ְַחנּו כ ְֹב‬ ֲ ‫ׂשרֵ נּו ִּכ ְבנֵיהֶם ָּבנֵינּו ְו ִהּנֵה‬ ָ ‫ְוע ַָּתה ִּכ ְבׂשַ ר ַאחֵינּו ְּב‬ ‫ִכּבָׁשֹות ְואֵין ְלאֵל יָדֵ נּו׃‬ ְ ‫ָדים ְוי ֵׁש ִמ ְּבנֹתֵ ינּו נ‬ ִ ‫ְּבנֹתֵ ינּו ַלעֲב‬ Now, is not our flesh like the flesh of our brothers, are not our children like their children? But see, we alone are forcing our sons and our daughters to be slaves and some of our daughters are being ravished, and there is nothing we can do about it;

5

Exploitation of Depopulated Land in Achaemenid Judah

161

What was going on here? There are two possibilities. The first is that people were selling their children to raise money to pay the rent on their fields and vineyards. It would have been for seven years at most, in conjunction with Neh 10:32, but even so it was horrible enough if the girls were being ravished. A second possibility is that the service refers to the ilku. If the mandattu was a surcharge, a rent over and above the sūtu rent, it was also a rent separate from the ilku—that is, from the service obligation typical on land assigned as fiefs by the king, satrap, or governor. Since this was land owned by the king and assigned to the Judeans, there was most likely an ilku-service also owed on it. For those who could afford it, the ilku would have been paid off in silver; for those who could not, it had to be paid off in actual service, and a son (and evidently sometimes even a daughter) would be recruited to pay it. We saw this in the text from the Murašu archive, where Gedaliah, the son of Rahim-ili, was the one who actually paid the ilku-service that his father owed. I have often suspected that the words of Samuel (1 Sam 8:11–17) referred to a Persian king, and not a Judean monarch:   11These will be the ways of the king who will reign over you: he will take your sons and appoint them to his chariots and to be his horsemen, and to run before his chariots; 12and he will appoint for himself commanders of thousands and commanders of fifties, and some to plow his ground and to reap his harvest, and to make his implements of war and the equipment of his chariots. 13He will take your daughters to be perfumers and cooks and bakers.  14 He will take the best of your fields and vineyards and olive orchards and give them to his courtiers. 15He will take one-tenth of your grain and of your vineyards and give it to his officers and his courtiers. 16He will take your male and female slaves, and the best of your cattle and donkeys, and put them to his work. 17He will take one-tenth of your flocks, and you shall be his slaves.

Our Fields and Our Vineyards Belong to Others, ‫אחִֵרים‬ ֲ ‫ּוכ ָרמֵינּו ַל‬ ְ ‫ּוׂשדֹתֵ ינּו‬ ְ At the end of v. 5, we read, “[O]ur fields and vineyards belong to others.” What is the purpose of this addition? It is already clear that the ְ ‫;)מּדַ ת ַה ֶּמל‬ land has been rented from the king (this is the meaning of ‫ֶך‬ ִ why then belabor the point that they do not own their own land? According to v. 4, they had to borrow silver to pay the king his rent, ְ ‫מּדַ ת ַה ֶּמל‬. the ‫ֶך‬ ִ The debt was usually owed at the time of the harvest, and if it was unpaid, then the creditor took control of the property. Guillaume (2010) claims that there was no economic crisis in 5th-century Judah because these were antichretic loans. Guillaume is right in seeing the loan as antichretic, but he errs in underestimating the effect of this type of loan.

162

Lisbeth S. Fried

These fiefs were inalienable, so that if the holder wanted to take out a mortgage it had to be an antichretic mortgage; that is, the creditor did not receive the land, only the usufruct, and the creditor had the right to the usufruct until the debt was paid. Deprived of the usufruct of his own property, he then had no way ever to repay the loan. The debtor was reduced to the status of a tenant, and as Wunsch (2002: 240) puts it, “[R]epayment of the debt becomes a matter of wishful thinking, rather than economic reality.” The creditor assumed the position of owner, paying the taxes and liens on the property, and enjoying the usufruct without actually holding title—thus the cry: “Our property is in the hands of others!” The only way out was for the debt to be abolished.

Nehemiah’s Solution (Nehemiah 5:6–13) Nehemiah’s solution is recorded in v. 11. ‫ּומאַת ַה ֶּכ֤סֶף ְוהַָּדגָן‬ ְ ‫ָׁשיבּו נָא ָלהֶם ְּכהַּיֹום ְׂשדֹתֵ יהֶם ּכ ְַרמֵיהֶם ז ֵיתֵ יהֶם ּובָּתֵ יהֶם‬ ִ‫ה‬ ‫ׁשים ָּבהֶם׃‬ ִ ֹ ‫ַּתם נ‬ ֶ ‫ֲׁשר א‬ ֶ ‫ִצהָר א‬ ְ ‫ה ִַּתירֹוׁש ְו ַהּי‬ [Nehemiah commands:] “Restore to them, this very day, their fields, their vineyards, their olive orchards, and their houses, and the interest on money, grain, wine, and oil that you have been exacting from them.” Nehemiah orders the return of fields, orchards, and vineyards. But what actually was returned? In antichretic loans, the creditor receives control of the land, all the usufruct, as well as the continued enjoyment of the usufruct until the debt is paid. The owner of the land pays rent to the creditor to continue living on it; he has become a tenant on his own land. Nehemiah here has ordered the creditors to restore control of the fields and vineyards to the debtors plus ownership of all the produce from them and to abolish any interest that the debtor owes, and to turn over any interest that the creditors may have gained from borrowing on it or from leasing it out to others. The abolishment of this debt should not surprise us. Jursa (2002: 212) suggests that local cancellations of debts and service obligations were well attested under the NeoAssyrians and would have continued under the Achaemenids as well.

Conclusion: Land Use in Depopulated Areas We may conclude that the land in Achaemenid Judah accrued to the Persian king by right of eminent domain when the Persians conquered Babylon. Indeed, that is the meaning of the phrase “the king’s ְ ‫מּדַ ת ַה ֶּמל‬. rent,” the ‫ֶך‬ ִ The land was parceled out into fiefs on which, most likely, as everywhere else, ilku-service was required in addition to the

Exploitation of Depopulated Land in Achaemenid Judah

163

king’s rent. This picture is consistent with what we know about land use everywhere in the Achaemenid Empire.

Bibliography Albertz, R. 1994 A History of Israelite Religion in the Old Testament Period, vol. 2: From the Exile to the Maccabees, trans. J. Bowden. Louisville: Westminster John Knox. Blenkinsopp, J. 1988 Ezra–Nehemiah: A Commentary. OTL. Philadelphia: Westminster. Briant, P. 1985 Dons de Terres et de Villes: L’Asie Mineure and le Contexte Achéménide. REA 87: 53–71. Cardasçia, G. 1951 Les Archives des Murašû, une Famille d’Hommes d’Affaires à l’Époque Perse (455–503 av. J.-C.). Paris: Impr. Nationale. Clines, D. J. 1984 Ezra, Nehemiah, Esther. NCBC. Grand Rapids, MI: Eerdmans. Dandamayev, M. A. 1967 Die Lehnsbeziehungen in Babylonien unter den ersten Achämeni­ den. Pp. 37–42 in Festschrift für Wilhelm Eilers: Ein Dokument der Internationalen Forschung zum 27. September 1967, ed. G. Wiessner and W. Eilers. Wiesbaden: Harrassowitz. 1974 The Domain-Lands of Achaemenes in Babylonia. AoF 1: 123–27. Driel, G. van 2002 Elusive Silver: In Search of a Role for a Market in an Agrarian Environment—Aspects of Mesopotamia’s Society. Nederlands Instituut voor het Nabije Oosten te Leiden 95. Leiden: Nederlands Instituut voor het Nabije Oosten. Faust, A. 2012 Judah in the Neo-Babylonian Period: The Archaeology of Desolation. SBLABS 18. Atlanta: Society of Biblical Literature. Gunneweg, A. H. J. 1987 Nehemia: Mit einer Zeittafel von Alfred Jepsen und einem Exkurs zur Topographie und Archäologie Jerusalems von Manfred Oeming. KAT 19/2. Berlin: Evangelische Verlag. Guillaume, P. 2010 Nehemiah 5: No Economic Crisis. JHebS 10: 1–21. DOI:10.5508/ jhs.2010.v10.a8. Hoftijzer, J., and Jongeling, K. 1995 Dictionary of the North-West Semitic Inscriptions. Leiden: Brill. Hoglund, K. G. 1991 The Achaemenid Context. Pp. 54–72 in Second Temple Studies, vol. 1: Persian Period, ed. P. R. Davies. JSOTSup 117. Sheffield: JSOT Press.

164

Lisbeth S. Fried

Jursa, M. 2002 Debts and Indebtedness in the Neo-Babylonian Period: Evidence from the Institutional Archives. Pp.  197–217 in Debt and Economic Renewal in the Ancient Near East, ed. M. Hudson and M. Van de Mie­ roop. International Scholars Conference on Ancient Near Eastern Economies 3. Bethesda, MD: CDL. 2010 Aspects of the Economic History of Babylonia in the First Millennium bc. AOAT 377. Münster: Ugarit-Verlag. Myers, J. M. 1965 Ezra–Nehemiah. AB 14. New York: Doubleday. Naveh, J., and Shaked, S. 2012 Aramaic Documents from Ancient Bactria (Fourth Century bce). London: The Khalili Family Trust. Ries, G. 1976 Die neubabylonischen Bodenpachtformulare: Münchener Universitäts­ schriften. Abhandlungen zur Rechtswissenschaft­lichen Grundlagenforschung 16. Berlin: Schweitzer. Reinmuth, T. 2002 Der Bericht Nehemias: Zur literarischen Eigenart, Traditionsgeschicht­ lichen Prägung und Innerbiblischen Rezeption des Ich-Berichts Nehemias. OBO 183. Göttingen: Vandenhoeck & Ruprecht. Rosenthal, F. 1995 A Grammar of Biblical Aramaic. 6th rev.  ed. PLO 5. Wiesbaden: Harrassowitz. Roth, M. T. 1997 Law Collections from Mesopotamia and Asia Minor. 2nd ed. WAW. Atlanta: Society of Biblical Literature. Rudolph, W. 1949 Esra und Nehemia Samt 3 Esra. Tübingen: Mohr Siebeck. Stolper, M. W. 1985 Entrepreneurs and Empire: The Murašû Archive, the Murašû Firm, and Persian Rule in Babylonia. Leiden: Nederlands Instituut voor het Nabije Oosten. Williamson, H. G. M. 1985 Ezra, Nehemiah. WBC 16. Waco, TX: Word. Wright, J. L. 2004 Rebuilding Identity: The Nehemiah-Memoir and Its Earliest Readers. BZAW 348. Berlin: de Gruyter. Wunsch, C. 2002 Debt, Interest, Pledge and Forfeiture in the Neo-Babylonian and Early Achaemenid Period: The Evidence from Private Archives. Pp. 221–55 in Debt and Economic Renewal in the Ancient Near East, ed. M. Hudson and M. Van de Mieroop. International Scholars Conference on Ancient Near Eastern Economies 3. Bethesda, MD: CDL.

The Achaemenid Policy of Reproduction Josef Wiesehöfer

Christian-Albrechts-Universität zu Kiel

Introduction (Music of pipes and flutes can be heard coming from a garden below them [Alexander and his companions at Babylon]. . . . Intrigued by the music, Alexander moves down a staircase into . . . a cool garden paradise. As Alexander descends, Darius’ lovers appear, in ones or twos, then threes, accumulating quickly like flowers in their curiosity to set eyes on the young conqueror. . . . To Alexander’s enchanted eyes, it’s a vision of Paradise more than a harem. The Persians have constructed a world from water pools, to food, fragrance, music, clothes, and love, all designed to seduce the senses.) Perdiccas (agog):  By the gods! What is this?! Craterus: I venture one for each night of the year! No wonder Darius ran—when he had this to come back to! Cassander (aside): Time for a woman, Alexander? Alexander: I think you’ll manage well without me, Cassander. (The fat, older eunuchs prostrate themselves, their world aflutter at this invasion of throbbing Greek masculinity. Women keep appearing from small apartments—partially concealed by foliage and rock—and several play flutes and various shapes of lyres, luring the men.) Leonnatus: Help me, Aphrodite! How will I ever go back to Lysiclea after this? Nearchus: I advise you not to touch, Leonnatus. I’ll take care of it for you. (The men ravish the women with their eyes, their tastes different, their lust similar. . . .)

No doubt, the shooting script of scene 36 of Oliver Stone’s Alexander serves all orientalistic clichés of a Near Eastern royal court—the film even exceeds the screenplay in that respect. Not for nothing did the New Yorker mark the cinematic representation of the conquest of the Achaemenid Empire by Alexander ironically as “Operation Persian Freedom” (Lane 2004: 125).1 And Lloyd Llewellyn-Jones (2010: 274) has 1.  I owe this quotation and the reference to the film scene to Llewellyn-Jones (2010: 244–45)

165

166

Josef Wiesehöfer

rightly stressed the meaning of the “harem women” in Stone’s display of “all the familiar Orientalist notions about the inferiority and picturesqueness of Eastern societies”: they not only bind together the two dimensions of space (“the call of far-distant lands and alien peoples”) and time (“the pursuit of the historical ‘other’”), they embody “most rigidly the mythic elements of the Western perception of the Orient: domestic submissiveness, erotic willingness, and exotic sensuality.” Stone’s preoccupation with an imaginary Persia does not relieve me of the duty to have a close look at the historical royal Achaemenid policy of reproduction and, along with that, at the role of women in the immediate ambit of the ruler. In my contribution, I am therefore going to discuss the following three topics: (1) The role of polygyny in the Iranian parts of the Achaemenid Empire: Was it an elite phenomenon or widespread? And what role did unions of Iranians and non-Iranians play in it? (2) The role of what many scholars and Oliver Stone call “harem”: Is it an appropriate term for the role of females and other people at court? And what sorts of hierarchy and modes of communication can we detect at the Achaemenid court? (3) The significance, not least, of reproduction in the royal household and its Iranian milieu documented by the Persepolis Fortification Tablets: How may we describe the political significance and the social status of the ruler’s multiple offspring from the numerous unions with wives and concubines? And what importance do the rulers assign to the children of their Iranian and non-Iranian subjects?

The Role of Polygyny in the Iranian Parts of the Achaemenid Empire Not least, our knowledge of the polygynous marriage and mating customs of the Achaemenid kings is due to the information given by the Greek classical sources. It was Walter Scheidel (2009b: 279) who, some years ago, stressed the fact that for the Greeks of the 5th and 4th centuries b.c.e. “not only despotism but also polygyny were marks of the ‘barbarian,’ ” institutions “which could then be embellished for increased entertainment value and harvested for moralizing or xenophobic value judgements.” Despite all topical biases—for example, the idea of female seclusion or the motif of the woman-grabbing enemy—there can be no serious doubts as to the historicity of polygyny and the transfer of captive women to the palaces of the king. Members of the Iranian aristocracy practised polygyny, too, however, on a proportionately smaller scale. And like Augustus, who took the necessary precautions against noble competition by restricting the aristocratic clientelae and by limiting the scope of noble manumission,

The Achaemenid Policy of Reproduction

167

the Achaemenid kings probably provided for the observance of differences in rank between themselves and their bandakā in the field of polygyny. Poor Persians most likely were traditionally monogamous; Hesiod’s advice to his brother Perses to admit only one wife to his oikos for economic reasons (Hesiod, Works and Days 405) was presumably also heeded by them. And as in Sasanian Iran, elite polygyny, despite royal and aristocratic appropriation of foreign women, may also have “skewed reproductive opportunities to such an extent that marriageable women were lacking in the lower classes” (Scheidel 2009b: 279). By the way, it would be interesting to ask how the monogamous principle of Greeks and Romans came to be so firmly established (in the West) even among (customarily polygynous) elites . . . how it co-existed with de-facto polygyny facilitated by sexual congress with chattel slaves . . . and how it became entrenched in Christian doctrine that survived the fall of the Roman state and ensured its survival and spread in later European (and subsequently world) history. (Scheidel 2009a: 140)

As for Persia, it must also be borne in mind that, as in Rome with its ius conubii, the Achaemenid Empire knew the right or the privilege of a so-called Persian marriage for foreigners, who were even allowed to apply for the recognition of their offspring as “Persian.” This privilege is recorded, to give just one example, for Miltiades’ son Metiochus (Herodotus 6.41), who had been deported to Persia in 493 b.c.e. In view of the role of ethnicity in the distribution of benefices and positions in the Achaemenid Empire, such a privilege is not surprising. However, the frequency of its being awarded, in contrast to the tradition in Rome, cannot be measured.

The Role of “Harem” Most recently, scholars have dealt extensively with the Achaemenid court and court society. Among other things, scholars heatedly debated and debate the right to use the term “harem” for the nucleus of the court. They are no longer discussing, of course, the pitfalls of the traditional naïve use of the term that fell into the trap of the respective clichés of the classical or early modern European sources—that is, ideas of a secluded female-only space or a form of purdah. Today, scholars such as Amélie Kuhrt or Pierre Briant regard the harem only as a by-product of Greek fantasy, whereas others, such as Lloyd Llewellyn-Jones (2010: 261; but see Briant 2013), think that, “in the absence of any other purposeful term, ‘harem’ is the most appropriate one to use to describe the domestic makeup of the Persian royal household” and that the royal

168

Josef Wiesehöfer

women of Achaemenid Persia “formed part of a strict hierarchical court structure that moved in close proximity to the king.” In this contribution, I am not able to elaborate on that controversy.2 However, the fact remains that, first, female members of the royal family (Elamite dukšiš; compare the term irti for “wife”)—in particular the “King’s mother,” the “King’s wife” (Greek gyne tou basileos), and the King’s “daughters”—played an important role at court (Brosius 1996; Madreiter and Schnegg forthcoming). The first was important because of being the mother of the reigning king, and the second as the mother of the prospective heir to the throne. Both had immediate access to the king, were able to take part in royal festivities and audiences, and were even allowed to dine with the king. Wives were normally selected from among the Persian nobility, but we also know of royal marriages to half-siblings. As in all monarchical systems, royal marriages often served political needs. In the case of Darius I, his marriage to Cyrus’s daughter Atossa helped him to connect his own, Achaemenid line with that of the Teispids. It comes as no surprise that Darius chose the offspring of that marriage, Xerxes, as heir to the throne, although Xerxes was not his oldest son. Darius’s marriage to the daughters of his fellowconspirators, Otanes and Gobryas, gave him the necessary support of the noble Persian families at the beginning of his reign. On the other hand, the endogamous marriages of members of the royal clan could help to control family lineages and to restrict other families’ influence on royal affairs. Second, the cuneiform tablets from Persepolis and Mesopotamia give evidence that royal wives and princesses were self-confident and economically independent women and did not at all live in a kind of monastic seclusion. Although very often anonymous (cf. P F a 5 et al.), some of them are named individually if they act on their own; Irdabama, Irtashtuna, and others are mentioned as property holders and/or estate owners who give direct orders to their bailiffs and stewards and have their own work force. They travel extensively, visit their estates, and sometimes even administer their wealth individually. Besides, apart from being participants in royal banquets and festivities, they seem to have had their own “table,” the economic base of which was “interwoven with the Persepolis economy at large” (Henkelman 2010: 732). Third, as Leslie Peirce (1993: 3) has observed for the Ottoman Empire, 2.  I have to confess, however, that I am a member of the group that speaks against the use of the term for reasons of content (it is not really able to reflect the variety of personal relations within Persian court society) and of methodology (there is the danger of false analogies and connotations).

The Achaemenid Policy of Reproduction

169

Sex for . . . any monarch in a hereditary dynasty, could never be purely pleasure, for it had significant political meaning. Its consequences—the production of offspring—affected the succession to the throne, indeed the very survival of the dynasty. It was not a random activity. . . . Sexual relations between the [ruler] and chosen women of the harem were embedded in a complex politics of dynastic reproduction.3

Last but not least, it has rightly been stated that Alexander’s seizure of the Persian royal household after Issus “jeopardized the legitimacy of Darius [III’s] reign” (Llewellyn-Jones 2010: 266). Besides, the Macedonian’s multiple marriages into the royal Achaemenid family and his alleged honoring of and respect for the “King’s Mother,” Sisygambis (which are said to have been answered with her preference for Alexander over her own son, Darius), make Alexander appear, as Briant (2002: 876) has rightly pointed out, the be the “last of the Achaemenids.”

The Significance of Reproduction in the Royal Household Let us now move to our last topic, the role of the offspring of royal, noble, and nonaristocratic unions. In contrast to Greece and Rome, the multiple offspring who resulted from the unions of Persian monarchs with their numerous wives and concubines were “free of the social stigma of bastardy” (Llewellyn-Jones 2013: 7). The example of the Great Kings Darius II and Artaxerxes II shows that even the sons of royal concubines had the potential to become king, although as an exception to the rule. Furthermore, any prince who was unable to reach the highest levels of power and authority still had the option of supporting his father politically or militarily in many ways and of gaining additional power and esteem. “In fact, this lack of social stigma over legitimacy propagated and unified the ruling eastern dynasties in tight family bonds, the mammoth scale of which was never experienced by Christian monarchies” (Llewellyn-Jones 2013: 7). Of course, the bottleneck of succession to the throne made the Achaemenid court a dangerous space and communication at court quite insincere (Wiesehöfer 2013). This is demonstrated in particular by several regicides and murders of kindred rivals to the throne committed within the royal family. However, in the end, reproductive capabilities went hand in hand with dynastic success. The Greeks, who were “obsessed with issues of legitimacy” (Llewellyn-Jones 2013: 7), never really understood the significance of polygyny with regard to the political accomplishments of the Persian dynasty. The same success can be observed in the Persian noble families and the Greek and biblical testimonies about the extinction of whole 3.  I owe the reference to this book to Llewellyn-Jones (2010).

170

Josef Wiesehöfer

resistant families by the Persian kings, originally meant to illustrate the pernicious power of despotic rulers, but in reality pointing to the significance of family bonds also in an aristocratic milieu. Thankfully, recent research on the Mass Wedding at Susa in 324 b.c.e. (Arr., Anab. 7.4.4–8; Just. 12.10.9; Plut. Alex. 70.2; Diod. 17.107.6; Curt. 10.3.12; cf. Athen. 12.538b–539a) is, in contrast to Oliver Stone’s idea of it, no longer obsessed by the idealistic interpretation of Alexander’s conquests as being essentially driven by his vision of a “unity of mankind.” Instead, it points to the practical use of the unions between Macedonian officers and Iranian noble women: the offspring of those unions, who like their mothers were forbidden to accompany their fathers back to Macedon, were meant to become members of a new imperial elite, raised in and familiar with the worlds of both parents. The study of the “Achaemenid policy of reproduction” does not exclusively deal (and that is my last point) with members of the elite; this is demonstrated by the following contemporary indigenous Iranian testimony: 61 marriš (of) beer Gobryas the ‘fire chief’ received, and gave (it) to post partum women. 40 women (who) bore males received each 1 marriš, and 42 women (who) bore female children received each 5 QA. For a period of 12 months they received. 23rd year.” (PF 1216)

This is a text from the Persepolis Fortification archive, an archive that illustrates in detail the workings of the Achaemenid (economic) administration in the imperial heartland in the years between 509 and 494 b.c.e. This text gives insight into the handling of local production, the labor force—to some extent involved in the construction of the royal terrace at Persepolis and the nearby city—bureaucracy, and the affairs of a variety of large estates in Fars. Many caveats have rightly been listed as far as this “dormant” archive (Henkelman) is concerned. The archive—which to be more precise is more of an interconnected “dossier,” as Amélie Kuhrt (2008: 568) has put it—does not provide a window on the Empire as a whole and is highly restricted in time and space. Important parts of the population of Fars, such as free peasants or pastoral peoples, do not appear in the texts. Whereas food, pottery, and leather production is well attested, there is only scant information on textile production or metal processing, which should have been quite important “industrial” sectors, and the material is also highly restricted as far as writing material and the overall mechanisms of the whole administrative system are concerned. “We only have brief notes, annual accounts, orders and memoranda needed by the administration to keep track of its outgoings and resources” (Kuhrt 2007: 764). The male and female workers of Persepolis, called kurtaš, were drawn from different age and status groups and ethnic backgrounds,

The Achaemenid Policy of Reproduction

171

and they were paid according to their skills and the level of their responsibility—we hear of supervisors, attendants, specialist artisans, and laborers. This is true for both the female and the male work force. As the text I quoted and other similar documents clearly show (see Brosius 1996: 171–82), new mothers and pregnant women of the so-called pašap and harrinup groups (i.e., those linked to members of the royal household and those without that connection) received special, higher rations for one month, consisting of grain or flour, beer or wine. Among them were female members of Darius I’s daughter Artystone’s (Irtashtuna’s) or Lady Irdabama’s labor force. However, and this is also proven by the text, the mothers of sons received twice the ration amount of mothers of baby girls. This does not necessarily reflect a preference for boys; it might be due to the idea that mothers of male offspring needed more nourishment. The children of the work force, whose existence does not necessarily point to families of workers—the number of adults outstrips that of children—are classified according to gender and payment; most likely, they were grouped into age sets, sometimes defined by height, and they seem to have been paid according to their age and their contribution to the workload. It has been postulated that the fact that the number of births of male children only slightly exceeds the number of girls born points to the nonexistence of female infanticide in Achaemenid Fars. However, such an idea mixes exposure and infanticide; and even for Greece and Rome, scholars have rightfully raised doubts as to disproportionate female exposure or infanticide. The mother’s rations, because of the consistent rules and principles they followed, point to an Achaemenid policy of fostering their subjects’ reproduction. Unfortunately, we are not able to decide whether this policy was just meant to enlarge the somehow dependent work force of the king and members of the royal house in Fars or was pursued overall for general demographic reasons and in other parts of the Empire. Scholars trying to imply that the Achaemenids had something like a demographic consciousness normally quote Herodotus 1.136: It is established as a sign of manly excellence next after excellence in fight, to be able to show many sons; and to those who have most the king sends gifts every year: for they consider number to be a source of strength. (ET G. C. Macaulay),

Strabo 15.3.17: They marry many wives and also maintain a number of concubines for the sake of having many children. Each year the king presents prizes to those who have many children. (ET A. Kuhrt)

and Plutarch, Alex. 69.1:

172

Josef Wiesehöfer

. . . the Persian kings, who, whenever they arrived in this province, presented each mother with a gold coin. (ET A. Kuhrt)

However, the texts say nothing more than that the Persian kings were interested in having as many soldiers, officials, and workers as possible (Briant 2002: 277ff.). The population size of the Persian Empire, although it would be the best indicator of economic performance, is unclear because of the problematic source material concerning birth, death, and migration (Wiesehöfer forthcoming). However, “whilst allowing for short-term variation, the balance of births and deaths must have been fairly stable in the long run” in the Empire; “even a seemingly moderate net shortfall of one birth per woman (of, say, four instead of five) would have halved a given population within three generations, whereas a net surplus of one birth per woman would have doubled it, neither of which was at all likely to happen” (Scheidel 2009a: 137). The Great Kings and their officials would surely have noticed both kinds of demographic problems. As for concrete royal Persian answers to problems of depopulation, we historians of Achaemenid history are therefore highly dependent on our archaeological colleagues’ expertise.

Bibliography Briant, P. 2002 From Cyrus to Alexander: A History of the Persian Empire. Winona Lake, IN: Eisenbrauns. 2013 Review of King and Court in Ancient Persia 559 to 331 bce: Debates and Documents in Ancient History, by Llewellyn-Jones. Bryn Mawr Review 9: 44. Brosius, M. 1996 Women in Ancient Persia. Oxford Classical Monograph. Oxford: Clarendon. Henkelman, W. F. M. 2010 Consumed before the King: The Table of Darius, That of Irdabama and Irtaštuna, and That of His Satrap, Karkiš. Pp.  667–775 in Der Achämenidenhof: The Achaemenid Court, ed. B. Jacobs and R. Rollinger. Classica et Orientalia 2. Wiesbaden: Harrassowitz. Kuhrt, A. 2007 The Persian Empire: A Corpus of Sources of the Achaemenid Period. Vol. 2. London: Routledge. 2008 The Persepolis Archives: Concluding Observations. Pp.  563–70 in L’archive des fortifications de Persépolis: État des questions et perspectives de recherche, ed. P. Briant, W. F. M. Henkelman, and M. W. Stolper. Persika 12. Paris: de Boccard. Lane, A. 2004 War-Torn. New Yorker. December 6. http://www.newyorker.com/ magazine/2004/12/06/war-torn.

The Achaemenid Policy of Reproduction

173

Llewellyn-Jones, L. 2010 “Help me, Aphrodite!” Depicting the Royal Women of Persia in Alexander. Pp. 243–81 in Responses to Oliver Stone’s Alexander: Film, History, and Cultural Studies, ed. P. Cartledge and F. R. Greenland. Madison, WI: University of Wisconsin Press. 2013 King and Court in Ancient Persia 559 to 331 bce: Debates and Documents in Ancient History. Edinburgh: Edinburgh University Press. Madreiter, I., and Schnegg, C. forthcoming  Sex and Gender. In A Companion to the Achaemenid Persian Empire, ed. B. Jacobs and R. Rollinger. Blackwell Campanion to the Ancient World. 2 vols. Malden, MA: Wiley-Blackwell. Peirce, L. P. 1993 The Imperial Harem: Women and Sovereignty in the Ottoman Empire. Oxford: Oxford University Press. Scheidel, W. 2007 Demography. Pp.  38–86 in The Cambridge Economic History of the Greco-Roman World, ed. W. Scheidel, I. Morris, and R. Saller. Cambridge: Cambridge University Press. 2009a Population and Demography. Pp. 134–45 in A Companion to Ancient History, ed. A. Erskine. Malden, MA: Wiley-Blackwell. 2009b Sex and Empire: A Darwinian Perspective. Pp.  255–324 in The Dynamics of Ancient Empires: State Powers from Assyria to Byzantium, ed. I. Morris and W. Scheidel. Oxford Studies in Early Empires. Oxford: Oxford University Press. Wiesehöfer, J. 2013 Ctesias, the Achaemenid Court, and the History of the Greek Novel. Pp. 127–41 in The Romance between Greece and the East, ed. T. Whitmarsh and S. Thomson. Cambridge: Cambridge University Press. forthcoming  Demography of the Persian Empire. In A Companion to the Achaemenid Persian Empire, ed. B. Jacobs and R. Rollinger. Blackwell Campanion to the Ancient World. 2 vols. Malden, MA: Wiley-Blackwell.

The Economy and Administration of Rural Idumea at the End of the Persian Period Diana Edelman

Faculty of Theology, University of Oslo Center for Advanced Study, Norwegian Academy of Science and Letters, Oslo

Among the six announced concerns for this volume devoted to the Persian-era economy of the southern Levant, my contribution will involve a subset of three: (1) when and why areas were redeveloped; (2)  land ownership and centralized governmental interventions; and (3) literary evidence for whether the traditional mainstays of the agricultural economy shifted in the Persian period. I will focus on the Idumean ostraca, which bear dates from 363 to 313/312 b.c.e., as my primary evidence when addressing these three concerns. Over 1,900 ostraca have been dug illicitly and sold to private collectors such as D. Jeselsohn (560 pieces), R. and L. Feuer and brother R. Forbes (468 pieces), S. Moussaieff (354 pieces), G. Chaya (88 pieces), and S.  Aḥituv (58 pieces). Fewer than 1,700 are legible. Just under 100 mention deliveries to or from a storehouse (Akkadian loanword maškantu/maškattu) in Makkedah or to or from Makkedah (Aramaic Mankedah), which has led a number of those who have studied the ostraca to suspect that they originate from this ancient site. Makkedah has been identified with modern Khirbet el-Qom in the Shephelah, seven miles southeast of Lachish and about nine miles west of Hebron (Porten and Yardeni 2006: 457; 2007: 126–27). The ostraca appear primarily to detail deliveries and occasionally dispersements of food-stuffs over a period of some 50 years in the province of Idumea, the date of founding of which is disputed. The earliest textual reference to the province is found in Diodorus Siculus (15.2.4) in his account of events that transpired in 312 b.c.e. The emerging majority position is that Idumea was established as a province by the Persians ca. 400–373 b.c.e., either in response to the loss of Egypt, which moved the border region effectively to the Beer-sheba Valley (Lemaire 2002: 231; 2003: 291; Sapin 2004: 109–10); or in response to an increase in the local population that made it profitable to impose taxation, combined with the rise in importance of the Qedarites and Nabateans in the region (de Geus 1979–80: 62); or in retaliation for Qedarite support of Egypt’s struggle for independence, as a result of which its territory 175

176

Diana Edelman

west of the Arabah was turned into a subprovince ca. 387–385 b.c.e. and into the province of Idumea before 363 b.c.e. (Lemaire 2002: 148, 2006: 418–19; Knauf and Guillaume 2014). Initial book-length studies of the Idumean ostraca were prepared in 1996 by I. Ephʿal and J. Naveh (hereafter cited as E/N) and another one also by A. Lemaire (hereafter L 1), with a subsequent book by Lemaire publishing additional ostraca in 2002 (hereafter L 2). Numerous articlelength studies have appeared presenting newly acquired ostraca as well. B. Porten and A. Yardeni have several preliminary studies in print (e.g., Porten and Yardeni 2006; 2007; 2008; 2009) and are in the process of publishing the final catalog volumes (Porten and Yardeni 2014; forthcoming). To set these ostraca in a Persian perspective, they should be viewed against the Persepolis Fortification Tablets (509–493 b.c.e.; Hallock 1969; Henkelman 2008: 67–179; 2013: 530–33) and the Persepolis Treasury Tablets (492–458 b.c.e.; Henkelman 2013: 534–38); the Murashu family archive from Nippur (454–404 b.c.e.) that details three generations of a family involved in fief and estate land management;1 the Elephantine ostraca and papyri (495–399 b.c.e.; AP; TAD A–D) from a Judean military colony on Elephantine Island in Egypt, just north of the First Cataract; the Aramaic documents of Bactria (Shaked 2004; Naveh and Shaked 2012), which consist of 30 rough drafts of documents written in Imperial Aramaic on leather, and 18 loans written on wooden tally sticks, all but one during the period 353–324 b.c.e.; and the Wadi Daliyeh papyri, which constitute documents created in Samaria that date between ca. 450 and 332 b.c.e. (Gropp 1986; 2001; Dušek 2007). While the first three of these other collections date to the beginning of the Persian Empire and the Bactrian and the Idumean to its end, all can provide valuable comparative data concerning any Persian policy of altering or maintaining local economic and administrative practices. For an inter- or intraregional view, the Idumean ostraca should be read together with the 45-odd Aramaic ostraca from Arad from the mid-4th century b.c.e., when it likely served as a Persian postal way-station with a granary and a storage shed for straw on site (Naveh 1981: 175–76); the 67 4th-century ostraca from Beer-sheba that lack the name of a king in the date formula (Naveh 1973; 1979), the late Persian and early Hellenistic 1.  They primarily were active as creditors for workers of agricultural enterprises in the lending and provision of equipment, seed, tools, irrigation, and animals for this purpose. Almost all of the documents record the creation or the discharge of claims and obligations; very few are administrative records detailing current incomes and outlays. For studies of this corpus, see, for example, Stolper 1985; Donbaz and Stolper 1997.

The Economy and Administration of Rural Idumea

177

ostraca from Tel Jemme (Naveh 1992); and single ostraca from Tel elFarʿah south, Tel ʿIra, Yatta, Raphia, and likely elsewhere. The Idumean ostraca have been used to try to understand local Idumean land ownership and land-use patterns. Lemaire (2002: 143–44, 229–30) and I (2005: 265–67) think that the ostraca generally, though not exclusively, reflect the payment of taxes-in-kind by local Idumeans to regional authorities who served as imperial agents, while Porten and Yardeni (2007: 135, 143, 154) have argued more recently that the use of local measures rather than Persian imperial measures must mean that these transactions were not official. They have proposed that the barley and grain were deliveries to dealers and that the storehouse at Makkedah was a granary and a market, apparently not a governmental regional storehouse where taxes in-kind were collected and then dispensed on site or sent to postal stations and other military or administrative facilities or divisions needing them. Ephʿal and Naveh (1996) have suggested that some ostraca provide evidence that villagers could borrow grain from local storehouses and repay the interest in kind, and others imply that land was owned by the crown and leased to locals. The following issues need to be addressed to evaluate the likelihood of the various proposals, which are not mutually exclusive: (1) Persian policy about the use of imperial measures; (2) Persian policy concerning land ownership, tenancy, and taxes; (3) the activities and responsibilities of agents of fiefs or estate managers and of governmental administrators; and (4) evidence of local markets or market towns.

Persian Policy regarding the Use of Imperial Measures While the Persians did not routinely impose their coinage system and its denominational weight standards across the territory it regulated, it does appear, at first glance at least, to have employed its weights and measurement systems for recording, tracking, and disbursing agricultural produce and yield throughout the Empire. The Persepolis texts indicate that the standard capacities in the capital for dry measures were the qa (ca. 1.2 liters), the bar (= 10 qa = ca. 12 liters), and the irtiba/ ʾartab (= 30 qa = 3 bar; ca. 36 liters; Hallock 1969: 72); another calculation has concluded the Elamite ʾartab was ca. 9.7 liters (Hinz and Koch 1987: 83), so that the seʾah would have been 3.23 liters and the qa 0.323 liter. There were 10 qa in a seʾah and 3 seʾahs in an ʾartab. Liquid quantities were measured by the qa (ca. 1.2 liters) and the marriš (= 10 qa = ca. 12 liters). A lesser-used system involved the daduya, which apparently was the equivalent of a ka, and the kurriya, which was the equivalent of a bar (Hallock 1969: 72). As indicated by the Mnesimachus inscription from the Temple of Cybele at Sardis dating to ca. 200 b.c.e., Persian

178

Diana Edelman

measurements continued to be used to express the number of ʾardabs of seed needed to sow crops in paradeisoi and gardens in a leasehold of royal land held by Mnesimachus in the Hellenistic period (1.15–16; Atkinson and Atkinson 1972: 47–48, 64–67; contra a late Persian date espoused by Briant 2002: 414). In Bactria, the measurements mentioned in two provisioning requests (C1, C3) include the gun, ʾardab, seʾa, griv, and ḥufn for dry measurements and the gun, seʾa, ḥufn, mari, and sp for liquid quantities, though the regular measures were the mari and sp. Due to limited information about the number of people splitting the rations, the calculation of equivalencies between measures can only be tentatively worked out. There appear to have been 10 ʾardab in a gun, 30 griv in a gun, and 300 ḥufn in a gun. Three griv = 1 ʾardab and 30 ḥufn = 1 ʾardab. There were 10 ḥufn in a griv. The griv appears to have been the local Bactrian equivalent of the Elamite seʾah, and the ḥufn was the same as the Elamite qa (Naveh and Shaked 2012: 37). In liquid measures, the mari, used exclusively for wine, was the equivalent of the Elamite marriš, or ca. 9.7 liters (Hinz and Koch 1987) and probably corresponded to a particular type of jar used to store wine. The sp was also probably some specific type of jar. The gun seems to have been the Bactrian equivalent of the ḥōmer used to measure liquids and the kor used for dry goods (Naveh and Shaked 2012: 37). The Aramaic papyrus fragments from Saqqara in Egypt contain many of the same measurements seen in the Bactrian papyri. Thus, mention is made of the ʾardab that measured flour (Segal 10:1; 24:7; 41:1, 4; 43b:5; 45a:6; 45b:2, 4, 6, 8; 46:1; 81:4; 89:1, 145:2); the griv used to measure flour, sesame, and oil (Segal 1983: 24:8; 41:1; 42b:1; 69a; 95b:4); the ḥufn used for flour, sesame, and salt ([?]; Segal 1983: 41:2, 68:6; 77a:2; 126:4); and the liquid mari unit used to measure quantities of beer (Segal 1983: 42a:3–4; 91:1–3). They lack the seʾah and use the ḥōmer instead of the gun or the kor (Segal 1983: 24:8; 69); oil is measured by the ḥōmer. The smaller qab unit appears once but, unfortunately, the substance being measured has not been preserved (Segal 1983: 51:10). In Babylonia, the system had been the qu (bowl = ca. 1 liter), the sutu (“vessel,” ca. 10 liters), the parsiktu = (ba-ri-ga in Sumerian; bushel = ca. 60 liters) and the gur/kurru (royal cube; ca. 300 liters, though calculated by Dandamaev as ca.180 liters). The Persian system was reportedly found in use in Babylonia under Cambyses, although the specific source is not cited and so cannot be verified (Briant 2002: 414). In Idumea, the dry and liquid measures were the qab (ca. 1.2 liters), with subunits of a quarter and an eighth, the seʾah (= 6 qab = ca. 7.2 liters), and the kor (= 30 seʾah = ca. 220 liters; Yardeni and Porten 2008: 741;

The Economy and Administration of Rural Idumea

179

for 6 qab = 1 seʾah in a commodities list from Wadi Murabbaʿat probably dating to the early 2nd century c.e., see Naveh 1981: 153). The qab and the seʾah were used to measure crushed barley/wheat, barley groats, flour, fine flour, oil, and wine, and the kor was used for large quantities of wheat and barley (e.g., E/N 13:3, 62:3, 99:2, 190:6). In the 45 Arad ostraca, which likely date from the same period as the Idumean ostraca and reflect administration in the same province, the same qab-seʾah system was used to dispense whole barley as animal feed for horses, donkeys, colts, and camels (e.g., ##1, 2, 3, 6, 12, 15, 16, 24, 25) and to dispense crushed barley (daqîr) to horsemen (e.g., ##7, 8) and crushed barley, wheat (ḥes), or flour (qomah) to individuals (e.g., ##10, 13, 28; Naveh 1981: 153–63, 175). Nothing was dispensed in the recovered ostraca in a quantity large enough to be measured in kors. However, the Beersheba ostraca include measurement of wheat by kor and seʾah (Naveh 1973: ##3, 5, 6, 13; 1979: ##29–30, 32, 47) and of barley by seʾah, qab, and peras (Naveh 1973: ##1, 2, 4; 1979: ##27–28, 39, 45, 50). They also include an undetermined measurement smaller than a seʾah that is abbreviated with a letter that resembles a Latin D, of which at least 12 constituted a seʾah (Naveh 1973: ##1, 3, 4). Thus, it was at least twice as small as a qab, six of which equaled a seʾah. The qab (2 Kgs 6:25) and seʾah are known from biblical texts (Gen 18:6; 1 Sam 25:18; 1 Kgs 18:32; 2 Kgs 7:1, 16, 18) as dry measures of grain, even if their quantities are not able to be deduced from the limited contexts of use. The kor appears as both a liquid measure (Ezek 45:14) and a dry measure (2 Chr 2:10; 27:5). It needs to be recognized, however, that none of these measures appears definitively in the monarchic-era ostraca from Samaria, Lachish, Arad, or Beer-sheba. The Samaria ostraca name products and give numerical quantities that seem to presume the number of jars in which the wine or oil was stored (Reisner 1924: 227–46). The Arad ostraca use a number of symbols before the numerical amounts. They clearly designate some form of quantification of a specified product, but there is no key that equates a given symbol with a specific quantity. In the Iron Age Hebrew ostraca from Beer-sheba, the abbreviation b occurs on ostracon 1, which probably represents a bat (Aharoni 1973: 71). But otherwise, it is not possible to confirm the names of the dry measurements to which the symbols that have been used correspond during the time of the Judahite monarchy. Naveh has assumed the terminology used in the biblical texts that are set in the monarchic era accurately reflect the measurements in use at the time, which is possible but equally may be incorrect, reflecting one or more measures in use in the Neo-Babylonian, Persian, or Hellenistic period(s), when texts were composed or updated. Subsequent practice

180

Diana Edelman

in the Persian-era ostraca was to use the first letter of a given quantity as shorthand for recording, which makes the proposed equivalencies more compelling. Assuming Naveh has correctly identified the symbols used for measurements, at Arad there are bats of wine or an unspecified substance (e.g., ##1–4, 8, 11, 61, 79), ḥōmers of flour, wine, or unspecified substance(s) (##1, 2, 8, 46), leteks of unspecified substances or wheat (##18, 33, 42, 83), ḥeḳats of barley, wheat, and an unspecified substance (##25, 32–34, 60, 76), ephahs of grain (#31), and possibly seʾahs of wheat or an unspecified substance (##33, 41; for symbol discussion, see Naveh 1981: 58). But the meaning of the symbol identified as a seʾah, as is the case with all the other symbols, has not been confirmed and may be incorrect. It appears rarely. With this caveat in mind, it can be tentatively suggested that, on some level, the monarchic-era terminology probably continued to be used in the successor provinces of Yehud and Idumea, once it was established, although the quantities or volume represented by the continuing terminology could have changed over time. But conclusive proof that this was the case is lacking at this time. It can be noted that additional dry and liquid measures that do not appear in the Idumean ostraca are used in the biblical texts: the log,2 the ḥōmer, the ephah (about a bushel or ca. 22 liters), the ʿissaron, which was one-tenth of an ephah (Lipschits et al. 2010: 457), the hin (ca. 3.89 2.  This measurement is attested in Elephantine letter AP 62, 63, 65, 66, 70, 79, 81; TAD C.3.28 and possibly also in the Saqqara ostraca, although the evidence is disputed (Segal 1983: 139, III). It is mentioned in an inscription on an alabaster jar found at Susa in the occupational level destroyed by Ashurbanipal in the 7th century b.c.e. but which likely originated as tribute sent in the 8th century b.c.e., which matches the style of writing (Renz 1995: 240–41): hn 1 wḥṣ y hlg wrbʿt hlg = 1 hin and one-half log and one-quarter log, showing it was already in use in the Iron Age. S. Aḥituv (2008: 242) suggests that the jar was booty taken from Judah and follows Naveh’s suggestion that the script points to the 7th century b.c.e. Its recovery from an 8th-century stratum provides the more definitive date, as long as the pottery used to date that level does not continue in use through a large portion of the 7th century, questioning the dating in the 8th century. No inscription from Cisjordan yet confirms the use of the log or hin in Judah or Israel in the Iron Age when either kingdom existed, in spite of the deduction of S. Mittmann (1997: 272) that the reference to “a quarter” on a flask from level II at Lachish, dating to the 7th century b.c.e., refers to one-quarter hin. His reconstruction is possible but not substantiated since the liquid volume was not specified in the inscription. It might have been a log or a hin or some other native measurement. Caution is needed, therefore, in concluding that this was a native measurement in either kingdom; it is equally possible that it was part of a measurement system introduced for administrative purposes by scribes of the Assyrian or Neo-Babylonian Empire who replaced the native system over time.

The Economy and Administration of Rural Idumea

181

liters), and the bat (ca. 19–24 liters; for the history of calculations, see Lipschits et al. 2010: 458–67), respectively. The ḥōmer is used as a measurement for grain yield in a papyrus contract from year 7 of Darius likely found at Elephantine (Bauer and Meissner 1936: 418, 423). However, it also appears in Arad Hebrew ostracon 2: “Give to the Kittiyim 1 bat and 2 of wine for the four days, and 300 of bread, and a full ḥōmer of wine” (Aḥituv 2008: 96). Thus, it probably reflects the continued use of Judean measurements by the relocated Judean mercenaries and their descendants. A kor would have comprised 10 ephahs when used to measure dry quantities, consistent with a base-ten system, and so was the equivalent of the ḥōmer, which also contained 10 ephahs as well as 10 bats (Ezek 45:11, 14).3 A recent article reinvestigating liquid volume measurement during the monarchy has concluded that the bat was the only measurement used in the monarchic period, while the log and hin were used in the Second Temple cult and, thus, were Persian-era introductions (Lipschits et al. 2010). The authors conclude that the bat was a type of storage jar that came to be mass-produced in a standard shape, with its capacity variance narrowed to within about 10%, in the late 8th century b.c.e. The jar continued in use in the 7th and early 6th centuries b.c.e. and was the jar type that bore both the lmlk and rosette stamps (Lipschits et al. 2010: 458). The bat is never found subdivided into smaller units in any of its 13 mentions in HB texts and was not used to transport liquids from one place to another (Lipschits et al. 2010: 468). How significant is the failure of the ḥōmer, ephah, hin, and bat to appear in the Idumean ostraca? The standard measurements that appear in the ostraca have exact correspondences or equivalencies in most instances to Persian measurements: 1 qab = 1 qa (ca. 1.2 liters); 5 seʾah = 1 ʾartab; 1 kor  = ca. 6.1 ʾartab. These made conversions from the local system to the imperial system easy, with the kor yielding slightly in excess of 6 artab, to the favor of the Persian administration. The ḥōmer and ephah, on the other hand, do not allow easy conversions; ca. 1.63 ephah = 1 ʾartab; ca. 16.36 ḥōmer = 1 ʾartab. For liquid measures, 10 qab = 1 marriš (ca. 12 liters); ca. 3.08 hin = 1 marriš; ca. 1.67 marriš = 1 bat (averaging 20 liters). Using the hin as a basis, the conversion rate would have favored the Persian system, which would have gained 0.08 per marriš; slightly less than the 0.10 realized when converting the kor to 3. The ḥōmer and smaller fractional griv are used as measurements for an unpreserved substance in the fragment of Saqqara Aramaic papyrus #24:8, of uncertain date but likely Persian or Hellenistic. The same tribute list includes 20 ardabs of alabaster of Syene (Segal 1983: 39–40). In #69a, oil is measured by ḥōmer and griv (Segal 1983: 92).

182

Diana Edelman

ʾartabs. The administrative representatives would have lost 0.33 had they worked with the bat as the local standard. If we acknowledge that precision was probably not a key factor in the measurement of ancient volumes, the hin, with its probable variation falling within 10% of the average corresponding Persian measurements, probably failed to appear because, measuring about one-third of a marriš, it was a nonstandard form for dispensing liquids (falling between the qab and the marriš). Similarly, the bat was larger by an additional two-thirds than the standard marriš and so apparently was not a conveniently sized jar in which to ship liquid commodities or to use to calculate Persian equivalencies. The conversion rates provide a logical explanation for the failure to continue to use the ḥōmer, ephah, and bat in the Persian province of Idumea but not for the discarding of the liquid hin measurement. The first three did not have an exact, nonfractional equivalency or one that deviated by only 0.10 maximum. Perhaps the reason for the failure to employ the hin had to do with the size of jars that came to be used as standard containers for wine, olive oil, and other forms of liquids in this era. This could profitably be checked but is beyond the scope of the current investigation. It has been suggested that, in Egypt, the Persian measurement system was imposed by Darius I (550–486 b.c.e.), which then yielded two types of ʾardabs that were used side by side. The old, native measurement of the khar “sack” that held about 80 liters of grain, which had been standard since the Old Kingdom period, was cut in half to produce a new local grain measurement called the ʾardab of ca. 40 liters, taking over a Persian term. The Ptolemies subsequently adopted the ca. 40-liter ʾardab, which consisted of ca. 80 hin, and were able to adapt their Greek choinix measure to the Egyptian hin at a rate of 2 : 1, making it the equivalent of a ḥofen. The newly created volume measurement of the ca. 40-liter ʾardab was used side by side in Egypt with the standard Persian ʾardab thought to have been introduced by Darius I, which measured ca. 30 liters in volume (= 60 hin = 30 ḥofen; Vleeming 1979: 97–99; 1981: 542–43; for a different proposal, which equates the ʾardab with the khar initially, with a noticeable diminution in value of the latter fairly quickly, see Briant 2002: 414). Thus, the Egyptian and Ptolemaic ʾardab held different volume capacities than the Persian ʾardab. One document in the Ananiah archive from Elephantine mentions a loan of emmer of 2 peras, 3 seʾahs (TAD B3.13; Brooklyn Museum Text 11), which would be repaid when the borrower received his ration from the treasury (ʾwṣr) of the king, which subsequently is referred to as the “house (byt) of the king” when describing the penalty for nonrepay-

The Economy and Administration of Rural Idumea

183

ment. The quantity or volume of a peras is unclear; rations were always dispersed in ʾardabs, so it is unknown how many peras made up an ʾardab, although it has been suggested that, since peras means “half,” two peras would have equaled an ʾardab and that quantities were often given in peras instead of ʾardabs (Porten 1968: 72).4 There were three seʾahs in an ʾardab in Egypt5 and at least four seʾahs in a peras. The loan of emmer mentioned above would have been at least four ʾardab. One document from Elephantine, AP 24, indicates that during the reign of Darius II (423–405 b.c.e.) at the fort at Syene, 22 men received a monthly ration of one ʾardab of barley, 2 received one and a half ʾardab, and 30 received two and a half ʾardab (Porten 1968: 72). The qab measure is mentioned on ostracon 35468a in the Museum of Cairo: “Prepare for me barley, one qab, and mustard grain (?), a small sack” (Aimé-Giron 1931: 6). This term also appears in one fragmentary contract, AP 45:8, where some form of payment-in-kind involving barley apparently was to be repaid at the rate of “one qab to one peras all the months and years” (Porten 1968: 72). Thus, mercenaries of Semitic descent in Egypt shared with Idumea the use of the qab and seʾah for dry measures in the Persian period. The Persian ʾartab, which was used to measure monthly food rations owed to troops and government dependents, is absent from the Idumean ostraca, as are the measures called bar and peras, which are not otherwise known in Bactria or Babylonia either. What emerges from this summary? First, since there is no reference to the ʾartab in the Idumean ostraca, the number of seʾahs that made up one ʾartab cannot be calculated. As a result, we cannot determine how standardized the equivalency systems might have been across the Empire. In Persia, Bactria, and Egypt, three seʾahs constituted an ʾartab, but did that also hold in Idumea? Interestingly, according to the Mishnah, the seʾah in Jerusalem was slightly larger than the desert seʾah, with five Jerusalemite seʾahs equaling six desert seʾahs (Yardeni and Porten 2008: 41). But that still leaves open how many Jerusalemite seʾahs would have constituted an ʾartab in the Persian period and whether the quantity shifted over time. It seems that local names of measures were used in Bactria but that they were able to be equated with the quantities and volumes of Persian equivalents. The failure for some well-known measures in the Hebrew Bible to appear in the Idumean ostraca, in a 4.  Kraeling (1953: 263) suggests the peras might have been half an omer or half an ʾardab; an ʾardab was one-tenth of an omer or 3 seʾahs. Note that the peras appears as a measure also in the Beer-sheba Aramaic ostraca (Naveh 1979: #39). 5.  One text from Elephantine refers to a “large qab,” indicating that the quantities within a named measure could vary according to which local system of measurement was being used (TAD D7.46:2; Yardeni and Porten 2008: 741).

184

Diana Edelman

territory that had once been controlled by the Kingdom of Judah, has been deduced to have been due to their inconvenient and unfavorable rate of equivalency exchange into Persian measurements. Before it is concluded, however, that all payment of taxes-in-kind and governmental transactions would have been made using Persian weights and measures and that local transactions also would have been conducted according to imperial standards, as assumed by Porten and Yardeni (2007: 135, 143, 154), the limited available evidence needs to be scrutinized carefully to ensure that the isolated evidence from Egypt, Sardis, and Babylonia is not limited to royal estates or to the delivery of official taxes-in-kind. Certainly, the evidence from Egypt is strongest concerning the imposition of Persian standards (e.g., TAD C3.14; 18; 25–28); nevertheless, the use of Persian standards to pay mercenaries in the employ of the Persian government at Elephantine is unsurprising and may represent a special situation that did not apply more generally across the native Egyptian population. The ʾardab-sized alabaster vessel dedicated by Darius I at Memphis to commemorate the death of the Apis bull in 488 b.c.e., which included the statement “72 hin” in its inscription, may well have been a political move to reinforce the theoretical introduction of Persian weights and measurements throughout the country (so, e.g., Agut-Labordère forthcoming). On a practical level, it seems to have signaled the introduction of the Persian administrative accounting system into the newly acquired territory to facilitate control of its regional economy (Henkelman 2013: 532). Yet, Aragonite vases with Egyptian hieroglyphic inscriptions have been found at Susa bearing the names of Darius, Xerxes, and Artaxerxes. Those bearing the names and regnal years of Xerxes and Artaxerxes have quadralingual inscriptions. Two give the contents in Egyptian measures, not Persian (Briant 2002: 451), which tend to emphasize the symbolic rather than practical nature of the first dedicated vessel. More evidence needs to be gathered, however, to see if all records in the country were kept using Persian weights and measures or if older systems were still employed on the local level and conversions made when and as needed. Various papyri written in Demotic indicate that temple payments and receipts continued the use of local weights and measures; quantities were recorded according to the weight/measure of a specific temple.6 All this evidence suggests, then, that local Egyp6.  I wish to thank G. Knoppers for asking his colleague, D. Redford, about how widespread the use of Persian weights and measures was in Achaemenid Egypt and for his pointing out that Demotic evidence indicates the ongoing use of local systems, even if he did not provide specific examples.

The Economy and Administration of Rural Idumea

185

tian measurements continued to be used in Egypt alongside newly introduced Persian measures. In addition, the Elephantine evidence suggests that immigrant Semitic mercenary populations tended to use their familiar measurement systems, particularly the qab and seʾah that overlapped with the Persian qa and seʾah, in combination with the Persian monthly rations system that dispensed dry foods in ʾardabs, rather than adopting any local Egyptian system. The one use of “the weight of Ptaḥ” for a loan of silver at Elephantine made by PN bar Yatmaʾ to a person whose name is lost but who received a salary from the treasury likely was made in the period between 460 and 450 b.c.e., when Egypt had successfully revolted against Persia. That rate is spelled out at 10 : 1 in the contract, which would allow for the contract to remain in effect even if the Persians regained control and the Persian weight standard came back into effect (AP 11; TAD B4.2; Cowley 1923: 32). In addition, it should be noted that the tribute system set up by Darius I made satraps responsible to the king for raising and delivering the tribute that had been set for their region. They then assigned different kings, dynasts, and governors a portion of the total amount due, who in turn divided the burden among their constituents (Briant 2002: 411, 440). M. Jursa (2011: 439–40) cites BM 42352+ as an example of a labor tax paid in silver by citizens in Sippar that financed the hiring of substitutes for a work gang that the governor of Babylon was ordered to provide to dig a canal in Elam in year 17 of Darius I. He understands the tablet to demonstrate that the extraction of taxes and labor services was accomplished in a decentralized way on a local level, once directions were issued from the highest levels of the Achaemenid administration. Bearing these examples in mind, we can see that the collection of taxes-in-kind at the local level may well have been done using local weights and measures understood by the peasants and eventually converted to Persian standards when the requisite amounts were processed as official annual tribute.

Persian Policy concerning Land Ownership, Tenancy, and Taxes It is widely agreed that the Persian administration established a cadastral land registry in each province, probably beginning in the reign of Darius and then, subsequently, as new lands were conquered (so, for example, Dandamaev and Lukonin 1989: 130; Briant 2002: 412–14). Preexisting boundaries, villages, cities, ethne, sanctuaries, or kingdoms were usually recognized by the administration, which were then used as the basis for taxation. The Great King established control, directly or indirectly, over land production and those working the land (Briant

186

Diana Edelman

2002: 419); he did not necessarily assume ownership of the land. It is also agreed, however, that the best portions of land belonged to the king, the temples, and the business houses, military elite, and civil servants of the royal and temple administration (Dandamaev and Lukonin 1989: 130). Lands could be confiscated by the king after a revolt or during war, thus ensuring that the best land came under direct royal control. Concessionary plots were taken from royal land and given/gifted to favorites or relatives of the king, or to colonists, military and nonmilitary. At Elephantine, for example, soldiers received rations in-kind, silver, and concessionary plots of land (for the latter, see AP 1:3–5, 7:2, 16:2; TAD B5.1:3–5, B7.2:2, A5.2:2; Briant 2002: 417). At Memphis, the “fields of the garrison” also appear to have been held by soldiers, who would have paid tax from their yields and may have had ḥkr agreements in place, which involved an annual rent payable in-kind, with the garrison as a corporate unit being assessed an annual tribute payment (Segal 1983: 31, #18, 51, #31; 39, #24; Briant 2002: 405, 417–18). There are 10 ostraca in Ephʿal and Naveh’s 1996 publication (##185– 95) that seem to reflect the existence of a cadastral registry in Idumea. They describe land parcels and then seem to give quantities of seed needed to sow the lands or anticipated tithes on yields of olives or fruits. While some of these refer to the holdings of individuals, two others from Lemaire’s 2002 publication detail temple holdings and apparently royal lands. Ostracon 186 is a good example of private individual holdings; it refers to Qausram’s lot and olive grove (E/N 72:2; 85:1? 120:2; 133:1; 154:3; 186:2; 194:2; L 2 4:1; 44:2; 238:3; 292:2; 310:1; L I 21:3; 58:3; 123:1); ostracon 188 mentions Barʾa’s vineyard (E/N 188:5; 271:3); ostracon 189 refers to ʿAliʾel’s orchard (E/N 69:1; 189:1; L 2 110:2; Lozachmeur and Lemaire 1996: 9:1), Samuk’s olive grove (E/N 11:2; 79:2; 150:1; 186:1; L 2 365:10; Lozachmeur and Lemaire 1996: 8:1), Binu’s vineyard, Sismay’s vineyards, and Hazaʾel’s “white” or full-sun field;7 and ostracon 190 mentions Huri’s vineyard (E/N 1:3; 3:2; 4:2; 25:2; 98:2; 137:1; 188:1; 190:5? L  2 2:3; 70:2; 176:1; 212:1; 245:3; 285:2; L  1 148:1), ʿAmi’s field, and the olive grove of rpydʾ. The widely discussed ostracon 283 in Lemaire 2002 details land located near or in conjunction with three temples to ʿUzza, Yaho, and Nabu: “the hill/ruin that is underneath/below (tahat min) the house/ Temple of ʿUzza, and the [measuring] band/cord of the house of Yaho, the marshy sterile land of Zabi, the terrace of the terebrinth, the devastated land of Saʿad/ru, the grave of Gilgul, the pond/pool of the house/ 7. In m. Šeb. 2:1 and in m. Moʿed Qaṭ. 1:4, a white field ‫ ׂשדה הלבן‬is one without shade for growing vegetables or grain; an orchard field ‫ ׂשדה האילן‬is one with shade (Ephʿal and Naveh 1996: 13).

The Economy and Administration of Rural Idumea

187

Temple of Nabu, and the grave of Yinqom” (Lemaire 2002: 149–50). These might be lands that were not able to be worked or were exempt from mandatory production and taxation. Another ostracon of great interest is ostracon 259 in the same 2002 volume, which details “the portion/division . . . lot/parcel of Seleucos son of Shabiru who is responsible for . . . from the vineyard to the pistachio tree/grove this is the house of the king and the wide place, ‫( רחובא‬rḥwbʾ), and the garden of eternity to . . . the rocky mass . . .” (Lemaire 2002: 135). This ostracon is the first indication that a royal estate possibly existed in Idumea, along with some sort of garden designated “the garden of eternity,” although it needs to be noted that the expression “house of the king” could be used to describe a governmental warehouse as well as a royal estate, as it did at Elephantine in seven instances (Porten 1968: 60; Briant 2002: 449; 461).8 Nevertheless, in this case, it may illustrate the propensity noted by Briant and Dandamaev and Lukonin for the crown to gain control of the best portions of land in a region. The likely estate features a wide or broad place in addition to the “eternal garden, possibly a grave,”9 and may also have included the vineyard and a pistachio grove. The Bactrian document that provides a provisioning list for Bayasa, also known as King Artaxerxes V, in his first year while he was in Maithanaka, passing from Bactra to Varnu (C1), allows the inference to be drawn that “white flour” (qmh. ḥwry) (C1:15, 34; C5:5; D2:2), in contrast to ordinary flour designated by the term [qmh.] damya (C1:16, 35; C3:21, 22, 38), was specifically used in the cult (for the terms, see Naveh and Shaked 2012: 34; for cultic use, see Shaked 2004: 40–41). Document C1: 37–39 specifies “for the temple of Bel, white flour 8 ʾardab, wine 15 mari” (Shaked 2004: 45–46; Naveh and Shaked 2012: 179). Lines 40–41 itemize offerings for a deity named Zyrw/Zydw, of white [flour] 3 ʾardab, wine 10 mari, while lines 42–43 itemize an offering/allotment (bagaya) for the cult tied to Vata, the Zoroastrian yazata associated with the wind and entitled to veneration: white [flour] 2 ʾardab, 2 griv, wine, 10 mari. Then, in lines 44–45 of the same document, 1 ʾardab of white flour and 3 mari of wine are designated for the yašta, the most important ceremony in the Zoroastrian cult (Shaked 2004: 46; Naveh and Shaked 2012: 36). It is only better-quality white flour that is used for cultic offerings, not the coarser grind of plain flour. It is noteworthy, however, that what has been interpreted as the finest grade of flour, smyd, is not routinely offered to the gods either; in the list, it is reserved for the king and is issued in the relatively small 8. See Cowley, AP 2:12, 14, 16; 3:13; 43:7, 10; and Kraeling 1953: 11:6. 9. For ʿolam as meaning “grave,” see Niehr 1997.

188

Diana Edelman

quantity of 3 ʾardab, 2 griv, in comparison with 63 ʾardab of white flour and 100 ʾardab of plain flour otherwise issued to the king in the Bactrian documents (Naveh and Shaked 2012: 34, 202–3). This raises the question whether this ingredient is a super finely ground flour or some other substance that has been pulverized into a powder form. It is not yet attested outside this single mention. In the Idumean ostraca, a number of references are made to quantities of a grade of flour called nšyp that was superior to ordinary flour (qmḥ; Lemaire 2002: 15).10 Whether it was being collected or dispersed, on analogy with the use of fine flour in the Bactrian correspondence for cultic offerings, it can be suggested tentatively that fine flour was required for use in the temples mentioned on ostracon 283 in the immediate region dedicated to ʿUzza, Yaho, and Nabu. The failure to mention any clan affiliations of the owners of the cadastral fields, vineyards, orchards, and olive groves is striking. This may well be significant, since the majority of the deliveries listed in the ostraca involve members of five “clans”: the sons of ʿAlbaʿal, the sons of Gir/Gur(u), the sons of Yehokal, the sons of Baʿalrim, and the sons of Qosi, and they are made after harvests, not year-round (Lemaire 2002: 230). Sometimes the individuals are said to belong to “the house” of a given clan rather than “its sons.” Perhaps this indicates that the lands not owned by clans that are detailed on the ten ostraca published by Ephʿal and Naveh discussed above were concessions from the crown to bureaucrats in the area, which might have been worked by their own personnel or rented to tenant farmers for a share of the annual yields. If so, the clan lands would have been detailed in the cadastral registry as well but would have represented preexisting allotments that probably had been reconfirmed once the province was established. The Bactrian documents mention three categories of Persian taxes: hlkʾ (A1:2, [12], 14), mndt mlkʾ (A8:2), and nhmrnytʾ, (A1:9, 11). The first is known from the book of Ezra (4:13, 20; 7:24) and from an Arshama letter (Driver 1957: no. 8:5); in the latter, it designates some kind of land tax. The satrap investigated the claim of Petosiri that an allotment of 30 ardab was owned by his father Pamun but had been abandoned after his death and orders that if the land remains either unassigned to another servant or attached to Arshama’s estate, Petosiri is to be instated as its owner and is to pay the hlkʾ tax to Arshama’s property at the same rate that had been paid by his father. This term has been linked to Akkadian ilku (Dandamaev and Lukonin 1989: 178–79, 194), which designated 10.  L 1 42:2, 128:2; L 2 7:3?, 9:3, 198:1, 300:2; E/N 3:3, 5:3, 6:2, 7:2, 26:2, 30:1, 48:2, 52:2; spelled nšp in E/N 80:2, S 3:2. The quantities vary from 2 seʾah (9:3) to 6 seʾah (198:1).

The Economy and Administration of Rural Idumea

189

personal service that landowners owed to the crown annually in the Neo-Assyrian administration (Postgate 1974: 80–81, 91, 222). However, by the Persian period, it seems to have changed in many instances, especially in Babylonia, into a monetary payment by landowners used to hire laborers to carry out work projects undertaken by the crown (Dandamaev and Lukonin 1989: 152, 178–79).11 The second term might or might not be the same as the ‫ מדה־בלו‬tax mentioned alongside the ‫ הלך‬tax in the same verses in Ezra, but the term certainly is identical to the king’s tax/tribute mentioned in Neh 5:4: middat hammelek. It has been proposed that, in the Bactrian documents at least, it designates rent owed on domains belonging to the king or satrap (Naveh and Shaked 2012: 30). In the biblical context, it appears instead to be a tribute levied on the province that was to be paid in silver, perhaps assessed per adult male or by the number in a given household.12 Many peasants would have had to convert agricultural produce into silver in order to meet the demand. In addition to liability to pay taxes-in-kind on crops and herds, smallholders were responsible for providing obligatory free labor (corvée) to the government annually and for paying various special assessments imposed on them. If we assume that the Assyrian taxation system was fairly representative of ongoing practice in the larger ancient Near East, the following types of assessments would have been obligatory on landowners: taxes on crops for human consumption from cultivated land = nusāhē and taxes on straw = šibšu, both of which normally were due at the harvest (Postgate 1974: 186–89). The straw would have been used as fodder and to make bricks (Postgate 1974: 197). Taxes were assessed as well on flocks, herds, and horses = ṣibtu. While taxes-in-kind, assessments in-kind and in silver and gold, and mandatory service owed the king could be delivered by individuals to local officials, in which case they were called pirru, the more common practice in Assyria was for administrators to go and round them up as part of a levy or requisition process, in which case, they were called bitqu (Postgate 1974: 60, 62, 93, 11.  However, this term also continued to be used in the early period to designate service duty to the crown owed by landowners, as indicated by a Babylonian contract from Nippur dated to 525 b.c.e. that divided a field among a number of individuals. There was a stipulation that each new landowner was to serve the king from their portion of land allotment (Contenau 1927: 203; Dandamaev and Lukonin 1989: 194). 12.  In Babylonia, there was a newly created tax called “transportation of payments-in-kind” (zebēlu ša upiyāta) in the Persian period, indicated by the use of a Persian term for “taxes-in-kind” (Briant 2002: 413, 440; no source citation is given other than tablets from Babylon). It may have been used more widely in the Empire.

190

Diana Edelman

106, 166). Personal service owed by landowners to the king was called ilku and tupšikku; the two are frequently paired (Postgate 1968: 9–16; 1974: 80–81, 91, 222). In addition to human levies to provide physical labor to repair buildings and metalwork or build roads, walls, dig canals, or provide military service, horses and mules were involved in the levies (bitqu) but probably not as part of the ilku service obligation (Postgate 1974: 61–62), while leather or perhaps uncured animal skins was included as a levied item (Postgate 1974: 50, 208). Tax collection was done by people bearing either the title qēpu or ša qurbūti (Postgate 1974: 194). The Bactrian correspondence indicates that the governor of Khulmi, Bagavant, had at his disposal a workforce that could be used to fix roofs on dilapidated buildings belonging to the satrap (A6), build walls and dig a moat around a fortress at Nikhshapaya (A4), or crush grasshoppers and harvest fields (A2, A4). One group of such men in question was from Herat (Hrkyn), who were to go to the “desert of Artadana” either to deliver vinegar or to remove drifting sand (A2) (Shaked 2004: 30–31; Naveh and Shaked 2012: 80–85). It remains unclear if these tasks were to be done by mercenary soldiers, who were not on campaign but still on the payroll, or by civilians fulfilling their annual corvée labor service to the crown. There are 41 Idumean ostraca that detail the supply of laborers; they do not always give the names of individuals but, rather, indicate their affiliation with one of the five clan groups, suggesting that the clan was the primary dispatcher (2006: 474). These clans would seem to be local residents of various Semitic backgrounds. Porten and Yardeni argue these are day laborers for two reasons. First, two workers, Zabdimilk of Baʿalrim and Ramel of Qoṣi, work on two successive days, Kislev 1 and 2. Second, members sent on a given day from the same clan were generally not recorded together on a single ostracon, but separately (Porten and Yardeni 2006: 474). When names are given, however, clan affiliation is usually also included; clan affiliation would not be needed if they were individual day-laborers earning for themselves, which seems to be what Porten and Yardeni intend. They later note that the term they render day-laborer, ‫פעל‬, does not appear in the Hebrew Bible with this meaning, but refers to earned wages in a single instance in the Hebrew Bible (Job 7:2), while the worker himself is designated in the same passage by the term ‫( שׂכיר‬2006: 481). While ‫ פועל‬comes to mean laborer in later Hebrew (e.g., Sir 19:1, 37:11; and the Dead Sea Scrolls), this might well have been the result of borrowing from Aramaic (so HALOT: s.v. ‫)פעל‬. In any event, the category of day-laborer needs to include those performing annual free labor owed to the crown and those who

The Economy and Administration of Rural Idumea

191

were being paid to perform such obligatory labor on behalf of others, in addition to those who were hired on an ad hoc basis by administrative representatives to perform various tasks requiring manual labor. Four ostraca (Porten and Yardeni 2006: table 3a: ##1, 2, 14, 41) list only the number of workers sent by a given clan: 11 (2006: table 3a: ##3, 5, 17, 18, 20, 25, 29, 31, 3, 39, 52) give more than one name from the same clan on a single ostraca, and 8 (2006: table 3a: ##45–51) contain a single name without clan affiliation recorded (Aydu/Iyadu x 3, Qosaz, Qoskahel, Zabdidah, Zabdiel, and PN). Of these, the latter 8 might have been paid day-workers who were being hired to work on an ad hoc basis. Those who worked in Kislev (December) may well have been used to plow and sow, as Porten and Yardeni suggest (2006: 479–80), but if so, they would be doing so on royal or governmental estate land as part of their obligatory labor service owed the crown. What about those who worked in all the other months, except Iyyar and Tebeth? The Persepolis Fortification and the Treasury archives indicate how Fars was administered. After being gathered in the districts, harvest and livestock were collected into warehouses overseen by two individuals: the manager of grains (tumara) and the authorizing agent for disbursements (ullira). Layover points along major roads were overseen by a warehouse head, which provided travel provisions to officials with proper authorization. Each year, accountants prepared an inventory of their warehouse that was sent to Persepolis. According to Briant, each “county seat” included several warehouses, each one dedicated to the collection and storage of a single produce: various grains, sesame, beer, wine, wheat, or livestock (sheep, cattle, camels, horses, and fowl). At Hadaran, in year 19 of Darius I (504 b.c.e.), for example, 2615.7 bar (ca. 14 tons) of grain were distributed under 11 categories to workers, horses, fowl, flocks, and herds, and to the royal warehouse (Hallock PF 1943). In 503 b.c.e., year 20 of Darius I, 303.5 marriš (ca. 5,500 liters) of wine were dispensed from a warehouse located at Sharamanda (Hallock PF 1954; Briant 2002: 424). At Dur, 10,685.7 bar of grain were dispensed in year 25 of Darius I (Hallock PF 1948). The Bactrian correspondence might provide additional insight into this process. In letter A6, the local governor of Khulmi, Bagavant, was under orders from the satrap of Bactria apparently, Akhvamazda, whose seat would have been Paktra/Bactra some 80 kilometers away, to deliver grain and sesame seeds to the satrap’s storage silo as part of an existing obligation (Shaked 2004: 30, 36). It is not clear if the reference to the satrap’s storage silo is to a personal facility on his private land or if it is a reference to a central facility within his administrative domain. If the latter, it would corroborate the practice of gathering foodstuffs

192

Diana Edelman

locally and then transporting them to a regional storage facility at the expense of the local governor. That facility might not be located in the satrapal seat but instead at a more convenient location within the province or in a subprovincial, regional facility. For further failure to comply, the satrap threatens the governor with the penalty of paying the entire amount of his realm/field/domain to the satrap’s realm/field/ domain/province. When we turn to the Idumean ostraca, it can be noted that more than grain is sent to and from Makkedah and its storehouse. While large quantities of wheat and barley are sent there, a sack of straw (E/N 38:2), wheat (E/N 15:2), barley groats (E/N 49:3; L 2 213:2), grgrn, perhaps dried figs (L 2 225:2),13 and possibly oil (E/N 124:2) are dispensed from Makkedah. Thus, there were either multiple storehouses around the town, each dedicated to a specific product or two, to and from which various perishables were sent and dispensed, or in this remote part of the Empire, a single warehouse could suffice to stockpile a range of foodstuffs for humans and animals. On the basis of the limited evidence available, we cannot determine how standardized administrative practices were across the Persian Empire. Two texts in Ephʿal and Naveh detail repayments on loans of wheat. One, ostracon 47, mentions 20 seʾah from the grain of the loan, the warehouse, suggesting that the loan was made by officials at the warehouse so that its repayment in kind would be sent back there (Ephʿal and Naveh 1996: 38): “La ʿadel son of Q[. . .] wheat: 20 seʾah, from the grain of the loan; the storehouse of ʿAdarbaʿal.” The other, ostracon 92, records two payments made by ʿAlqaus of the sons of Qoṣi, leyad Hazael. “On the 11th of Ab, year 7, ʿAlqaus of the sons of Qoṣi, to the hand of Ḥazael, from the loan: wheat: 10 seah; to the storehouse, wheat: 20 seah” (Ephʿal and Naveh 1996: 52). Unfortunately, these texts do not help us understand if the loans were taken out by tenants renting crown or private lands or by free-holders who had fallen short in their previous harvest and needed seed to sow for the next season. The anticipated agricultural yield was normally tenfold in a year with adequate water. In Babylonia, barley was sown at a rate of 112 liters per hectare, with an average return yield of 1,575 liters per hectare (Dandamaev and Lukonin 1989: 130). This, in theory, should have allowed sufficient surplus to pay off the loan eventually, if that were the desired objective, and still have enough to live on or sell after paying taxes or rent. 13.  So, e.g., Aḥituv (2008: 184), citing m. Peʾah 8:5. But he also notes a connection with olives in Isa 17:6 and Sir 50:10 and so a possible alternate meaning of pickled olives. Lemaire (2002: 77, 204) suggests “shriveled olives.”

The Economy and Administration of Rural Idumea

193

According to evidence gathered by Dandamaev and Lukonin (1989: 131–33), an Aramaic papyrus from Egypt, probably Elephantine, in the time of Darius I stated that the tenant of a field had to sow it with his own grain but turn over half the harvest to the lease-giver, while in Babylonia, the tenant only paid one-third of the harvest to the person who held the lease (Bauer-Meissner 1936: 414, 418). Demotic contracts in Egypt were drawn up for one year, with rents ranging from onethird to one-half of the harvest (Seidl 1968: 59–62). In Babylonia, the granting of seed grain and oxen together with the land was standard for tenants, while the use of oxen was rare in Egypt (Dandamaev and Lukonin 1989: 133). In Babylonia, on the other hand, oxen could be rented by contract. The Murashu firm rented an ox for three years for 10 kur (1,800 liters) of barley a year and four oxen with harness gear for 3 years for 150 kur of barley a year (Hil­precht and Clay 1898: 52, 89, as cited by Dandemayev and Lukonin). Whether the two loans noted in the Idumean ostraca were to freeholders or to tenants who were expected to furnish their own seed for sowing, they were taken out of the warehouse in Makkedah and most likely represent transactions between borrowers and representatives of the Persian bureaucracy rather than estate managers or middle-men, since the repayment on the first ostracon at least was made in-kind to the central warehouse. Nevertheless, it might be possible that Hazaʾel had brokered the loan as an estate manager and was having the payment in-kind credited to an account he had established at the warehouse, for which he would have been paid the conversion value in silver or gold for all foodstuffs he sent there. The failure of the phrase “balance carried forward” to appear on any of the ostraca in Idumea should not be assumed to indicate that no accounts were maintained at the storage facilities in Makkedah. It is known from the Persepolis texts and may have appeared in the official ledgers for which the ostraca served as chits that gave vital but incomplete information.

Agents of Fief or Estate Managers and of Governmental Administrators In light of the Murashu archive from Babylonia, we should perhaps consider whether any of the agents to whom the payments and deliveries were made were fief or estate managers who would have been relied upon to collect land taxes due to the crown from small-holders in addition to rents owed by locals who may have leased plots of land owned by civil servants, high court officials, warriors, or the Persian royal family.

194

Diana Edelman

Agents have been identified in the Idumean ostraca by the use of the phrase ʿal yad (Porten and Yardeni 2007: 127). According to Porten and Yardeni, payments were made via agents such as Zabdi (ISAP 900–901; Porten and Yardeni 2006: 471). They also have concluded that he was the chief administrator of the storehouse in Makkedah in the year 342 b.c.e. who calculated the amount of grain delivered by Natansidq of Qosi (L 2 82; Porten and Yardeni 2006: 471–72). However, it seems the terms do not necessarily mean “agent” in the sense of a representative of an estate management firm or an appointed tax collector. The expression ʿal yad might as easily refer to one of three types of individual in the present context. The first might be the delivery person who physically moved the goods from the supplier to the receiver. His name would be needed in case a discrepancy arose at either end in the quantity of the shipment. Second, it might be possible to understand the agent to be an official who had delivered a prior request to the supplier for a needed quantity of goods, who is now being referenced in the record of the successful delivery of the goods for future cross-referencing in the official accounts. Finally, in light of the common practice in Assyria of sending out tax agents to collect goods owed, it might be possible to suggest that the person whose name comes after lĕyad or ʿal yad was a tax collector. A more detailed study is needed of the range of people who appear in the ostraca in conjunction with the  phrases ʿal yad or lĕyad to try to determine their functions and possible status as governmental tax-collection agents or agents of private estate managers. A study also needs to be made of any potential differences in meaning between the phrases lĕyad and ʿal yad. Ephʿal and Naveh note that lĕyad can mean “to the hand of,” “to,” or “under the authority of,” while ʿal yad can mean “to,” “under the authority of,” or “at the guidance of,” so the terms seem to be interchangeable (Ephʿal and Naveh 1996: 14–15). Once the full corpus of ostraca is conveniently available in a single publication, it will be easier to undertake such a study within this particular set of texts and then comparatively among other Persian era Aramaic administrative texts. Any of the three interpretations could be a variation on the use of the “sealed document” system in Fars, which tracked all quantities transported through the supply pipeline to warehouses. After an incident in which a courier fled with a document, orders were given to include the name of the person responsible for its delivery on future tablets documenting the transport of goods (Briant 2002: 424–25). Only two ostraca out of the 1,700 refer specifically to the delivery of produce-in-kind as constituting tribute (‫ אׁשכר‬ʾaškar). “On the 10th of Kislev, year 2 [.] ʿAlqaus son of Huri brought in, as the tribute of

The Economy and Administration of Rural Idumea

195

Philip, to the hand of ʿAbdisi, oil: 3 seʾah” (Ephʿal and Naveh 1996: 54, #98). The other reads: “wheat: 23? seʾah, tribute” (Ephʿal and Naveh 1996: 78, #168). It might be significant that the delivery took place under the Ptolemaic administration and so might have reflected a change in accounting notation to include the purpose or nature of the delivery, though other ostraca also date to this period and do not designate deliveries as such. Or perhaps a single zealous scribe decided to be thorough in this instance for some reason. Given the shorthand nature of the ostraca that were intended for scribes and accountants fully versant in their jobs, the failure to indicate on the ostraca that the majority of deliveries of wheat and barley were probably payments of taxes-inkind is unsurprising and should not be seen to indicate that they served other purposes.

Evidence of Local Markets or Market Towns The suggestion by Porten and Yardeni that Makkedah represented a local market with its own storehouse, perhaps even a market town, raises an interesting possibility for which, unfortunately, there is very little evidence in the southern Levant during the preceding Iron Age or the Persian period. It is known that private trade had piggy-backed off official gift exchanges between allies and tribute payments by vassals to overlords, joining caravans that were moving items between countries and selling private goods such as fabrics, jewelry, perfumes, and cosmetics for profit along the way (Edelman 2006). But this is different from a local emporium or a local food market, where surpluses could be sold or bartered for other goods or foodstuffs. Would such local markets have employed scribes to record transactions and to collect a percentage on behalf of the crown? In a study of earlier Assyrian administration, Postgate (1974: 198) proposed that the taxes on foodstuff-in-kind and straw would not have been transported to a provincial capital for deposit but instead would have been stored more locally, in provincial towns in granaries located there. Such towns would have been fortified where necessary and would also have served as local centers of government, with an official who administered the local supplies according to central requisitions and requirements. Briant argues that Persian control over local regional economies involved the establishment of a monopoly over grain reserves (Briant 2002: 453–54). In support, he cites an Athenian decree, the date of which is broken away but that is thought to date to the mid-4th century b.c.e. (349/348?). The Athenian Council honors the satrap Orontes from Mysia in Asia Minor and bestows upon him Athenian citizenship because

196

Diana Edelman

“he is a good man toward the People . . . both now and in former times” (IG II2: 207, lines 5–6; translation cited is by S. Lambert, Attic Inscriptions Online). The stele goes on to specify that the honor was bestowed because Orontes had sold wheat to feed Athenian troops fighting in the straits. “[T]he money for the receipt of the grain shall be paid from the military fund . . . to the army encamped with Chares and Charidemos and Phokion” (lines 28–29). This event on its own does not confirm, however, that there was a Persian monopoly on surplus grain stored in Persian royal granaries or a policy prohibiting local markets. This is an inference he has drawn, rightly or wrongly. Briant then cites a document from western Asia Minor from the last quarter of the 4th century b.c.e. in the early Hellenistic period, which appears in Welles as 3:80–85 (ca. 303 b.c.e.; Welles 1934: 19, 22–23; Briant 2001: 453) probably as proof of the continuation of Persian policy under the Seleucids. In the latter inscription, ambassadors from the town of Lebedos had requested permission from Antigonus the OneEyed to maintain wheat reserves themselves in their town, allowing individuals to borrow from a money reserve set up to import grain because a shortfall was anticipated in light of the growing population. The merchants could then resell it at a profit throughout the year, returning the capital with the interest accrued to the fund at the end of each year. Antigonus states on this stele that in the past “we were un[willing] that [any] city should undertake the importation of grain or maintain a (subsidized) grain supply.” He then notes, “for crown [land] is near [and if a need] of grain arose, we think there could easily be brought from [there whatever] one wished.” His reluctance to change the past procedures is explained away on two counts: “[for we were not willing to have the] cities spend for this purpose large sums of money unnecessarily” and “[o]ur anxiety on this point was due [to a desire] to serve the [cities], since you and everyone else [knows that there is] no private profit [for us] in the business, but we maintain the regulation [in the hope that] the cities may become free from debt.”14 The two objections were hollow since grain would be needed and must be acquired from somewhere, and the proposed plan actually provides annual income for the city from the interest paid back alongside the returned principle at the end of each year. So any existing city debt could be cleared more quickly. The claim there is no private profit in selling grain from the nearby crown land is false; the crown would benefit from the sale. 14.  I am indebted to Professor Lynette Mitchell, Department of Classics, University of Exeter, for the translation.

The Economy and Administration of Rural Idumea

197

In the end, permission was granted, specifying that duty was to be paid to the market commissioner on all grain that was sold, whether imported into the city and sold in the marketplace or sold directly by local farmers to others. Thus, the crown ultimately stood to benefit in two ways from this reversal of former policy: from selling some of its surplus grain to importers and by collecting duties on grain sold from private sources. According to this document, then, towns in Asia Minor had been required to buy needed grain from the administrators of the royal granaries exclusively and were not allowed to stockpile independently for “personal” use or to sell off at a tidy profit such communal reserves to others in need in the Empire or outside it. Briant’s argument that this policy had remained in place from the previous Persian administration is not air-tight but is cogent, in spite of the possibility that the Seleucid administration could have altered earlier Persian policy by introducing a ban on local markets that they subsequently reversed. W. F. M. Henkelman (2005: 153 and n. 35) builds on Briant’s deduced lack of free markets as a deliberate Persian imperial policy. He cites the Fortification texts from Persepolis as evidence that the local food and meat producers had fixed prices dictated by the central administration in Persepolis on any trade they undertook, effectively imposing a trade monopoly on its hinterland that was regulated as exchange. The rate was 3 : 1 for barley exchanged for sawur wine and 30 quarts of wine for one sheep (Henkelman 2005: 145–48, 30 marriš = 1 sheep; Hallock 1969: 17). Again working with the Fortification texts, Henkelman focuses on a group that involves a fixed exchange rate of 100 quarts of grain to one sheep or goat (10 bar = 1 sheep or goat; Hallock 1969: 17; and PF 278, 352, 363, 364, 588) destined for sacrifice and then suggests that a similar kind of Persian monopoly regulated dealings with neighboring pastoralist tribes, such as the Ouxians and Elymaians, with the exchange rates set by the Persian authorities (Henkelman 2005: 153–56, 163–64). He admits that the animals could have come from internal royal herds so that the exchange would have been one on paper, but he interprets the annual gift-giving by the Persian king to the leaders of these groups as a move intended to formalize relations that included the ability to engage in exchanges/“trade” in the border region between Fars and Khuzestan (Henkelman 2005: 160–61, 163–64). While policy could have varied in different regions in the Persian Empire, it seems likely that the same policy concerning profits to be made on sales from crown monopolies on surplus grain stockpiled in Persian administrative facilities would have been in effect on the southern border of the Empire. This was an area of marginal rainfall, where crop failure or low annual yields would have provided a steady stream

198

Diana Edelman

of local customers for stockpiled grain, as would have the Arabian caravans traversing the Beer-sheba Valley en route to Gaza. A desire to maximize imperial income likely would have led to the use of this policy throughout the Empire. In light of this evidence and likely Persian fiscal policy, I think it is highly unlikely that a municipal or private warehouse existed in Makkedah or any other town in Idumea. It is unclear whether the official Persian warehouse in Makkedah would have “bought” or engaged in exchange to acquire surplus grain from locals or not. In theory, the practice of creating reserves throughout the Empire that could be sold by the crown as needed to citizens or to foreign groups might have resulted in the purchase of surplus grain to augment the taxes-in-kind owed annually to the crown by all landholders. In addition, the creation of provision dumps in Idumea for militia on the march would have been part of the logistical strategy to reconquer Egypt. If any of this were the case, however, it is odd that no price has been included on the ostraca if any of the grain deliveries were sales of surplus grain. One might have expected a price, since there would only have been a single buyer in every instance who would have fixed the rate. But perhaps it would have been fixed, as the examples cited by Henkelman above indicate was the practice in the heartland, and so calculated subsequently and recorded in the official ledger, with the silver, either in coin or by weight, sent to the supplier or collected subsequently. The absence of any indication of accounting balances, such as the Aramaic phrase equivalent to Elamite šutur daka, “balance carried forward,” also might be coincidental but might equally tend to indicate that the practice at Makkedah was not to set up accounts for individuals or clan groups that could be credited and drawn against, as needed.

Conclusion The Idumean ostraca can provide insight into the routine workings of administrative officials who collected and dispensed primarily foodstuffs for animal and human consumption. The evidence provided by the ostraca raises tantalizing possibilities concerning different forms of landholdings in the region, which included freeholding traditional clans, possibly concessionary estates and farms, and a possible royal estate. These ostraca seem to indicate the collection and delivery of goods from producers in some cases as well as their direct delivery by producers in others to a single, governmental warehouse in Makkadeh, although a collection of more specialized warehouses dedicated to specific types of foodstuffs, all located in Makkedah, cannot be ruled out. The use of tax collection agents is likely indicated by the phrases lĕyad or ʿal yad, although some agents of estate managers might also be involved

The Economy and Administration of Rural Idumea

199

as third parties. Taxes-in-kind were levied on barley, wheat, and straw and measured using a local system that probably carried on after the end of the Judahite monarchy, even though definitive proof is lacking, but one that did not correspond capacity-wise with Persian volumes, except in the case of the qab and qa. A careful look at the quantities of flour, barley groats, oil, and wine dispensed might indicate the presence of governmental personnel entitled to daily, weekly, or monthly rations, but most of the legible ostraca deal with the intake of goods rather than their disbursement, so more can be learned about the latter practices in the texts from Persepolis, for example. No doubt, once the entire corpus is published in a single, convenient study, more information of interest concerning the economy and administrative systems used in Idumea in the late Persian and early Hellenistic periods will be extracted. This cursory, preliminary study has hopefully pointed to potential fruitful avenues of comparison and investigation.

Bibliography Agut-Labordère, D. forthcoming  Political Piety-for-Predation Policy: Persian and Egyptian Gods during the Achaemenid Period. In Religion in the Persian Period: Emerging Judaisms and Other Trends, ed. D. V. Edelman, A. Fitz­ patrick-McKinley, and P. Guillaume. Orientalische Religionen in der Antike. Tubingen: Mohr Siebeck. Aharoni, Y. 1973 The Hebrew Inscriptions. Pp.  71–78 in Beer-Sheba I: Excavations at Tel Beer-Sheba (1969–1971 Seasons), ed. Y. Aharoni. Tel Aviv: Tel Aviv University Press. Aḥituv, S. 2008 Echoes from the Past: Hebrew and Cognate Inscriptions from the Biblical Period. Jerusalem: Carta. Aimé-Giron, M. N. 1931 Textes araméens d’Égypte. Services des Antiquitués de l’Egypt. Cairo: Institut Français d’Archéologie Orientale. Atkinson, K. T., and Atkinson, K. M. T. 1972 A Hellenistic Land-Conveyance: The Estate of Mnesimachus in the Plain of Sardis. Historia 21: 45–72. Bauer, H., and Meissner, B. 1936 Ein aramäischer Pachtvertrag aus dem 7. Jahre Darius’ I. Pp. 414–24 + 2 plates in Sitzungsberichte der Preussischen Akademie der Wissenschaft, Philisophisch-historische Klasse. Berlin: Akademie der Wissenschaften. Briant, P. 2002 From Cyrus to Alexander: A History of the Persian Empire, trans. P.  T. Daniels. Winona Lake, IN: Eisenbrauns.

200

Diana Edelman

Contenau, G. 1927 Contrats néo-babyloniens de Teglathphalasar III à Nabonide. Paris: Louvre Museum. Cowley, A. E. 1923 Aramaic Papyri of the Fifth Century b.c.: Edited, with Translation and Notes. Oxford: Clarendon. Repr., Osnabruck: Zeller, 1967. Dandamaev, M., and Lukonin, V. 1989 The Culture and Social Institutions of Ancient Iran, trans. P. L. Kohl and D. J. Dadson. Cambridge: Cambridge University Press. Donbaz, V., and Stolper, M. W. 1997 Istanbul Murašû Texts. Uitgaven van het Nederlands HistorischAchaeologisch Instituut te Istanbul 79. Leiden: Nederlands Historisch-Archaeologisch Instituut te Istanbul. Driver, G. R. 1954 Aramaic Documents of the Fifth Century b.c. Oxford: Clarendon. Dušek, J. 2007 Les manuscrits araméens du Wadi Daliyeh et la Samarie vers 450–332 av. J.-C. CHANE 30. Leiden: Brill. Edelman, D. V. 2005 The Origins of the ’Second’ Temple: Persian Imperial Policies and the Rebuilding of Jerusalem. BibleWorld. London: Equinox. 2006 Tyrian Trade in Yehud under Artaxerxes I: Real or Fictional? Independent or Crown-Endorsed? Pp. 207–46 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and J. Blenkinsopp. Winona Lake, IN: Eisenbrauns. Ephʿal, I., and Naveh, J. 1996 Aramaic Ostraca of the Fourth Century bc from Idumaea. Jerusalem: Israel Exploration Society / Magnes. Geus, C. H. J. de 1979–80 Idumaea. Jaarbericht Vooraziatisch-Egyptisch Gezelschap ‘Ex Oriente Lux’ 26: 53–74. Gropp, D. M. 1986 The Samaria Papyri from Wâdī ed-Dâliyeh: The Slave Sales. Ph.D. dissertation. Harvard University. 2001 Wadi Daliyeh II: The Samaria Papyri from Wadi Daliyeh. DJD 28. Oxford: Clarendon. Guillaume, P. 2012 Land, Credit and Crisis: Agrarian Finance in the Hebrew Bible. Bible World. London: Equinox. Hallock, R. T. 1969 Persepolis Fortification Tablets. OIP 92. Chicago: University of Chicago Press. Henkelman, W. F. M. 2005 Animal sacrifice and “External” Exchange in the Persepolis Fortification Tablets. Pp. 137–65 in Approaching the Babylonian Economy: Proceedings of the START Project Symposium Held in Vienna, 1–3 July 2004,

The Economy and Administration of Rural Idumea

201

ed. H. D Baker and M. Jursa. Studies in the Economic History of First Millennium Babylonian 2. AOAT 30. Munster: Ugarit-Verlag. 2008 The Other Gods Who Are: Studies in Elamite-Iranian Acculturation Based on the Persepolis Fortification Texts. Achaemenid History 14. Leiden: Nederlands Instituut voor het Nabije Oosten. 2013 Administrative Realities: The Persepolis Archives and the Archaeology of the Achaemenid Heartland. Pp. 528–46 in The Oxford Handbook of Ancient Iran, ed. D. T. Potts. Oxford: Oxford University Press. Hilprecht, H. V., and Clay, A. T. 1898 Business Documents of Murashû Sons of Nippur Dated in the Reign of Artaxerxes I (464–424 b.c.). The Babylonian Expedition of the University of Pennsylvania: Series A, Cuneiform Texts 9. Philadelphia: University of Pennsylvania Press. Hinz, W., and Koch, H. 1987 Elamisches Worterbuch, vol. 1: A–H; vol. 2: I–Z. Berlin: Reimer. Jursa, M. 2011 Taxation and Service Obligations in Babylonia from Nebuchadnezzar to Darius and the Evidence for Darius’ Tax Reform. Pp. 431–48 in Herodot und das Persische Weltreich: Akten des Internationalen Kolloquiums zum Thema “Vorderasien im Spannungsfeld klassischer und alt­ orientalischer Überlieferungen,” Innsbruck, 24.–28. November 2008, ed. R. Rollinger, B. Truschnegg, and R. Bichler. Wiesbaden: Harrassowitz. Knauf, E. A., and Guillaume, P. 2014 A History of Biblical Israel. Worlds of the Ancient Near East and Mediterranean 1. Sheffield: Equinox. Kraeling, E. G. 1953 The Brooklyn Museum Aramaic Papyri: New Documents of the Fifth Century b.c. from the Jewish Colony at Elephantine. New Haven, CT: Yale University Press. Lambert, S. n.d. Attic inscriptions online: Honours for Orontos. https://www .atticinstriptions.com/instription/IGII31/295.html. Lemaire, A. 1996 Nouvelles inscriptions araméennes d’Idumée au musée d’Israël. Transeuphratène Supplement 3. Paris: Gabalda. 2002 Nouvelles inscriptions araméennes d’Idumée Tome II. Transeuphratène Supplement 9. Paris: Gabalda. 2003 Nabonidus in Arabia and Judah in the Neo-Babylonian Period. Pp. 285–98 in Judah and the Judeans in the Neo-Babylonian Period, ed. O. Lipschits and J. Blenkinsopp. Winona Lake, IN: Eisenbrauns. 2006 New Aramaic Ostraca from Idumea and Their Historical Interpretation. Pp.  413–56 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. Lipschits, O.; Koch, I.; Shaus, A.; and Guil, S. 2012 The Enigma of the Biblical Bath, and the System of Liquid Volume Measurement during the First Temple Period. UF 42 (2011): 453–78.

202

Diana Edelman

Lozachmeur, H., and Lemaire, A. 1996 Nouveaux ostraca araméens d’Idumée (Collection S. Moussaieff). Semitica 46: 123-52. Mittmann, S. 1997 Sakraler Wein und die Flüssigmasse Hin und Log. Pp.  269–80 in Ana Śadî Labnāni lū Allik: Festschrift für W. Röllig, ed. B. PongratzLeisten, H.  Kühne, and P. Xella. AOAT 247. Neukirchen-Vluyn: Neukichener Verlag. Naveh, J. 1973 The Aramaic Ostraca. Pp.  79–82 in Beer-Sheba I: Excavations at Tel Beer-Sheba (1969–1971 Seasons), ed. Y. Aharoni. Tel Aviv: Tel Aviv University Press. 1979 The Aramaic Ostraca from Tel Beer-Sheba (Seasons 1971–1976). TA 6: 182–98. 1981 The Aramaic Ostraca from Tel Arad. Pp. 153–76 in Arad Inscriptions, trans. J. Ben-Or, ed. A. Rainey. Jerusalem: Israel Exploration Society. 1992 Aramaic Ostraca and Jar Inscriptions from Tell Jemmeh. ʿAtiqot 21: 49–53. Naveh, J., and Shaul, S. 2012 Aramaic Documents from Ancient Bactria (Fourth Century b.c.e.) from the Khalili Collections. London: The Khalili Family Trust. Niehr, H. 1997 Zur Semantik von nordwestsemitischen ʿlm als ‘Unterwelt’ und ‘Grab.’ Pp. 295–305 in Ana Śadî Labnāni lū Allik: Festschrift für W. Röllig, ed. B. Pongratz-Leisten, H. Kühne, and P. Xella. AOAT 247. Neukirchen-Vluyn: Neukichener Verlag. Porten, B. 1968 Archives from Elephantine: The Life of the Ancient Jewish Military Colony. Berkeley: University of California Press. Porten, B., and Yardeni, A. 1986–99  Textbook of Aramaic Documents from Ancient Egypt. 4 vols. Jerusalem: Academon. 2006 Social, Economic and Onomastic Issues in the Aramaic Ostraca of the Fourth Century b.c.e. Pp. 457–88 in Judah and the Judeans in the Persian Period, ed. O. Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns. 2007 Makkedah and the Storehouse in the Idumean Ostraca. Pp. 125–70 in A Time of Change: Judah and Its Neighbours in the Persian and Early Hellenistic Periods, ed. Y. Levin. LSTS 65. London: T. & T. Clark. 2008 The Chronology of the Idumean Ostraca in the Decade or So after the Death of Alexander the Great and Its Relevance for Historical Events. Pp. 237–49 in Treasures on Camels’ Humps: Historical and Literary Studies from the Ancient Near East Presented to Israel Ephʿal, ed. M. Cogan and D. Kahn. Jerusalem: Magnes. 2009 Dating by Grouping in the Idumean Ostraca: Six Commodity Dossiers Dating to the Transition Years from Artaxerxes II to Artaxerxes III. ErIsr 29: 144–83.

The Economy and Administration of Rural Idumea 2014

203

Textbook of Aramaic Ostraca from Idumea, vol. 1: Dossiers 1–10: 401 Commodity Chits. Winona Lake, IN: Eisenbrauns. forthcoming  Textbook of Aramaic Ostraca from Idumea, vol. 2: Dossiers 11–50: 265 Commodity Chits. Winona Lake, IN: Eisenbrauns. Postgate, J. N. 1969 Neo-Assyrian Royal Grants and Decrees. Studia Pohl 1. Rome: Pontifical Biblical Institute. 1974 Taxation and Conscription in the Assyrian Empire. Studia Pohl 3. Rome: Pontifical Biblical Institute. Reisner, G. A. 1924 Inscriptions. Pp. 227–51 in Harvard Excavations at Samaria 1908–1910, vol.  1: Text, ed. G. A. Reisner, C. S. Fisher, and D. G. Lyon. HSS  1. Cambridge: Harvard University Press. Renz, J. 1995 Inschriften des letzten Viertels des 8. Jhdts. Pp. 145–241 in Handbuch der althebräischen Epigraphik, vols. 1–2, ed. J. Renz and W.  Röllig. Darmstadt: Wissenschaftliche Buchgesellschaft. Sapin, J. 2004 La frontièrejudéo-iduméene à IVe s. avant J.-C. Transeuphratène 27: 109–54. Segal, J. B. 1983 Aramaic Texts from North Saqqâra with Some Fragments in Phoenician. Texts from Excavations 6; Excavations at North Saqqâra Documentary Series 4. London: Egypt Exploration Society. Seidl, E. 1968 Ägyptische Rechtsgechichte der Saiten- und Perserseit. 2nd rev.  ed. Ägyptische Forschungen 20. Glückstadt: Augustin. Shaked, S. 2004 Le satrape de Bactriane et son gouverneur: Documents araméens du IVe s. avant notre ère provenant de Bactriane. Persika 4. Paris: de Boccard. Stolper, M. W. 1985 Entrepreneurs and Empire: The Murašû Archive, the Murašû Firm, and Persian Rule in Babylonia. Uitgaven van het Nederlands HistorischAchaeologisch Instituut te Istanbul 54. Leiden: Nederlands Historisch-Archaeologisch Instituut te Istanbul. Vleeming, S. P. 1979 Some Notes on the Artabe in Pathyris. Enchoria 9: 93–100. 1981 The Artaba, and Egyptian Grain-Measures. Pp. 537–45 in Proceedings of the Sixteenth International Congress of Papyrology: New York, 24–31 July 1980, ed. R. S. Bagnall et al. American Studies in Papyrology 23. Chico, CA: Scholars Press. Welles, C. B. 1934 Royal Correspondence in the Hellenistic Period: A Study in Greek Epigraphy. New Haven, CT: Yale University Press.

Yardeni, A., and Porten, B. 2007 The House of Baalrim in the Idumean Ostraca. Pp.  99–147 in New Seals and Inscriptions: Hebrew, Idumean, and Cuneiform, ed. M.  Lubetski. Sheffield: Sheffield Phoenix. 2008 Two Become One: A Unique Memorandum of Obligation. Pp. 733– 51 in Birkat Shalom: Studies in the Bible, Ancient Near Eastern Literature, and Postbiblical Judaism Presented to Shalom M. Paul on the Occasion of His Seventieth Birthday, ed. C. Cohen et al. 2 vols. Winona Lake, IN: Eisenbrauns.

Forces of Decline and Regeneration: A Socioeconomic Account of the Iron Age II Negev Desert Yifat Thareani

Hebrew Union College, Jerusalem

The Iron Age IIb–c (ca. 750–586 b.c.e.) was the first period in which the Negev Desert frontier saw the growth and decline of a flourishing settlement system. These processes were closely tied to the rise of the Neo-Assyrian Empire, which marked the beginning of a new political reality in the southern Levant. This was a period that saw both imperial and local state interests stimulating the development of large-scale urban systems in areas not previously settled. This was also the era in which the foundations were laid for subsequent southern Levantine desert frontier control strategies. The current study is an attempt to outline the sociopolitical and economic factors at play in this critical moment of Negev history, to examine external and internal forces shaping life on the desert frontier, and to place these in their broader historic-cultural context.

Forces of Regeneration If we assume that the development and intensification of Negev settlement in antiquity was not simply the result of improved climatic conditions (for a broad discussion and references, see Thareani 2011b), other factors—political, economic, or social—are responsible for the major cultural changes that took place in this arid region during Iron Age IIb. To isolate the cultural mechanisms that created and stimulated this process, we should briefly review the settlement history of the Negev in the period under discussion. Settlement History of the Iron Age IIb–c Negev During the period between the 8th and early 6th centuries b.c.e., the Negev saw unprecedented settlement florescence (fig. 1). Of the 50 late Iron Age sites documented in the Negev, 10 were large. Some of these were central sites: Tel ʿIra, Tel Malhata, Tel ʿAroer, and Beer es-Sebaʿ. Five can be classified as forts: Ḥorvat Radum, Ḥorvat ʿUza, Tel Arad, 207

208

Yifat Thareani

Fig. 1.  Map of Iron Age II Negev settlement system.

Ḥorvat ʿAnim, and Ḥorvat Tov. One was an administrative center, Tel Beer-sheba. Forty small, unfortified sites have also been identified in the region (Thareani-Sussely 2007b). This expansion in Negev settlement is usually attributed to a dramatic change in the status of the region during the Iron Age IIb. The advent of the Assyrian Empire—through conquest, annexation, and submission of the Levantine kingdoms— was followed by a period of political stability and economic prosperity commonly referred to as the pax Assyriaca (Tadmor 1975). The Assyrian conquest opened the way for new economic opportunities. The reduction in the scale and intensity of conflicts between Judah and its neighbors, all subordinated to Assyria, stimulated the development of long-distance trade relations and of settlement systems in frontier zones such as the Negev. The position of the Assyrian Empire also had a crucial impact on the sense of safety and economic prosperity of adjacent regions, some of which had never been permanently settled. Under imperial patronage, client kingdoms such as Judah could develop large hierarchic settlement systems and become integrated in long-distance trade networks. The cause for the growing commercial activity reflected in the archaeological record of this period should be sought in the growing demand in regional markets for products and goods transferable from Arabia. Situated between Arabia and Transjordan to the southeast and east, the coastline with its ports to the west, Palestine to the north, and

Socioeconomic Account of the Iron Age II Negev Desert

209

Fig. 2.  Map of the Arabian trade system.

Egypt and the desert to the south, the Negev played an important role in transferring Arabian products and goods to the coast and farther north (see fig.  2; Naʾaman 1987; Finkelstein 1992; 1995; Singer-Avitz 1999; Thareani-Sussely 2007a). Economic Resources and Their Impact on Urbanization Processes in the Negev The economic links that developed over time between the Negev and adjacent regions became a dominant factor in the urbanization processes stimulated by governmental economic interests in the desert frontier. Eventually, this development culminated in the adoption of sedentism on the part of local seminomadic groups and a change in the local subsistence pattern. While most scholars agree that climatic conditions in the ancient Negev resembled those of the modern era (Liphschitz and Waisel 1973a; Rubin 1989; 1996; Finkelstein 1995: 32–35, 155–57), the carrying capacity and agricultural potential of Iron Age Negev soils is a subject of great debate. Several archaeologists have limited the evidence for Iron Age IIb–c trade contacts with Arabia to a few luxury items, claiming that settlement growth at the end of the Iron Age Negev cannot be explained

210

Yifat Thareani

solely in terms of the Arabian trade system (Finkelstein 1994: 175–81; Faust and Weiss 2005: 74–75; Katz 2008: 29). Rather, in these researchers’ view, the growing prosperity of the desert frontier settlements was largely due to an increase in Judean demand for grain during the 7th century b.c.e.—itself a direct result of the critical loss of Judah’s fertile Shephelah lands during Sennacherib’s 701 b.c.e. campaign. Calculative studies have put forward an average annual Negev production figure of approximately 5,000 tons of grain (wheat and barley) during this period (Herzog 1994: 126–27; Finkelstein 1994: 177). This argument suffers from several difficulties. Most importantly, we do not have sufficient data to calculate the carrying capacity of Negev lands (Rubin 1990: 186). However, we do know that the structure of the loess soil that characterizes the Negev was not capable of supporting a large-scale agricultural crop (Yair 2005: 109–10). Given this, the types of storehouses and granaries that have been discovered at Negev sites—the same types found elsewhere in the southern Levant—might best be interpreted as storage facilities for grain and surpluses collected in other regions and redistributed to the Negev by the Judean administration. Moreover, archaeological excavations and surveys clearly show that settlement growth in the Negev had already begun earlier in the 8th century, prior to Sennacherib’s campaign and when the Shephelah settlement system was still functioning (Thareani-Sussely 2002: 144–45; 2007b: 73–74; Finkelstein and Naʾaman 2004; Faust 2008). Thus, crop distress in 7th-century b.c.e. Judah should not be held up as the motivator of Negev settlement growth. In sum, it is doubtful that the Negev’s climate has changed significantly in the last 5,000 years or that its agricultural potential has ever constituted the main reason for political powers’ economic interest in that region. The cause for growing settlement of the Negev should be sought instead in the intersection of internal and external developments that took place in Judah during the 8th century b.c.e. Put simply, the focus of attraction to the Negev was its function as a crossroads and link between different geographical zones. Archaeological finds and historical documents reflecting various periods in the Negev attest a wide range of products and goods there originating both in neighboring regions and farther afield: bitumen and salt from the Dead Sea, copper from Timna and Faynan, and vessels and other objects imported from distant lands such as Arabia, Mesopotamia, Phoenicia, and the Aegean, along with agricultural products brought from northern Palestine (Singer-Avitz 1999; Thareani 2011b: 275–76, 285–86, 314 with additional sources). Given this evidence, we can safely conclude that the most important economic resource of the Negev Desert frontier was its strategic location and role as a political

Socioeconomic Account of the Iron Age II Negev Desert

211

and economic intermediary in the interregional exchange system. This frontier’s function as a transitory zone at the very limit of multiple settled regions, the importance of land trade in antiquity, and the Negev’s position as a buffer zone between various political entities all contributed to and intensified its role as intermediary. The Economic System of the Iron Age IIb–c Negev Desert Frontier A multifaceted economic pattern based on agriculture, trade, and trade services is reflected through the archaeological finds and historical evidence of Iron Age IIb–c Negev. In their seminal article, Bienkowski and van der Steen (2001: 31–33) outlined several income sources that the seminomads of this region exploited in the preindustrial era. Examination of the archaeological finds from several Iron Age IIb–c Negev sites suggests similarities—with several reservations and updates. Pastoralism.  Archaeozoological analysis clearly shows that sheep and goats formed the main component in the faunal assemblage of Negev sites in all periods (Dayan 1999: 483; Motro 2011: 268, 274–78). Secondary products such as milk and wool were also of great importance in the local subsistence economy. Nineteenth-century accounts indicate that the Negev Bedouin practiced pastoralism and that their flocks included sheep, goats, and camels (Burckhardt 1822: 402; Marx 1967: 10). The prioritization of sheep and goats over cattle is explained by the relatively high cost of cattle husbandry and its need for plentiful water. Based on his analysis of the faunal assemblages at Tel Beer-sheba (Stratum II), Sasson (2004: 142–44) argued that the subsistence strategy was based on an autarchic economy and optimal use of animal products—a perception that ignores matters of context and other economic evidence found in Beer-sheba itself, as well as in other Iron Age II Negev sites. Liphschitz and Waisel (1973a: 100, 103), for instance, identified burned unipherus beams in Stratum II at Beer-sheba to be evidence for a relatively high economic potential. It appears that various objects from Beer-sheba (Singer-Avitz 1999) and ʿAroer (Thareani 2011a) present a more balanced picture and suggest an open economy in the Iron Age IIb–c Negev. Agriculture.  Even though agriculture was never the main economic resource in the Negev, we should not rule out the possibility that this subsistence strategy was practiced for purposes of self-consumption. The dietary staples of the local inhabitants were typical of other regions in Palestine and included wheat, barley, olives, dates, and lettuce (Liphschitz and Waisel 1973a: 103; 1973b: 31–35, tables 1–2). That said, travelers in the region during the Late Ottoman period noted that agriculture was mainly focused on the vicinities of Gaza and Hebron,

212

Yifat Thareani

while only a few fields were kept around Beer-sheba, and their yield was limited to a size that satisfied the self-consumption needs of the local Bedouin families (van de Velde 1854: 160–62). An issue relevant to both economic and cultural considerations is the presence or absence of pig bones. Most faunal assemblages from Iron Age II sites in Judah lack pig bones (Hesse 1990; Dayan 1999: 484–85, table 16.2; Tcherov and Drori 1983; Sapir-Hen et al. 2013). This dearth is often explained as a result of the Judahite taboo against the consumption of this animal. However, pig bones—many bearing butchery marks—comprise 2.5% of the faunal remains from Tel ʿAroer (Motro 2011: 268, 278, table 3.60). Taken in the context of the diverse material culture from Iron Age II Negev sites (see next section below), these pig bones imply a varied population that included a non-Judean element. Trade.  From the 8th century b.c.e. on, faunal evidence from the Negev shows an increase in the number of camel bones (Wapnish 1981; Dayan 1999: 480–87; Kolska Horwits 1999: 488–94; Jasmin 2006: 147; Motro 2011: 268, 276–77, table 3.60). The appearance of the camel— an animal whose breeding conditions and maintenance are relatively expensive—is associated with the integration of the Negev into the framework of the Arabian trade system (Finkelstein 1992). The extent and form of trade in and through the Negev was dictated to a large extent by its proximity to important trade routes, which resulted in local societies developing a variety of trade-related occupations—e.g., tribal elites, traders, tax collectors, caravaneers—who were active agents in the organization of the trade network. Trade services developed in and outside towns: trade quarters such as the extramural neighborhood at ʿAroer (Thareani-Sussely 2008), commercial centers such as the store avenue at Mampsis and the open spaces at Shivta (Shereshevski 1991: 63), caravansaries such as those at ʿAroer (Thareani-Sussely 2007a) and Mampsis (Negev 1988: 191–94), and markets such as that of the Bedouin in Beer-sheba (Ben-David 1990: 191). This infrastructure provided food and shelter for trade caravans passing through the Negev and enabled the concentration of products and goods in towns, where they could be sold or transported onward to Judah, Egypt, the Transjordan, or the coastal ports. Local Traditions and New Styles in Iron Age IIb–c Negev Sites The material culture from Iron Age IIb–c Negev sites reflects the preservation of local traditions and repertoires unique to this region. At the same time, assemblages often include imports and imitations of non-local types, such as the “Assyrian Ware.” Judean Repertoire.  The tendency toward preservation is reflected in material culture forms shared with other regions of Judah. We see

Socioeconomic Account of the Iron Age II Negev Desert

213

this in architecture, the weight system, the use of Judean names and script, Judean Pillar Figurines, and the large amount of ceramics, which is typical of unequivocally Judean sites (Stern 2001: 151–63). Moreover, the fact that 35 Negev settlements are listed among the towns of the tribe of Judah (Josh 15:21–32) attests that the area was included within the Judean Kingdom by the time this text was composed in the 7th century b.c.e. (Alt 1925; Naʾaman 2005: 345–47). Edomite Repertoire.  So-called Edomite Ware is also evident in Negev assemblages. These vessels are characterized by distinctive forms, some of which were painted black, white, brown, or red. Although this pottery is known mainly from Edomite sites in Transjordan, it is also found in smaller quantities in the Iron Age II Negev, Arava, and Sinai, at sites such as: ʿAroer (Thareani 2011: 120–57), Qitmit (Freud and BeitArieh 1995: figs. 4.2:17; 4.4:11; 4.14:12; 4.17:37; 4.23), Beer-sheba (SingerAvitz 2004), Malhata (Beit-Arieh and Freud 2015), ʿIra (Freud 1999: 194, figs.  6.83:5; 6.90:4, 16), Tell el-Khleifeh (Pratico 1993: 47, pl.  37-8-12), Kadesh Barnea (Bernick-Greenberg 2007a: 168–70, figs. 11.74–77), and Hazeva (Ben-Arieh 2011: 132-44, figs. 20–29). Neutron activation and petrographic analyses have established that the “Edomite” element in the Negev ceramic repertoire is mostly of local origin (Gunneweg et al. 1991: 249; Gunneweg and Mommsen 1995: 280–86; Cohen-Weinberger 2011: 188; Iserlis and Thareani 2011: 179– 87). Handmade and wheel-made Edomite cooking pots, some of which were imported from the Edomite plateau are also prominent (BernickGreenberg 2007b: 192–93; Iserlis and Thareani 2011: 184–85). Several pieces of evidence are important for our understanding of the nature of interactions with the Edomites: the unique Edomiteoriented assemblage from the cultic site of Ḥorvat Qitmit (Beck 1995: 186–87; 1996; Beit-Arieh 1995: 261–62, 264–67, 303–5); the use of the name qos—a major Edomite god—not only in the cultic context of this site (Beit-Arieh 1995: 267) but also in an administrative context at Arad (Aharoni 1981: 26); the qosa seal found at ʿAroer (Avigad and Sass 2011) and another seal from Hazeva bearing the Edomite inscription lmsqt ben wehazam (Naveh 2001: 197–98); the mention of the Edomite-oriented names ʿaznael and danael at Tel Malhata (Ziffer 2015); and the content of the Edomite ostracon from Ḥorvat ʿUza (Beit-Arieh 2007: 332–35; Naʾaman 2012: 214–16, 225), where an Edomite transport of grain to the fortress is documented. Arabian Repertoire.  Ostraca bearing Arabian signs have thus far been found at Beer-sheba (Singer-Avitz 1999: 50–52) and ʿAroer (Thareani 2011a: 228, pls. 4–5; 231.1; 207.1). A seal incorporating an Arabian name was unearthed at Hazeva (Naveh 2001: 197–98). Small limestone altars bearing organic residue have been recovered at ʿAroer (Thareani

214

Yifat Thareani

2011a: 206–7, figs. 3.96–97; pls. 49.8, 77.3), Beer-sheba (Singer-Avitz 1999: 41–44, figs. 11–12), and Qitmit (Beit-Arieh 1995: 275–76, fig. 6.5). This body of evidence suggests the presence of people of Arabian origin in the late Iron Age II Negev. Although the number of Arabian artifacts in the Negev Desert frontier is too meager to allow firm conclusions to be drawn, the few objects that have been discovered along the route leading from Arabia through the Negev Desert to the Coastal Plain imply that Arabian traders and experts settled in population centers in Judah and the neighboring regions and administered their business from there. Concurrently, the Judean Kingdom and its elites had a clear economic interest in the exotic commodities being brought from the south. Imported Objects from Assyria.  During the 8th–7th century b.c.e., the Negev also saw the appearance of Assyrian luxury items, trade-related objects, and other imported goods, such as a duck-shaped weight at ʿAroer (Thareani 2011a: 209, fig. 3.98, pl. 1.3), a crouching lion weight found in the Arad sanctuary (Herzog 2002: 80, fig. 35.3; Ornan 1997: 276–77), ceramic vessels imitating Assyrian metal bowls and bottles (Naʾaman and Thareani-Sussely 2006: with bibliographical references), a piece of an Assyrian glass cup (Barag 2011: 259–60, pls. 8, 48.2), and an Assyrian cylinder seal from Suhu (Rainey 1973: 61–70). In light of this multifaceted picture reflecting preservation of local traditions alongside booming trade relations and diverse material culture, we should ask: what can the archaeological and historical records tell us about the Assyrian control strategy? Assyrian Imperial Strategy for Controlling the Negev Desert Frontier While many scholars have emphasized the economic importance of the Arabian trade network to the Assyrian Empire (Tadmor 1966: 89–90; Otzen 1979: 255–56; Ephʿal 1982: 93–94; Edens and Bawden 1989; Naʾaman 1987: 8–11; Finkelstein 1995: 146–49; Parpola 2003: 103–4; Jasmin 2006), there has been only limited discussion of the control strategy that the Assyrians practiced in their southwestern frontier (the work of Naʾaman [1979] forms an exception). The paucity of scholarly attention to this question derives mainly from a near-complete absence of written sources. Nevertheless, the integration of various historical sources concerning the Assyrian, Roman–Byzantine, and Ottoman periods and of material culture from several Iron Age IIb–c Negev sites does shed new light on the nature of interaction between the Assyrian overlords and the local population in the desert frontier.

Socioeconomic Account of the Iron Age II Negev Desert

215

Unlike northern Palestine, the western Negev and the Coastal Plain where the Assyrians annexed the conquered territories and imposed Assyrian town planning, architecture, and other material culture manifestations (Ben-Shlomo and Van Beek 2014; Peersmann 2000; KoganZehavi 2006), the Negev—where architecture manifests a strong local tradition, and Assyrian artifacts are limited to a few imported items or local imitations—clearly constitutes a different case. But if the Assyrians did not enforce a permanent physical presence in the Negev, how then did they control this remote frontier zone and sustain their economic and political interests in the region? Ideologically, the Assyrians saw the frontier as a boundary between the civilized world and its chaotic periphery. This dichotomy marked the opposition between the known and the unknown, between the reassuring and the hostile (Liverani 1979: 306; Tadmor 1999: 60; Parker 2001: 265). This was especially true with regard to the desert, as demonstrated in Esarhaddon’s succession treaty (Parpola and Watanabe 1988: 45): May Sin, the brightness of heaven and earth, clothe you with leprosy and forbid your entering into the presence of the gods or the king. Roam the desert like the wild ass and the gazelle.

Contrary to Assyrian propaganda, however, the archaeological and historical evidence shows that Assyria maintained a flexible policy toward its imperial frontiers. The imperial strategy for controlling one such frontier, the Negev—which also had the important function of safeguarding the Egyptian border and the Arabian trade routes—is reflected in treaties with local Arab tribes. The royal annals state that Tiglath-pileser “appointed the [Idi]biʾilu as the Gatekeeper facing Egypt” (Tadmor 1994: 168–69, line 6). Tadmor and others have concluded from such historical evidence that the Assyrians established the Idibiʾlu as a proxy authority in the Besor region (Tadmor 1966: 89–90; Naʾaman 1979: 80–86; Ephʿal 1982: 93–94, 103–5). Another example of this strategy is provided by the campaign of Sargon II to the Brook of Egypt (the Besor region). The Assyrian king states that (Naʾaman 1979: 71): . . . (people) from the land [GN I deported and in] a land whose [place they have never known, in the city GN] situated on the border of the Brook of Eg[ypt, territory which is located on the shore of the] Western [sea] I set[tled them, in the hands of so and so], the sheikh of Laban, [I entrusted them].

This campaign included the settlement of exiles on the border with Egypt, giving that entire region to the Arab prince of the city of Laban (Naʾaman 1979: 71, 77). Sargon further boasted (Tadmor 1966: 92;

216

Yifat Thareani

Naʾaman 1979: 84): “I opened the sealed-off harbour [kāru] of Egypt, mingled Assyrians and Egyptians together and made them trade with each other.” These two quotations describe the political system that was created in the southwestern reaches of the Assyrian Empire in the second half of the 8th century b.c.e.—a system under which Arab tribal leaders settled in urban environments and served Assyrian imperial economic and political interests. The creation of a network of Assyrian fortresses in the Edomite plateau and especially the establishment of an Assyrian government center in Busayra (Bienkowski 2002: 57–110) support this. Against this background, it is reasonable to assume that the set of treaties that the Assyrians signed with the tribal leaders came at a certain price: something was given to the tribal elites in return for their loyalty. Assyrian administrative records describe precious items issued to delegations visiting the Assyrian court. Among references to emissaries from Ekron, Bit-Ammon, and the Phoenician coastal cities are mentions of gifts granted to Arab emissaries. The following lines constitute a representative example (Fales and Postgate 1992: 75–76):   [X] rings of gold, 1⁄3 sheqel for those who brought the tribute with him: total, Ilâ-nasaka, the sheikh of the Arabs.   [X] gold rings, small, Ter-ilaʾi, chief of the pack animals of the Arabs . . . a silver ring, 1⁄4 sheqel, for his Arab associates.

Foreign dignitaries visited the Assyrian capital frequently to pay tribute. These visitors were rewarded with rich garments, footwear, and feasting at the state’s expense (Postgate 1974: 113–14, 127). Historical documents and archaeological evidence from the Negev and other frontier zones in later periods illustrate similar modes of interaction between imperial powers and local populations. The Longue Durée of Imperial Control in the Negev The Roman–Byzantine Period.  Textual sources such as Procopius1 (6th century c.e.) indicate that the Byzantine imperial government in Constantinople negotiated with the chiefs of the Negev’s local tribal groups, the Saracens, to protect the frontier zone from the inroads of pastoralists and to secure the pilgrimage and trade routes—i.e., the central authority entered into a treaty with the tribal authority. Involvement with the imperial power intensified the social status of the tribal elites—an arrangement that guaranteed the central authority peace and political stability on the frontier, as well as the collaboration of the seminomads in protecting trade and pilgrim routes. In return, 1.  Procopius, The History of the Wars, Book 1, chapter 19.7–16, 181.

Socioeconomic Account of the Iron Age II Negev Desert

217

the tribal elites benefited from certain advantages (Mayerson 1989: 73; Shahîd 1995: 972–76; Isaac 1998: 417, 442–43). In several cases, the seminomadic pastoralists became an integral part of the regional sedentary system (Graf 1997: 366–69), whereas in others some of the local tribes settled in an urban environment (Shahîd 2002: 1–4). A crucial piece of evidence illustrating the important role that seminomads played in the Negev relates to a major figure in the history of Palestine during the Byzantine era: Abu-Karib, who served as a watchman over the Arabian frontier. Procopius’s “King of the Saracens,” AbuKarib was appointed by Justinian as Phylarch of Palestina Tertia, and at a certain point controlled Arab tribes in the Sinai Peninsula, the Negev, the southern part of Provincia Arabia, the region south of the Arnon, and part of the northern Hejaz (Shahîd 1989: 89–90; 2002: 28–33). In this context, a remark by the Sinai monk Ammonius (in Agnes 1912: 2) is worth quoting: “And after a few days suddenly many of the Saracens fell upon us; because at that time the king of the Saracens had died, he was the guardian of the desert.” Such accounts illustrate a reality in which the leaders of the Saracens enjoyed exaggerated material benefits, usually an allowance in money or in kind and in return provided transportation, delivery, road guidance, and animals. In most cases, peace and stability were maintained as long as treaties were respected by all sides. Nevertheless, the alliances made between townsmen, the central authority, and the Saracens were repeatedly broken. While some groups of a tribal origin based themselves in desert settlements, integrated with the townspeople, and protected caravans, groups of a similar origin raided the very same towns and markets, threatened their people, and robbed caravans in broad daylight (see, for example, how Theophanous describes the revenge of the Saracens against the central authority in reaction to the intermission of subsidies; Mayerson 1994: 55–56). However, although they were aware of this potential for nomadic hostility, the Romans could not completely prevent the penetration of tribes into settled lands, and the bulk of their efforts focused on protecting the settlements rather than the border itself (Mayerson 1986: 39–40; 1989: 73–75; Shahîd 1995: 972–74).2 2.  This imperial strategy is also evident from other frontier zones during the Roman–Byzantine period—the steppes of Tripolitania, for example. Textual sources and archaeological remains indicate that the Roman authorities encouraged the local tribal elites there, who resided on the frontier, to profit from association with the seat of imperial power. They offered them social advancement in exchange for loyalty, thereby incentivizing them to abandon nomadism in favor of a more sedentary way of life. In this way, tribal elites received lands from the Romans, creating a new class of seminomads turned landowners (Grahame 1998: 97–100).

218

Yifat Thareani

The Ottoman Period.  In the years following the establishment of Beer-sheba (1900), the Ottoman governor granted the local Bedouin tribes the right to elect the new town’s governor and his deputies, provided that the strongest candidate among the tribes was selected (GalPeer 1991: 33; Elkayam 1994: 116). The sheikhs of the large tribes profited from the high price that the Ottoman government paid for their lands. The council of Beer-sheba purchased lands from the Bedouin and incentivized them to settle in the town. In fact, the sheikhs were thus encouraged in order to facilitate control over them and indirectly over the tribes that they ruled. As a result of this, local tribal elites maintained homes among the inhabitants of Hebron and Gaza. Many Bedouin attended the Ottoman ceremonies, and their leaders were granted gifts and awards (Gal-Peer 1991: 33–34)3, while most of the local population continued practicing the traditional seminomadic mode of subsistence (Berman 1965: 316– 17; Gal-Peer 1991: 33). In addition, the Ottomans established a tribal school for the district’s Bedouin youth (1906). The privilege of studying at this school was reserved for the sheikhs’ and nobles’ sons who lived in Beer-sheba and its environs. Then, in 1913, in order to facilitate attendance by the youths who lived in more-remote areas, a boarding school was constructed adjacent to the original building. Behind its ostensible wish to bestow education on the offspring of the tribal elites, the Ottoman imperial authority was acting on its need to control the desert people, produce disciplined Ottoman civilians, and encourage a Bedouin transition to a sedentary way of life (Abu-Rabia 2001: 19–21). A longue durée historical view reveals close relations between imperial and local desert elites’ economic and political interests. It is clear that the Assyrians, Romans, Byzantines, and Ottomans treated tribal elites and their leaders as landowners and high-status nobles. By these means the empire created a new social class—a strategy that assured the support of the tribal leaders and enabled peace and internal stability in the empires’ remote frontiers. While tribal leaders and their nobles moved to permanent settlements where they closely protected their clans’ economic interests in the trade systems, large elements of the tribes continued in their traditional roles, which involved pastoralism, trade, and caravan raids. 3.  Other examples of this Ottoman policy are seen in the plains east of Aleppo, Syria, and in Sudan. There the Ottoman rulers encouraged leaders and nobles among the local Bedouin tribes to settle in lands defined as being under direct imperial proprietorship. While most of the tribe continued to raise livestock in the semiarid hinterland, the sheikhs were given a central role in the wool trade of Aleppo’s markets (Ahmed 1973: 83; Lewis 1987: 42–54; Masters 1999: 73–75).

Socioeconomic Account of the Iron Age II Negev Desert

219

Local Elites in the Service of the Imperial Power during the Late Iron Age Like their predecessors, the Assyrians had a clear interest in keeping the desert frontier as stable as possible. Peripheral states, including Judah, were subordinated rather than ruled directly. In this way, such kingdoms were made into buffer zones and economic intermediaries between Assyria, the Arabian tribes, and Egypt (Postgate 1992: 252–55; Parpola 2003: 103–4). Recognizing the delicate state of interaction between the settled population and nomads in the Negev, the Assyrians preferred to exploit a preexisting system of exchange without fundamentally altering it. But they also realized that the most efficient way to spread imperial influence into the recently conquered territories was to persuade their inhabitants to ally themselves with the Empire. Thus, local elites became the focus of Assyrian attention (Parpola 2003: 101–5). The Judean elite were one such group. Assyrian religious and ideological influence on the Judahite ruling class is demonstrated by locally manufactured seals and by art and local architecture (Cogan 1974: 91– 95; Stern 2001: 14–41; Ornan 2005: 168–82). Vassal rulers were explicitly told by the Assyrians to propagate Assyrian ideals among their people, and the foreign material culture forms attracted Judah’s wealthy class (Parpola 2003: 101 n. 5). The indirect control that the Assyrian Empire exerted in the desert frontier also left its mark on the elites of the Iron Age II Negev’s local seminomadic tribes (the “Edomites”). Assyrian cultural influence can be thus discerned in the archaeological record, mainly in the form of elite goods and styles, which included imported luxury items and their local imitations. Such Assyrian imports recovered at Negev sites were manifestations of status, derived from these objects’ scarcity and symbolic value as emblems of power and prestige. The ceramic repertoire and imported prestige objects also suggest assimilation. Assyrian material culture was not replicated but borrowed selectively, particularly in the case of objects that symbolized wealth and authority (Naʾaman and Thareani-Sussely 2006: 72–73). However, it is reasonable to assume that in an era of Assyrian globalization and an increasing tendency toward multiculturalism, anti-imperial voices were also heard in Judah (Machinist 1983). Under Assyrian patronage—which showed a vibrant interest in the Arabian trade—the Judean ruling class, local seminomadic groups, Arabs and others had no choice but to coexist. Their activities were controlled by the Assyrian imperial bureaucracy (even if the latter was in large part not physically present in the region). As a result, the Late Iron Age Negev—home to Judeans, local seminomadic tribes of an Edomite cultural orientation, as well as to Arabs and other groups—

220

Yifat Thareani

comprised a multicultural society in which peoples of diverse origins, traditions, and social strata coexisted. At the same time, the Assyrian imperial drive to establish settlements in the desert frontier zone exacerbated a local need to address the presence of foreigners, to cope with issues surrounding the limits of local authority and towns’ autonomy, and the influence and control of the dominating power. The Identity of Negev Dwellers under the Pax Assyriaca The manner in which the Edomite group coped with issues of social boundaries in this period of Assyrian expansion is reflected through the appearance of decorated Edomite pottery of local origin in Judeandominated ceramic assemblages from Negev sites. Ethnographic studies among African tribal societies have shown that material culture plays an important role in preserving the identity of social groups in frontier zones, particularly in times of heightened interaction with other social groups (Shack 1973: 251; Wiessner 1983: 271– 72; Emberling 1997: 309–10; Grahame 1998: 97–100). One of the most common means by which groups and individuals typically declare social identity is decoration. Tribal groups in frontier zones use decorated items as a social strategy to preserve group boundaries (Wiessner 1983: 270; Hodder 1985: 155–59). Studies of the material culture of prehistoric tribal societies have also illustrated how regionalism in material culture marks territory and social boundaries (Graves 1991; O’Shea and Milner 2002). In such cases, decoration constitutes a reaction to the presence of other groups and a means for preserving social identity in a dynamic multicultural reality. Edomite decorated ware is commonly found in direct or indirect association with hand- and wheel-made cooking pots imported from the Edomite plateau, as well as ostraca bearing the Edomite script and seals and inscriptions with Edomite names. Also appearing at this time are pig bones, in what are clearly Judean towns; they are therefore often associated with minorities of an “Edomite” cultural orientation. Quantitative analysis of the Edomite decorated vessels from Tel ʿAroer shows that the repertoire is limited to vessels for food preparation, consumption, and ritual activities (Thareani 2010: 51; 2011a: 156–57). The fact that these vessels were found in domestic and public contexts alike supports the view that such eating, drinking, and other ritual activities were at least in part communal. The majority of decorated Edomite vessels appear in the extramural areas of the town of ʿAroer, implying that the decorated vessel assemblage—which constitutes a small proportion of the site’s ceramics—reflects the existence of a population minority that performed its social activities on the margins of the town rather than at its center.

Socioeconomic Account of the Iron Age II Negev Desert

221

It seems that the decorated Edomite vessels preserved the group members’ sense of belonging and identity. Equally, the decorated vessels marked the limits of tribal territory and manifested social boundaries within the urban framework. As I have suggested elsewhere (Thareani 2010: 51–52), in the context of the desert frontier’s ever-growing multi­ culturalism and interaction with the outside world, the decoration and use of serving, eating, and ritual vessels of a distinctive style reflect a sharpening of tribal identity. The use of typical Edomite material culture forms was intended to preserve certain social behavior and to transmit a message about the ethnic identity of their users in a milieu in which subgroups from local tribes settled in an urban environment and interacted with a variety of ethnic groups within the framework of the Arabian trade system. The archaeological evidence left to us by desert frontier dwellers indicates expressions of identity that are both flexible and dynamic. Items of material culture from Negev sites reflect a combination of several social strategies present in the region. In times when local seminomadic groups chose to collaborate with external forces and adopt more sedentary ways of life, archaeology provides clues concerning the ethnic identity and origin of those settlers. During such periods, the imperial presence intensified, and the need to assert ethnic identity markers increased. Such activity was founded in cultural traditions that enabled the protection and highlighting of ethnic identity—under an imperial patronage that allowed such manifestations.

Forces of Decline Archaeological Evidence for the Final Destruction of the Iron Age II Negev After some 150 years of settlement growth and economic prosperity, the Iron Age II Negev settlement system came to a sudden end. Heavy destruction layers containing stone collapse, burned wooden beams, ash remains, and smashed complete vessels (all dated to the early 6th century b.c.e) have been found at most of the region’s central sites and fortresses: Tel ʿIra Stratum VI (Ayalon 1999: 45–49; Beit-Arieh 1999: 176–77; Biran 1999: 115); Tel ʿAroer Stratum IIb (Thareani 2011a: 25–33, 49–55, 107–8, 111–12); Tel Malhata Stratum III (Beit-Arieh 1998: 34–35); Tel Arad Strata VII–VI4 (Aharoni 1981: 150; Herzog 2002: 102); Ḥorvat 4.  Strata VII–VI at Arad are referred to here as a single stratigraphic horizon, in light of stratigraphic, ceramic, and paleographic critiques (Mazar and Netzer 1986: 90; Zimhoni 1997: 204–5; Stern 2001: 158–59) of the excavators’ interpretation (Aharoni 1981: 8, 149; Herzog 2002: 40–49).

222

Yifat Thareani

ʿUza Stratum III (Beit-Arieh 2007: 23–24, 53), and at the single-period site of Ḥorvat Qitmit (Beit-Arieh 1995: 12). In one case, Tel ʿAroer Stratum IIb, a pattern of abandonment was detected in the extramural road station of Area C, on the bank of Nahal ʿAroer (Thareani 2011a: 66–67). At Ḥorvat Radum, ʿUza’s neighbor to the south, only a few sherds and several Hebrew ostraca were detected in the gate and on the floors of inner rooms (Beit-Arieh 2007: 306–10, 314). The evidence from these two sites suggests convergences of population around central towns in times of warfare. Iron Age II Negev Destruction in Research History Archaeologists, historians, and biblical scholars have long debated the question of when the Negev Desert ceased to be part of the Kingdom of Judah. One school of thought dates the destruction of Negev towns to the Babylonian campaign of 598 b.c.e., arguing that the Negev was already cut off from Judah in 598/7 b.c.e. as part of a punitive campaign carried out by Nebuchadnezzar (Alt 1925: 108; Noth 1958: 283–84; Welten 1969: 166; Bartlett 1982: 23; 1989: 149–50). Another view puts the conquest and destruction of the Negev in the year 587 b.c.e., contemporary with the destruction of Jerusalem by Nebuchadnezzar (Kochavi 1970: 23; Biran and Cohen 1981: 272; Beit-Arieh 1985: 20, 25; 1986: 33, 35; 1987: 35; Naʾaman 1987: 15). According to a third viewpoint, the destruction of Negev towns was subsequent to both the Babylonian campaign and the destruction of Jerusalem (Lipschits 2005: 144–46, 181–82). The identity of the Negev towns’ destroyers is also a point of contention. Early historical research into this followed Jeremiah’s lamentation prophecy (Jer 13:18–19), which states that it was the Babylonian King Nebuchadnezzar who destroyed the Negev (Alt 1925: 108; Noth 1958: 283–84; Welten 1969: 166; Bartlett 1982: 23; 1989: 149–50). However, while some scholars have treated the biblical text as evidence for the complete or near-complete obliteration of early 6th-century b.c.e. Judean towns (Albright 1949: 142; Stern 2001: 304–31), others have suggested that this literary source is a postexilic creation that does not reflect the historical reality (Blenkinsopp 2002: 177–78).5 Alongside the Babylonian explanation, the antagonistic mention of Edom in the prophetic and other biblical books—Ezek 35:1–36, Psalm 137, and especially Obad 11–14, where Edom is singled out—contributed significantly to the more common view that relations between Edom and Judah were hostile and that it was in fact the Edomites who destroyed the Negev towns (Mazar 1963: 5–6; Myers 1971: 390–92; Lemaire 1977: 192–93; Aharoni 1982: 278–79; Kletter 1995: 24). This argument was further supported by the paleographic evidence of Ostracon 5.  And see Stern’s rejoinder to Blenkinsopp (Stern 2004).

Socioeconomic Account of the Iron Age II Negev Desert

223

24 from Arad, which contains an order to the commander of the fortress, Eliashib, to send reinforcements from Arad and Qinah to the commander of the garrison at Ramat-Negeb, in advance of an attack expected there (Aharoni 1981: 46–49): From Arad 50 and from Kin[ah] . . . and you shall send them to RamatNegeb by the hand of Malkiyahu the son of Qerabʾur and he shall hand them over to Elishaʾ the son of Yirmiyahu in Ramat-Negeb, lest anything should happen to the city. And the word of the king is incumbent upon you for your very life! Behold, I have sent to warn you today: [Get] the men to Elishaʾ: lest Edom should come there.

Looking at this evidence together with the Edomite ostracon from Ḥorvat ʿUza, some scholars have suggested that there was a temporary Edomite conquest of the Judean fortresses prior to their final demise (Aharoni 1981: 150; Blenkinsopp 2002: 186; Beit-Arieh 2007: 332–35). However, this argument is not supported by the evidence from Negev sites, where only one major destruction horizon is detected. Moreover, the presence of a military ostracon written in a foreign language should not necessarily be taken as evidence for military conquest. Rather, this could reflect the hiring of mercenaries of various origins by Judah to operate its forts—a common practice during Iron Age II (Faust 1995: 85– 87; Thareani-Sussely 2008: 201, 209; Naʾaman 2012: 225). Furthermore, the ostraca represent Judean-dominated fortresses hosting groups of diverse ethnic origins, such as the Kittim (ktym), most likely of Aegean origin (Aharoni 1981: 12–13; Herzog et al. 1984: 31; Fantalkin 2001: 140– 42). In my opinion, the paleographic evidence from Negev fortresses implies that local seminomadic desert dwellers were integrated with the administration and maintenance of the forts. It is worth noting here an argument put forward by Ben Zvi, whereby the exceptionally hostile description of Edom in the book of Obadiah is derived from its traditional “brotherhood” with Israel rather than a deep-rooted historical hatred of Edom in postmonarchic Judah (Ben Zvi 1996: 230–46). Against this background, it seems that the traditional scholarly view of the Babylonian Empire and the Edomite kingdom as demolishers of Judah’s southern frontier relies on literary evidence, the meaning and date of which are speculative at most. I therefore believe that archaeology can shed new light on the old question of the identity of the Iron Age II Negev system destroyers. Archaeological evidence from Iron Age IIb–c Negev sites, including the so-called Edomite vessels, locally made and imported cooking pots, ostraca, and seals bearing Edomite script and names, as well as pig bones bearing cutting marks illustrate the crucial role that the “Edomite” ethnic group played in the sociopolitical history of the region.

224

Yifat Thareani

Being a minority that settled in the desert frontier of the Judean Kingdom at a time of Assyrian hegemony and interacted with a variety of ethnic groups within the framework of the Arabian trade network, these Edomite-oriented seminomadic tribes experienced the need to sharpen their tribal identity and delineate their social and cultural boundaries (Thareani-Sussely 2008; 2010). But in spite of the multicultural experience that brought economic prosperity and political stability to this arid region, the material culture from Negev sites suggests that relationships between local seminomadic and sedentary populations were not egalitarian. In such a social environment, ethnic tension was inevitable. The recurring theme of fractious relations between seminomadic tribes and the territorial states in which they find themselves has been described by Rowton (1976: 240), who noted that western Asian seminomadic tribes have tended to strengthen their autonomy whenever the power of the central authority is weakened. The third quarter of the 7th century b.c.e. witnessed the withdrawal of the Assyrian Empire from the southern Levant. Judean society was then divided between pro-Egyptian and pro-Babylonian parties. This weakening of central authority led to increasing tensions between the belabored Judean state and the local seminomads of the Negev. In the absence of any imperial protection, the Judean capacity to guard and patrol the desert frontier was lost. It may be that the increasing ethnic tensions and political uncertainties opened the way for the local Edomite groups of the Iron Age II Negev to challenge the ruling Judeans. The Babylonian campaign of 587 b.c.e. could have put such pressure on the authorities in Jerusalem that they could no longer control and look after the state’s territories. The first to suffer were the frontier areas. Local seminomads took advantage of Judah’s political and administrative weakness, collaborated with the Babylonian armies, conquering the region and destroying the seats of Judean power. Whether the local Edomites collaborated with Babylonian troops or with Judah’s neighbors (who had waited long for such an opportunity) is not clear. It is also difficult to ascertain precisely when the Negev settlement system was destroyed, although it seems to have ceased functioning as part of the Judean Kingdom around the same time as the destruction of Jerusalem. The conflict between the Judean rulers and local seminomads was motivated by an increasingly divisive mentality on the part of the latter in terms of asserting their own identities, as well as by their desire to enlarge their profits from trade and other activities previously controlled by Judah. According to this reconstruction, the Iron Age II Negev towns were destroyed as a result of an internal socioethnic conflict between local

Socioeconomic Account of the Iron Age II Negev Desert

225

seminomadic and sedentary groups, which produced a relatively rapid, short, unorganized process of destruction. In this view, far-reaching political changes that took place in Judah toward the end of the 7th and the early 6th centuries b.c.e. provided the opportunity for local seminomads of Edomite orientation to take revenge on their Judean neighbors.

Bibliography Abu-Rabia, A. 2001 A Bedouin Century. New York: Berghahn. Agnes, L. W. 1912 The Forty Martyrs of the Sinai Desert. Cambridge: Cambridge University Press. Aharoni, Y. 1981 Arad Inscriptions. Jerusalem: Israel Exploration Society. 1982 The Archaeology of the Land of Israel: From the Prehistoric Beginnings to the End of the First Temple Period. Philadelphia: Westminster. Ahmed, A. M. 1973 Tribal and Sedentary Elites: A Bridge between Two Communities. Pp. 75–96 in The Desert and the Sown: Nomads in the Wider Society, ed. C. Nelson. Berkeley: University of California Press. Albright, W. F. 1949 The Archaeology of Palestine. Baltimore: Penguin. Alt, A. 1925 Judas Gaue unter Josia. PJ 21: 100–116. Avigad, N., and Sass, B. 2011 An Edomite Seal. P. 227 in Tel ʿAroer: An Iron Age II Caravan Town and a Hellenistic and Early Roman Settlement in the Negev. Avraham Biran (1975–1982) and Rudolph Cohen (1975–1976) Excavations, ed. Y.  Tha­ reani. Annual of the Nelson Glueck School of Biblical Archaeology 8. Jerusalem: The Nelson Glueck School of Biblical Archaeology. Ayalon, E. 1999 Area C. Pp.  40–59 in Tel ʿIra: A Stronghold in the Biblical Negev, ed. I.  Beit-Arieh. Monograph Series of the Institute of Archaeology of Tel Aviv University 15. Tel Aviv: Institute of Archaeology, Tel Aviv University. Barag, D. 2011 A Neo-Assyrian Glass Bowl Fragment. Pp. 259–60 in Tel ʿAroer: An Iron Age II Caravan Town and a Hellenistic and Early Roman Settlement in the Negev. Avraham Biran (1975–1982) and Rudolph Cohen (1975–1976) Excavations, ed. Y. Thareani. Annual of the Nelson Glueck School of Biblical Archaeology 8. Jerusalem: Hebrew Union College. Bartlett, J. R. 1982 Edom and the Fall of Jerusalem, 587 b.c. PEQ 114: 2–24. 1989 Edom and the Edomites. JSOTSup 77. Sheffield: JSOT Press.

226 Beck, P. 1995

Yifat Thareani

Catalogue of Cult Objects and Study of the Iconography. Pp. 27–197 in Ḥorvat Qitmit: An Edomite Shrine in the Biblical Negev, ed. I. BeitArieh. Monograph Series of the Institute of Archaeology 11. Tel Aviv: Institute of Archaeology, Tel Aviv University. 1996 Ḥorvat Qitmit Revisited via ʿEn Hazeva. TA 23: 102–14. Beit-Arieh, I. 1985 An Edomite Ostracon from Ḥorvat ʿUza. TA 12: 96–101. 1986 The Ostracon of Ahiqam from Ḥorvat ʿUza. TA 13: 32–38. 1987 Tel ʿIra and Ḥorvat ʿUza Negev Sites in the Late Israelite Period. Cathedra 42: 34–38. [Hebrew] Beit-Arieh, I., ed. 1995 Ḥorvat Qitmit: An Edomite Shrine in the Biblical Negev. Monograph Series of the Institute of Archaeology 11. Tel Aviv: Institute of Archaeology, Tel Aviv University. 1999 Tel ʿIra: A Stronghold in the Biblical Negev. Monograph Series of the Institute of Archaeology 15. Tel Aviv: Institute of Archaeology, Tel Aviv University. 2007 Ḥorvat ʿUza and Ḥorvat Radum: Two Fortresses in the Biblical Negev. Monograph Series of the Institute of Archaeology 25. Tel Aviv: Institute of Archaeology, Tel Aviv University. Beit-Arieh, I., and Freud, L., eds. 2015 Tel Melḥata: A Central City in the Biblical Negev. Monograph Series of the Institute of Archaeology 32. Tel Aviv: Institute of Archaeology, Tel Aviv University. Ben-Arieh, S. 2011 A Temple Furniture from a Favissa at ʿEn Hazeva. ʿAriqot 68: 107–75. Ben-David, J. 1990 The Negev Bedouin: From Nomadism to Agriculture. Pp.  181–95 in The Land That Became Israel: Studies in Historical Geography, ed. R. Kark. Jerusalem: Magnes. Ben-Shlomo, D., and Van Beek, G. W., eds. 2014 The Smithsonian Institution Excavation at Tell Jemmeh, Israel, 1970–1990. Washington, DC: Smithsonian Institution Scholarly Press. Ben Zvi, E. 1996 A Historical-Critical Study of the Book of Obadiah. Berlin: de Gruyter. Berman, M. 1965 The Evolution of Beersheba as an Urban Center. Annals of the Association of American Geographers 55: 308–26. Bernick-Greenberg, H. 2007a The Ceramic Assemblages and the Wheel-Made Pottery Typology. Pp. 131–86 in Excavations at Kadesh Barnea (Tell el-Qudeirat) 1976–1982, ed. R. Cohen and H. Bernick-Greenberg. IAA Reports 34. Jerusalem: Israel Antiquities Authority.

Socioeconomic Account of the Iron Age II Negev Desert

227

2007b The Negebite Ware Typology. Pp.  187–210 in Excavations at Kadesh Barnea (Tell el-Qudeirat) 1976–1982, ed. R. Cohen and H. BernickGreenberg. IAA Reports 34. Jerusalem: Israel Antiquities Authority. Bienkowski, P., ed. 2002 Busayra Excavations by Crystal M. Bennett, 1971–1980. British Academy Monographs in Archaeology 13. Oxford: Oxford University Press. Bienkowski, P., and Steen, E. van der 2001 Tribes, Trade and Towns: A New Framework for the Late Iron Age in Southern Jordan and the Negev. BASOR 323: 21–47. Biran, A. 1999 Area L–Area M. Pp. 115–28 in Tel ʿIra: A Stronghold in the Biblical Negev, ed. I. Beit-Arieh. Monograph Series of the Institute of Archaeology 15. Tel Aviv: Institute of Archaeology, Tel Aviv University. Biran, A., and Cohen, R. 1981 ʿAroer in the Negev. ErIsr 15: 250–73. [Hebrew, English Summary] Blenkinsopp, J. 2002 The Bible, Archaeology and Politics; or The Empty Land Revisited. JSOT 27: 169–87. Burckhardt, J. L. 1822 Travels in Syria and the Holy Land, Comprehending an Account of Those Territories in the Hedjaz Which the Mohammedans Regard as Sacred. London: John Murray. Cogan, M. 1974 Imperialism and Religion: Assyria, Judah and Israel in the Eighth and Seventh Centuries b.c.e. SBLMS 19. Missoula, MT: Scholars Press. Cohen-Weinberger, A. 2011 Provenance of the Clay Artifacts from the Favissa at ʿEn Hazeva. ʿAriqot 68: 185–89. Dayan, T. 1999 Faunal Remains: Areas A–G. Pp.  480–87 in Tel ʿIra: A Stronghold in the Biblical Negev, ed. I. Beit-Arieh. Monograph Series of the Institute of Archaeology 5. Tel Aviv: Institute of Archaeology, Tel Aviv University. Edens, C., and Bawden, G. 1989 History of Tayma and Hejazi Trade during the First Millennium bc. JESHO 32: 48–103. Elkayam, M. 1994 Forty Years of Jewish Settlement in Gaza. Netzarim: Center for the Commemoration of the Jews of Gaza. [Hebrew] Emberling, G. 1997 Ethnicity in Complex Societies: Archaeological Perspectives. JAR 5: 295–344. Ephʿal, I. 1982 The Ancient Arabs: Nomads on the Border of the Fertile Crescent, 9th–5th Centuries b.c. Jerusalem: Magnes.

228

Yifat Thareani

Fales, F. M., and Postgate, J. N. 1992 Imperial Administrative Records, part 1: Palace and Temple Administration. Helsinki: Helsinki University Press. Fantalkin, A. 2001 Mezad Hashavyahu: Its Material Culture and Historical Background. TA 28: 3–165. Faust, A. 1995 The Rural Settlement in the Land of Israel during the Period of the Monarchy. M.A. Thesis. Bar-Ilan University. Ramat-Gan. [Hebrew, English Summary] 2008 Settlement and Demography in Seventh-Century Judah and the Extent and Intensity of Sennacherib’s Campaign. PEQ 140: 168–94. Faust, A., and Weiss, E. 2005 Judah, Philistia, and the Mediterranean World: Reconstructing the Economic System of the Seventh Century b.c.e. BASOR 338: 71–92. Finkelstein, I. 1992 Ḥorvat Qitmit and the Southern Trade in the Late Iron Age II. ZDPV 108: 156–70. 1994 The Archaeology of the Days of Manasseh. Pp.  169–87 in Scripture and Other Artifacts: Essays on the Bible and Archaeology in Honor of Philip J. King, ed. M. Coogan, J. C. Exum, and L. E. Stager. Louisville: Westminster. 1995 Living on the Fringe: The Archaeology and History of the Negev, Sinai and Neighbouring Regions in the Bronze and Iron Ages. Monographs in Mediterranean Archaeology 6. Sheffield: Sheffield Academic Press. Finkelstein, I., and Naʾaman, N. 2004 The Judahite Shephelah in the Late 8th and Early 7th Centuries bce. TA 31: 60–79. Freud, L. 1999 The Iron Age Pottery. Pp. 189–289 in Tel ʿIra: A Stronghold in the Biblical Negev, ed. I. Beit-Arieh. Monograph Series of the Institute of Archaeology 15. Tel Aviv: Institute of Archaeology, Tel Aviv University. Freud, L., and Beit-Arieh, I. 1995 Pottery. Pp. 209–57 in Ḥorvat Qitmit: An Edomite Shrine in the Biblical Negev, ed. I. Beit-Arieh. Monograph Series of the Institute of Archaeology 11. Tel Aviv: Institute of Archaeology, Tel Aviv University. Gal-Peer, I. 1991 The Establishment of Beer Sheva in the Ottoman Period. Pp. 31–37 in Beer Sheva and Its Sites, ed. G. Biger and E. Schiller. Jerusalem: Ariel. [Hebrew] Graf, D. F. 1997 Rome and the Arabian Frontier: From the Nabataeans to the Saracens. Ashgate: Variorum. Grahame, M. 1998 Rome without Romanization: Cultural Change in the Pre-desert of Tripolitania (First–Third Centuries ad). Oxford Journal of Archaeology 17: 93–111.

Socioeconomic Account of the Iron Age II Negev Desert

229

Graves, M. W. 1991 Pottery Production and Distribution among the Kalinga: A Study of Household and Regional Organization and Differentiation. Pp. 112– 43 in Ceramic Ethnoarchaeology, ed. W. A. Longacre. Tucson: University of Arizona Press. Gunneweg, J., et al. 1991 “Edomite,” “Negevite” and “Midianite” Pottery from the Negev Desert and Jordan: Instrumental Neutron Activation Analysis Results. Archaeometry 33: 239–53. Gunneweg, J., and Mommsen, H. 1995 Instrumental Neutron Activation Analysis of Vessels and Cult Objects. Pp.  280–86 in Ḥorvat Qitmit: An Edomite Shrine in the Biblical Negev, ed. I. Beit-Arieh. Monograph Series of the Institute of Archaeology 11. Tel Aviv: Institute of Archaeology, Tel Aviv University. Herzog, Z. 1994 The Beer-Sheba Valley: From Nomadism to Monarchy. Pp. 122–49 in From Nomadism to Monarchy: Archaeological and Historical Aspects of Early Israel, ed. I. Finkelstein and N. Naʾaman. Jerusalem: Yad BenZvi / Israel Exploration Society. 2002 The Fortress Mound at Tel Arad: An Interim Report. TA 29: 3–109. Herzog, Z., et al. 1984 The Israelite Fortress at Arad. BASOR 254: 1–33. Hesse, B. 1990 Pig Lovers and Pig Haters: Patterns of Palestinian Pork Production. Journal of Ethnobiology 10: 195–225. Hodder, I. 1985 Boundaries as Strategies: An Ethnoarchaeological Study. Pp. 141–59 in The Archaeology of Frontiers and Boundaries, ed. S. W. Green and S. M. Perlman. Orlando, FL: Academic Press. Isaac, B. 1998 The Near East under Roman Rule. Leiden: Brill. Iserlis, M., and Thareani, Y. 2011 Petrographic Analysis. Pp. 179–87 in Tel ʿAroer: An Iron Age II Caravan Town and a Hellenistic and Early Roman Settlement in the Negev. Avraham Biran (1975–1982) and Rudolph Cohen (1975–1976) Excavations, ed.  Y.  Thareani. Annual of the Nelson Glueck School of Biblical Archaeology 8. Jerusalem: The Nelson Glueck School of Biblical Archaeology. Jasmin, M. 2006 The Emergence and First Development of the Arabian Trade across the Wadi Arabah. Pp.  143–50 in Crossing the Rift: Resources, Routes, Settlement Patterns and Interaction in the Wadi Arabah, ed. P. Bienkowski and K. Galor. Levant Supplementary Series 3. Oxford: Oxbow. Katz, H. 2008 A Land of Grain and Wine . . . a Land of Olive Oil and Honey: The Economy of the Kingdom of Judah. Jerusalem: Yad Ben Zvi. [Hebrew]

230

Yifat Thareani

Kletter, R. 1995 Selected Material Remains of Judah at the End of the Iron Age in Relation to Its Political Borders. Ph.D. dissertation. Tel Aviv University. [Hebrew, English Summary] Kochavi, M. 1970 The First Season of Excavations at Tell Malhata. Qadmoniot 9: 22–24. [Hebrew] Kogan-Zehavi, E. 2006 Tel Ashdod. Hadashot Arkeologiyot 118. http://www.hadashot-esi.org .il/report_detail_eng.asp?id=340&mag_id=111. Kolska Horwits, L. 1999 Faunal Remains: Areas L and M. Pp. 488–94 in Tel ʿIra: A Stronghold in the Biblical Negev, ed. I. Beit-Arieh. Monograph Series of the Institute of Archaeology 15. Tel Aviv: Institute of Archaeology, Tel Aviv University. Lemaire, A. 1977 Inscriptions Hebraiques, vol. 1: Les Ostraca. Paris: Cerf. Lewis , N. N. 1987 Nomads and Settlers in Syria and Jordan, 1800–1980. Cambridge: Cambridge University Press. Liphschitz, N., and Waisel, Y. 1973a Analysis of the Botanical Material of the 1969–1970 Seasons and the Climatic History of the Beer-Sheba Region. Pp. 97–105 in BeerSheba I: Excavations at Tel Beer-Sheba, 1969–1971 Seasons, ed. Y. Aharoni. Monograph Series of the Institute of Archaeology 2. Tel Aviv: Institute of Archaeology, Tel Aviv University. 1973b Dendroarchaeological Investigations in Israel. IEJ 23: 30–36. Lipschits, O. 2005 The Fall and Rise of Jerusalem: Judah under Babylonian Rule. Winona Lake, IN: Eisenbrauns. Liverani, M. 1979 The Ideology of the Assyrian Empire. Pp. 297–317 in Power and Propaganda: A Symposium on Ancient Empires, ed. M. T. Larsen. Copenhagen: Akademisk Forlag. Machinist, P. 1983 Assyria and Its Image in the First Isaiah. JAOS 103: 719–37. Marx, E. 1967 Bedouin of the Negev. Manchester: Manchester University Press. Masters, B. 1999 Aleppo: The Ottoman Empire’s Caravan City. Pp. 17–78 in The Ottoman City between East and West: Aleppo, Izmir, and Istanbul, ed. E. Eldem, D. Goffman, and B. Masters. Cambridge: Cambridge University Press. Mayerson, P. 1986 The Saracens and the Limes. BASOR 262: 35–47.

Socioeconomic Account of the Iron Age II Negev Desert 1989

231

Saracens and Romans: Micro-Macro Relationships. BASOR 274: 71–79. 1994 Monks, Martyrs, Soldiers and Saracens: Papers on the Near East in Late Antiquity (1962–1993). Jerusalem: Israel Exploration Society in Association with Yale University. Mazar, A., and Netzer, E. 1986 On the Israelite Fortress at Arad. BASOR 263: 87–90. Mazar, B. 1963 ʿEin Gedi Archaeological Excavations 1961–1962. Yediot 28: 1–15. [Hebrew] Motro, H. 2011 Archaeozoological Analysis of the Faunal Remains. Pp. 265–97 in Tel ʿAroer: An Iron Age II Caravan Town and a Hellenistic and Early Roman Settlement in the Negev. Avraham Biran (1975–1982) and Rudolph Cohen (1975–1976) Excavations, ed. Y. Thareani. Annual of the Nelson Glueck School of Biblical Archaeology 8. Jerusalem: The Nelson Glueck School of Biblical Archaeology. Myers, J. M. 1971 Edom and Judah in the Sixth–Fifth Centuries b.c. Pp. 377–92 in Near Eastern Studies in Honor of William Foxwell Albright, ed. H. Goedicke. Baltimore: Johns Hopkins University Press. Naʾaman, N. 1979 The Brook of Egypt and Assyrian Policy on the Border of Egypt. TA 6: 68–90. 1987 The Negev in the Last Century of the Kingdom of Judah. Cathedra 42: 4–15. [Hebrew] 2005 Ancient Israel and Its Neighbors: Interaction and Counteraction. Collected Essays, vol. 1. Winona Lake, IN: Eisenbrauns. 2012 A New Look at the Epigraphic Finds from Ḥorvat ʿUza. TA 39: 212–29. Naʾaman, N., and Thareani-Sussely, Y. 2006 Dating the Appearance of Imitations of Assyrian Ware in Southern Palestine. TA 33: 61–82. Naveh, J. 2001 A Sixth Century bce Edomite Seal from ʿEn Hazeva. ʿAtiqot 42: 197–98. Negev, A. 1988 The Architecture of Mampsis: Final Report, vol. 1: The Middle and Late Nabatean Periods. Qedem 26. Jerusalem: Hebrew University of Jerusalem. Noth, M. 1958 The History of Israel. London: Black. O’Shea, J. M., and Milner, C. M. 2002 Material Indicators of Territory, Identity and Interaction in a Prehistoric Tribal System. Pp. 200–226 in The Archaeology of Tribal Societies, ed. W. A. Parkinson. Archaeological Series 15. Ann Arbor: International Monographs in Prehistory.

232

Yifat Thareani

Ornan, T. 1997 Mesopotamian Influence on the Glyptic of Israel and Jordan in the First Millennium b.c. Ph.D. dissertation. Tel Aviv University. [Hebrew, English Summary] 2005 The Triumph of the Symbol: Pictorial Representation of Deities in Mesopotamia and the Biblical Image Ban. Fribourg: Academic Press. Otzen, B. 1979 Israel under the Assyrians. Pp.  251–61 in Power and Propaganda: A Symposium on Ancient Empires, ed. M. T. Larsen. Mesopotamia 7. Copenhagen: Akademisk Forlag. Parker, B. J. 2001 The Mechanics of Empire: The Northern Frontier of Assyria as a Case Study in Imperial Dynamics. The Neo-Assyrian Text Corpus Project. Helsinki: Helsinki University Press. Parpola, S. 2003 Assyria’s Expansion in the 8th and 7th Centuries and Its Long-Term Repercussions in the West. Pp.  99–111 in Symbiosis, Symbolism and the Power of the Past, ed. W. G. Dever and S. Gitin. Winona Lake, IN: Eisenbrauns. Parpola, S., and Watanabe, K., eds. 1988 Neo-Assyrian Treaties and Loyalty Oaths. Helsinki: Helsinki University Press. Peersmann, J. 2000 Assyrian Magiddu: The Town Planning of Stratum III. Pp. 524–34 in Megiddo  III: The 1992–1996 Seasons, ed. I. Finkelstein, D. Ussishkin, and B. Halepern. Monograph Series of the Institute of Archaeology 18. Tel Aviv: Institute of Archaeology, Tel Aviv University. Postgate, J. N. 1974 Taxation and Conscription in the Assyrian Empire. Rome: Pontifical Biblical Institute. 1992 The Land of Assur and the Yoke of Assur. World Archaeology 23: 247–63. Pratico, G. D. 1993 Nelson Glueck’s 1938–1940 Excavations at Tell el-Kheleifehh: A Reappraisal. ASOR Archaeological Report 3. Atlanta: Scholars Press. Procopius 1914–40  History of the Wars: Books I and II, trans. H. B. Dewing. Cambridge: Harvard University Press. Rainey, A. 1973 The Cuneiform Inscription on a Votive Cylinder from Beer-Sheba. Pp. 61–70 in Beer-Sheba I: Excavations at Tel Beer-Sheba; 1969–1971 Seasons, ed. Y. Aharoni. Monograph Series of the Institute of Archaeology 2. Tel Aviv: Institute of Archaeology, Tel Aviv University. Rowton, M. 1976 Dimorphic Structure and the Tribal Elite. Pp. 219–57 in El-Bahit: Festschrift Joseph Henninger. Studia Insttuti Anthropos 28. St. Augustin bei Bonn: Verlag des Anthropos-Instituts.

Socioeconomic Account of the Iron Age II Negev Desert

233

Rubin, R. 1989 The Debate over Climatic Changes in the Negev, Fourth–Seventh Centuries c.e. PEQ 121: 71–78. 1990 The Negev as a Settled Land: Urbanization and Settlement in the Desert in the Byzantine Period. Jerusalem: Yad Ben-Zvi / Israel Exploration Society. 1996 Urbanization, Settlement and Agriculture in the Negev Desert: The Impact of the Roman–Byzantine Empire on the Frontier. ZDPV 112: 49–60. Sapir-Hen, L., et al. 2013 Pig Husbandry in Iron Age Israel and Judah. ZDPV 129: 1–20. Sasson, A. 2004 Animal Husbandry (Caprine and Cattle) in Light of Zooarchaeological Research in Stratum II (8th Century bce) at Tel Beer-Sheba. Ph.D. dissertation. Tel Aviv University. [Hebrew, English Summary] Shack, W. A. 1973 Urban Ethnicity and the Cultural Process of Urbanization in Ethiopia. Pp. 251–85 in Urban Anthropology: Cross-Cultural Studies of Urbanization, ed. A. Southall. New York: Oxford University Press. Shahîd, I. 1989 Byzantium and the Arabs in the Fifth Century. Washington: Dumbarton Oaks Research Library and Collection. 1995 Byzantium and the Arabs in the Sixth Century, vol. 1. Washington, DC: Dumbarton Oaks Research Library and Collection. 2002 Byzantium and the Arabs in the Sixth Century, vol. 2/1: Toponymy, Monuments, Historical Geography, and Frontier Studies. Washington, DC: Dumbarton Oaks Research Library and Collection. Shereshevski, J. 1991 Byzantine Urban Settlements in the Negev Desert. Beer-Sheva 5. BeerSheva: Beer-Sheva University Press. Singer Avitz, L. 1999 Beersheba: A Gateway Community in Southern Arabian LongDistance Trade in the Eighth Century b.c.e. TA 26: 3–74. 2004 “Busayra Painted Ware” at Tel Beersheba. TA 31: 80–89. Stern, E. 2001 Archaeology of the Land of the Bible, vol. 2: The Assyrian, Babylonian, and Persian Periods, 732–332 bce. New York: Doubleday. 2004 The Babylonian Gap: The Archaeological Reality. JSOT 28: 273–77. Tadmor, H. 1966 Philistia under Assyrian Rule. BA 29: 86–102. 1975 Assyria and the West: The Ninth Century and Its Aftermath. Pp. 36– 43 in Unity and Diversity: Essays in the History, Literature, and Religion of the Ancient Near East, ed. H. Goedicke and J. J. M. Roberts. Baltimore: Johns Hopkins University Press. 1994 The Inscriptions of Tiglath Pileser III King of Assyria. Jerusalem: The Israel Academy of Sciences and Humanities.

234 1999

Yifat Thareani

World Dominion: The Expanding Horizon of the Assyrian Empire. Pp. 55– 62 in Landscapes: Territories, Frontiers and Horizons in the Ancient Near East. Papers Presented to the XLIV Rencontre Assryiologique Internationale, Venezia, 7–11 July 1997, part 1, ed. L. Milano et al. Padua: Sargon. Tcherov, E., and Drori, I. 1983 Economic Patterns and Environmental Conditions at Hirbet el-Msas during the Early Iron Age. Pp. 213–22 in Ergebnisse der Ausgrabungen auf der Hirbet el-Msas (Tel Masos) 1972–1975, ed. V. Fritz and A. Kempinski. Wiesbaden: Harrassowitz. Thareani, Y. 2010 The Spirit of Clay: “Edomite Pottery” and Social Awareness in the Late Iron Age. BASOR 359: 35–55. 2011a Tel ʿAroer: An Iron Age II Caravan Town and a Hellenistic and Early Roman Settlement in the Negev. Avraham Biran (1975–1982) and Rudolph Cohen (1975–1976) Excavations. Annual of the Nelson Glueck School of Biblical Archaeology 8. Jerusalem: The Nelson Glueck School of Biblical Archaeology. 2011b Towns in the Desert: Geographical, Economic and Sociopolitical Perspectives. Ph.D. dissertation. Tel Aviv University. [Hebrew, English Summary] Thareani-Sussely, Y. 2002 Core and Periphery—A Case Study: The Arad-Beersheba Valley at the End of the Iron Age. M.A. Thesis. Tel Aviv University. [Hebrew, English Summary] 2007a Ancient Caravanserai: An Archaeological View from ʿAroer. Levant 39: 123–41. 2007b The “Archaeology of the Days of Manasseh” Reconsidered in Light of the Evidence from the Beersheba Valley. PEQ 139: 69–77. 2008 Desert Outsiders: Extra-Mural Neighborhoods in the Iron Age Negev. Pp. 197–212 in Bene-Israel: Studies in the Archaeology of Israel and the Levant during the Bronze and Iron Ages in Honour of Israel Finkelstein, ed. A. Fantalkin and A. Yasur Landau. Leiden: Brill. Velde, C. W. M. van de 1854 Reis door Syrie en Palestina in 1851 en 1852. Utrecht: Kemink. Wapnish, P. 1981 Camel Caravans and Camel Pastoralists at Tell Jemmeh. JANES 13: 101–21. Welten, P. 1969 Die Königs-Stempl. Wiesbaden: Harrassowitz. Wiessner, P. 1983 Style and Social Information in Kalahari San Projectile Points. AA 48: 253–76. Yair, A. 2005 Hillslope Hydrology Water Harvesting and Areal Distribution of Some Ancient Agricultural Systems in the Northern Negev Desert. Beer-Sheva 17: 91–111. [Hebrew]

Socioeconomic Account of the Iron Age II Negev Desert Ziffer, I. 2015

235

Iron Age Stamp Seals, a Cylinder Seal and Impression. Pp. 514–23 in Tel Melḥata: A Central City in the Biblical Negev, ed. I. Beit-Arieh and L. Freud. Monograph Series of the Institute of Archaeology 32. Tel Aviv: Institute of Archaeology, Tel Aviv University. Zimhoni, O. 1997 Studies in the Iron Age Pottery of Israel. Journal of the Institute of Archaeology of Tel Aviv University: Occasional Publications 2. Tel Aviv: Tel Aviv University Press.

The Rural Economy of Judah during the Persian Period and the Settlement History of the District System Oded Lipschits Tel Aviv University

Introduction The most conspicuous phenomenon that took place in Judah following the destruction of Jerusalem was the sharp deterioration of urban life as contrasted with the stability of the rural settlements north of the capital, in the region of Benjamin, as well as in the area south of the city between Bethlehem and Beth-zur.1 The settlement center of gravity moved from Jerusalem to its close periphery, and a new pattern of settlement was created; the old core became depleted, and the nearby periphery continued its existence almost unchanged, even if in much lower numbers. The more remote peripheral regions of the kingdom in the southern Shephelah to the southwest, in the Negev to the south, as well as in the Judean Desert, the Jordan Valley, and on the western shore of the Dead Sea to the east collapsed—probably in a longer, more complex process, with ruinous consequences. The Negev and the southern Shephelah became cut off from the Persian period province of Judah, and in a slow progression the new province of Idumea emerged (Lipschits 2005: 149–84, 206–57; 2011a: 66–72; 2011b: 188–94, with additional bibliographical sources; cf. Kloner and I. Stern 2007).2 Throughout the Persian period, the rural settlement in the close periphery around Jerusalem continued to characterize the province of Yehud. Despite the rebuilding of Jerusalem and the restoration of its status as the cultic center of the province, at least until the 2nd century b.c.e., the settlement in Judah remained static; there was no strengthening of urban life (Lipschits 2003: 323–76; 2005: 206–71; 2011b: 188–94). 1.  On this subject, see: Hoglund 1991: 57–60; Carter 1994; 1999: 137–66; Lipschits 1999; 2003: 326–55; 2005: 240–61; Milevski 1996–97: 7–29; Faust 2007; and see the summary in Lipschits 2011a: 66–72; 2011b: 188–94. 2.  On the creation of the Idumean province, see: Lipschits, Shalom, Shatil, and Gadot 2014: 147–49.

237

238

Oded Lipschits

The Origin and History of the Persian Period Rural Settlement in Judah The rural settlement and the related economy that existed in Judah in the Persian and Hellenistic periods characterized that of the land as a whole until the establishment of the Hasmonean Kingdom in the late 2nd century b.c.e. Both settlement and economy were rooted in the settlement, economy, and administration of Judah after it became a vassal Assyrian kingdom; and especially in the history of the land after Sennacherib’s campaign (701 b.c.e.) and the loss of the Shephelah (Koch and Lipschits 2013: 66–67; and see further below); as well as the consequences of the Babylonian conquest of Jerusalem in the early 6th century b.c.e. and the transition of Judah from a vassal kingdom to a Babylonian province. During this period, the Judahite population that continued to subsist in the northern Judean highlands, as well as in the Benjamin region, continued to preserve its economy, ways of life, administration, and material culture. The area of Benjamin had become an important element of the Judean economy and administration in the 7th century b.c.e., probably after the Shephelah was lost. The biblical description of the Benjamin region during the Babylonian campaign against Jerusalem and following the destruction of the city, along with survey and excavation data of the main sites explored in this area attest that the region was not destroyed with Jerusalem. We may surmise that even before the fall of Jerusalem the Babylonians had chosen Mizpah (Tell en-Nasbeh) as the alternative capital of the Babylonian province and had already appointed Gedaliah as its first governor (Lipschits 1999; 2005: 68–125, 237–49). The area to the south of Jerusalem, with Ramat Raḥel at its center, probably had the same fate as the Benjamin region (Lipschits 2005: 250–58). The Rephaim Valley, with its rich alluvial soil and moderately terraced slopes has historically been one of the prosperous agricultural districts in the environs of Jerusalem, vital to the economy of the city. The mounting archaeological data from this area, underscored by the many agricultural installations and small farmsteads found in and around the valley confirm that the periods during which the Rephaim Valley flourished agriculturally are the same periods during which there was construction at Ramat Raḥel—that is, from the late Iron Age to the Persian period, with no signs of a hiatus (Lipschits and Gadot 2008: 88–96). There is clear evidence to associate the development of the Rephaim basin and the agricultural area to the south of Jerusalem, especially in the 7th century b.c.e.,3 to the emergence of Ramat Raḥel 3.  For the date of the establishment of the rural settlement around Jerusalem to the 7th century b.c.e., see Gadot 2015: 3–26.

Rural Economy and Settlement History

239

as an administrative center in the region under Assyrian rule and with the organization of royal estates in the Kingdom of Judah—probably after the period when Judah became an Assyrian vassal kingdom—and even more so after the loss of the Shephelah. The Rephaim basin and the rural settlement around Jerusalem flourished in the 7th century b.c.e. more than in any other period in the history of Judah (Gadot 2015: 3–26), and this settlement phenomenon fits the centralized processing demonstrated by the concentration of winepresses not associated with village infrastructure, the process of organized decanting and shipping of the wine, and the function of Ramat Raḥel in all of the periods in question as an administrative center in the region (Lipschits and Gadot 2008: 88–96; Lipschits, Gadot, Arubas, and Oeming 2014: 28–74; Gadot 2015). The archaeological data are clear: Judah continued to exist as a rural society when it became a Babylonian vassal kingdom and then a province. Its economy was based on agriculture and the production of agricultural products, just as it had been during the 7th century b.c.e. when it was an Assyrian vassal kingdom. Understanding the foundation for an agricultural-based economy and administration are essential to understanding the history of the region under the rule of the Assyrian, Babylonian, and Persian empires and the central role of Ramat Raḥel to this system. The corpus of stamped and incised jar handles found in Judah and especially in and around Jerusalem and Ramat Raḥel is a key to such an understanding. About 3,000 stamped jar handles were discovered in Judah during archaeological excavations and surveys of the 600 years when the kingdom and then the province of Judah were under the rule of the empires (Lipschits forthcoming). This is precisely the period when Ramat Raḥel existed as the region’s administrative and main collection center for agricultural products—primarily wine and oil stored in jars. No other Judahite site, not even Jerusalem, can challenge Ramat Raḥel’s record: Over 300 stamped handles from the late Iron Age were found at the site (Lipschits, Gadot, Arubas, and Oeming 2011: 16–17; idem 2014: 34–37, 39, 64), including lmlk and “private” stamp impressions (late 8th and early 7th centuries b.c.e., see Lipschits, Sergi, and Koch 2010: 3–32); concentric-circle incisions (mid-7th century b.c.e., see: Lipschits, Sergi, and Koch 2011: 7–8), and rosette stamp impressions (late 7th–early 6th centuries b.c.e., see Koch and Lipschits 2013: 60–61). In the Babylonian, Persian, and Hellenistic periods, Ramat Raḥel was the main center of stamped jar handles, with about 77 lion-stamped handles dated to the 6th century b.c.e. (Lipschits 2010: 17–19), more than 300 yhwd stamp impressions dated to the late 6th to mid-2nd centuries b.c.e. (Lipschits and Vanderhooft 2011: 107–10), and 33 yršlm stamp impressions dated to the 2nd century b.c.e. (Bocher and Lipschits 2013: 103–4). All in all,

240

Oded Lipschits

the phenomenon of stamped jar handles collected and stored at Ramat Raḥel continued for more than half a millennium in this steady administrative system.4 Just as the stamped-jar administrative and economic system endured throughout the 600 years of Judah’s existence under the rule of the empires, we can also assume that the system of agricultural settlements, royal estates, administrative centers, and districts in Judah (as is known from the Persian period) was the product of an already well-established system and that its roots were in the late 8th and mainly the 7th century b.c.e. It can also be assumed that the system of districts has a long history, when it also adapted to the circumstances of the existence of Judah as a province under Babylonian and Persian rule.

The Origin and History of the Persian Period Districts in Judah

ְ ‫ ֶּפל‬appears eight times in the Old Testament, all of them The term ‫ֶך‬ in Nehemiah 3 (vv. 9, 12, 14–18). Most scholars understand this term as “district,” basically after the primary meaning of the Akkadian pilku (von Soden 1965: 863).5 If this is the case, of the 41 people and groups who assumed responsibility for rebuilding each of the 41 segments of the walls of Jerusalem,6 7 segments were built under the responsibility and with the support of official functionaries who were in charge of districts or subdistricts: “Rephaiah, son of Hur, the official of half the district of Jerusalem” (v. 9); “Shallum, son of Hallohesh, the official of half the district of Jerusalem” (v. 12a); “Malchijah, son of Rechab, the official of the district of Beth-haccherem” (v. 14); “Shallun, son of Colhozeh, the official of the district of Mizpah” (v. 15); “Nehemiah, son of 4. The renewed excavations at Ramat Raḥel (Lipschits, Gadot, Arubas, and Oeming 2011: 16–17; 2014: 22–26) and the final publication of the architecture and finds from Aharoni’s excavations (Lipschits, Gadot and Freud forthcoming) have made it possible to reevaluate the archaeology of the site and its significance vis-ávis the political history of Judah as a province in the Achaemenid empire. 5.  Williamson (1985: 206); Blenkinsopp (1988: 235–36); Weinfeld (2000: 249); Lipschits (2005: 170–71, and n. 125). Demsky (1983: 242–44), followed by Graham (1984: 57) maintain that this is an administrative term that describes a group of laborers who have been conscripted for work duty. Bowman already commented on this subject (1954: 267) and was supported by Carter (1999: 80), but see the critique by Williamson (1985: 206), Naʾaman (1995: 21 n. 39), Weinfeld (2000: 249–50), and Cogan (2006: 89 n. 10). Only seven sections of the wall were built by this kind of “tax payer,” while all the other 34 sections were built by volunteers. It also necessitates the premise that there was a complex administrative system in which every district had supervisors for the conscripted work, and this indicates that in Persian period Judah there should have been a great deal of similar building projects. 6.  For a summary discussion of the list, see Lipschits 2012: 73–99.

Rural Economy and Settlement History

241

Azbuk, the official of half the district of Beth-zur” (v. 16); “Hashabiah, the official of half the district of Keilah—for his district” (‫;ל ִפ ְלּכֹו‬ ְ v. 17b, and see below); Bavvai, the son of Henadad, the official of half the district of Keilah” (v. 18). Five half-districts are mentioned in this list: two half-districts of Jerusalem, two half-districts of Keilah, and only one half-district of Bethzur (Naʾaman 1995: 21).7 Two other complete districts are mentioned: Beth-haccherem and Mizpah. In addition to the above districts and half districts, various cities that were within the borders of the districts were mentioned, among them Jericho (v. 2), Tekoa (vv. 5, 27), Gibeon and Mizpah (v. 7), and Zanoah (v. 13). One section of the wall was even built under the responsibility of Ezer the son of Jeshua who was “the official of Mizpah” (the city itself; v. 19). Overall, 11 of 41 sections of the wall were built by people and groups affiliated with a specific place or region within the province. This makes it difficult to see the list as a representative sampling of the settlements that existed at that time or to draw from it any clear-cut geographical and geopolitical conclusions.8 In this light, it seems that only a few officials of the province are mentioned in this list and probably other officials, administrators, and leaders of the nation did not take part in the endeavor.9 This has led to a dispute among scholars concerning the internal divisions of the province, the borders of each district, and which districts should be added to the list. Thus, for example, Klein (1939: 34), AviYonah (1984: 21), Blenkinsopp (1988: 232–33), and other scholars added one district (Jericho), and E. Stern (1982: 248–49) added 2 (Jericho and Gezer), extending the limits of the province westward.10 Aharoni (1962: 7.  Kallai’s deliberation (1960: 93) on the possibility that there was only one halfdistrict of Beth-zur is surprising. 8.  Understanding the list in this way, one should not accept the attempt by various scholars (Klein 1939: 3–4; Demsky 1983: 242; Ofer 1993: 1.37–38) to emphasize the list’s geographical significance, assuming that it is a representative sampling of the settlements that existed at the time of Nehemiah, and assembling the names of all the settlements that appeared on the list, without distinguishing among the various categories presented in the text (districts, settlements of origin for specific people, and settlements from which groups of builders came). 9.  This sheds light on the emphasis placed on the Tekoites and the gap between them and “their nobles” (3:5), especially considering that the Tekoites managed to finance “another section” (v. 27). In the same context, one may understand that the reference to the priests as “the men of the plain” (v. 22) is an allusion to the other groups of priests who did not support the project. 10. See Carter (1999: 90 n. 48), against Stern’s idea to include Gezer in the boundaries of the province.

242

Oded Lipschits

338) divided the district of Mizpah into 2 and postulated that only the Beth-haccherem district had not been divided. Blenkinsopp (1988: 232–33), in contrast, proposed that there were 6 districts, each one in turn divided into 2 (12 half-districts). However, the only geographical conclusion that may be drawn from the list is that there were 5 districts in Judah (Naʾaman 1995: 21). There is no evidence of any additional districts, and methodologically, one may certainly not add districts on the pretext of “filling in gaps” or with the preconceived notion of the limits of the province or of the limits of the districts within it.11 There is certainly no place for the assumption that the limits of the province extended beyond the boundaries of the preexilic Kingdom of Judah, particularly to the west and north,12 and any reconstruction must assume that the borders remained stable throughout the Babylonian and Persian periods.13

The Location and Borders of the Persian Period Districts in Judah The above basic positions, combined with the geographical analysis of the region, the information about the borders of Judah at the end of the Iron Age, and knowledge of the historical events in the Babylonian and Persian periods14 enable the following reconstruction of the districts that existed in the province of Yehud, their location, borders, and agricultural and economic potential. The Mizpah District was largely congruent with the First Temple region of Benjamin and reflected the continuity of the region’s boundaries between the 7th and 5th centuries b.c.e. The people of Jericho and Senaʾah probably also belonged to this district,15 but the attempt to in11.  On this subject, see Abel (1933–38: 2.120–22); Kallai (1960: 87–94), and see the summary by E. Stern (1982: 245–49) and his criticism of the attempt by scholars to solve the problems raised by the list in this way (1982: 247–48). 12.  See, e.g., E. Stern 1982: 245–49. In contrast, see Schwartz 1988: 4–5; Naʾaman 1995c: 19–23. 13.  There are exceptions to this view at various points, where scholars try to account for the geographical difficulty and to explain it by a temporary change in the borders. On the Lod-Ono region, see Bright 1959: 384; on the area of Gezer, see Kochman in Heltzer and Kochman 1985: 113. On this area in the Persian period, see Lipschits 1997. 14.  See the detailed discussion in Lipschits 2005: 149–84. 15.  This premise is opposed to the attempt by Klein (1939: 3–4), Avi-Yonah (1984: 21), and Blenkinsopp (1988: 232–33) to add the Jericho District as the sixth district in the province of Yehud, as well as E. Stern’s attempt (1982: 247–48) to add Gezer as another district in addition to Jericho. On Senaʾah, see: Lipschits 2005: 157.

Rural Economy and Settlement History

243

clude the Ono-Lod region or parts of Samaria in this area should be dismissed.16 Mizpah was probably the major city in the district.17 The historical boundary between the Benjamin region/Mizpah District and “the environs of Jerusalem”/Jerusalem District (see below) is the line created by connecting the southern settlements in the Benjamin tribal inheritance. Mozah and Kiriath-jearim are mentioned in the list of the cities of Benjamin (Josh 18:26, 28).18 Gibeah of Saul and Anathoth are added to the list of cities of Benjamin that appear in Isa 10:28–32.19 These cities create a line that lies 5 km from Jerusalem, in a northeast– northwest direction, and geographical logic would have it that this area was bounded on the northeast by Wadi Og and on the northwest by Wadi Soreq. The Jerusalem District lay south of the Mizpah District and probably included the city limits and its close environs. Taking into account the surrounding districts, we may surmise that the Jerusalem District extended eastward to the edge of the desert and to Mount Scopus in the north. To the south, the Jerusalem District did not extend beyond the Rephaim Valley, so its borders were very close (ca. 2–3 km) to the city itself. The Beth-haccherem District lay south of Jerusalem, with the Rephaim Valley at its core and Ramat Raḥel as its main administrative center.20 The people of Tekoa were probably included in this district, which like the two districts north of it, also reached the edge of the mountainous region on the west and was bordered by the Keilah District (Kallai 1960: 92–93, with additional sources).21 16.  See, e.g., Kallai’s proposal, 1960: 87–94; and in contrast, see Naʾaman 1995: 20–24; and Lipschits 1997, with additional bibliography. 17.  On the status of the Mizpah during the Babylonian and Persian periods, see: Lipschits 2005: 237–40. 18.  On the repetitive mention of Kiriath-jearim (Josh 15:60; 18:28) in the list of cities of Judah and Benjamin, see the explanation offered by Naʾaman 1991: 8–10. The place, in any case, was considered part of Benjamin (Josh 9:17; 18:28; Ezra 2:25; Neh 7:29). 19.  On Josh 18:28, see Naʾaman 1991: 8–10; and on the part of the list that was apparently omitted from the book of Joshua, see Naʾaman 1986: 229 n. 45; 1991: 25–26. The affiliation of Anathoth, Jeremiah’s birthplace, with the Benjamin tribal legacy is also highlighted at the height of the Babylonian siege, preceding the destruction of Jerusalem (Jer 32:7–15). 20.  On the identification of Ramat Raḥel as the biblical Beth-haccherem, see Lipschits and Naʾaman 2011: 65–86; Lipschits, Gadot, Arubas, and Oeming 2014: 12–14. 21.  One must dismiss Avi-Yonah’s suggestion (1984: 21) of including the Emmaus and Chephira regions in this district. These two regions were historically part of the Mizpah District. Similarly, Kallai’s proposal (1960: 94) to include the

244

Oded Lipschits

The two districts of Jerusalem and Beth-haccherem probably evolved from and were a split off the geographical unit known in the First Temple period as “the environs of Jerusalem.” This unit may be connected to the growth of towns, villages, hamlets, and agricultural installations to the west and south of Jerusalem during the late 8th and especially during the 7th century b.c.e. (Barkay 1985: 399–401, 466–67; Naʾaman 1991: 9, 13–14; and see recently Faust 2013: 13–14, with additional sources),22 which is the same period when similar growth can be seen in the region of Benjamin. This is probably the source of the distinction made by Jeremiah (17:26; 32:44; 33:13) between “the cities of Judah” and “the cities of the mountain” in the division of the areas of the kingdom into 6 regions (Barkay 1985: 399–401; Ofer 1993: 1.42). In agreement with Alt (1934: 324–28; cf. Naʾaman 1991: 15–16), “the cities of Judah” can be identified with the area to the south and west of Jerusalem, down to the Bethlehem region and seen as evidence of the distinction between the area surrounding Jerusalem and the area south of it. The region known as “the environs of Jerusalem” is also described in Jeremiah 40–42 as the area where Gedaliah the son of Ahikam was active after the destruction of Jerusalem by the Babylonians (Lipschits 2005: 152–54). It can be assumed that this geographical unit known as “the environs of Jerusalem” was divided by the Babylonians into two administrative units after the destruction of Jerusalem, when the main devastation was in the city and in its close periphery, and when the areas to the north of the city (the area of Benjamin) and to the south of it (the area to the west and to the south of Ramat Raḥel) continued to exist as before (Lipschits 2005: 206–18, 237–58). It was the historical period when the area around Jerusalem was separated and became an independent district (the District of Jerusalem), and the area to the south of it became the District of Beth-haccherem as indicated in the list of Nehemiah 3 and as mentioned again at the beginning of the list of “the sons of the singers” (Neh 12:28–29). The Beth-zur District is the southernmost area in the Judean Hills. On the east is En-gedi, and it is bounded on the west by the Keilah District. It is reasonable to assume that the border between these two districts northeastern Shephelah (the region of Gezer and Ekron) in this region should be rejected (Carter 1999: 90 n. 48). 22.  Gadot’s study (2015) makes it clear and strengthens previous similar opinions (see, e.g., Naʾaman 2007b: 42; Faust 2013: 22) that the main development of this area was during the 7th century and not in the late 8th century b.c.e., as was previously assumed. This conclusion has many important historical implications; see further below.

Rural Economy and Settlement History

245

ran east of the Adulam-Zanoah line. The southern border probably ran in the narrow strip between Beth-zur and Hebron, where it remained until the days of the Hasmoneans (Lipschits 2005: 146–49).23 The three districts in the Judean Hills, from the area north of Hebron to the area of Benjamin, reinstated the historical situation of the distinction between the southern Judean Hills, the center of which is Hebron, and the north-central Judean Hills, the center of which is Jerusalem, as can be reconstructed by historical and archaeological sources, perhaps even from the beginning of the second millennium b.c.e. There are deep economic, political, and social differences between these two areas, and these can be reconstructed from the archaeological data from the Late Bronze Age as well as from information on the settlement patterns of recent generations. The parallel settlement and social orientation of recent times supports the premise that the Hebron region is related to the southern region as far as the Beer-sheba–Arad valleys and that the area from Beth-zur northward is linked to the central mountain area (Naʾaman 1991: 361–80; Ofer 1993: 1.103–4, with additional bibliography). The Keilah District was the only stronghold that the province of Yehud had in the Shephelah, with the Valley of Elah and the city of Azekah in its southwestern corner. This district was bounded on the west by Ashdod, and to the south it was bounded by the region controlled from Maresha. On the north, the boundary did not extend beyond Gezer, which lies outside the limits of the province (Naʾaman 1995: 16–24; Carter 1999: 90 n. 48; Lipschits 2005: 147–48).

Agricultural and Economic Potential of the Districts in Judah in Light of the Topographic and Environmental Conditions, the Archaeological Data, and the History of the Region The District of Mizpah It is reasonable to assume that the growth of Jerusalem in the late 8th and 7th centuries b.c.e. brought about a strengthening of economic relations between Jerusalem and the region of Benjamin, inasmuch as “greater Jerusalem” needed a regular supply of agricultural produce. The grains that grew in the plains of the Benjamin region and the olive orchards and vineyards nearby were an important source of this supply. At the same time, the dissimilar fate of the area and the political features that made it distinctive on the eve of and subsequent to the 23.  I cannot accept Carter’s suggestion (1999: 98–99) that Hebron be included within the borders of the province.

246

Oded Lipschits

destruction of Jerusalem attest to the fact that the residents of this region preserved their distinction within the Kingdom of Judah, and, perhaps because of this, the Benjamin region became the center of the new province under Babylonian rule (Lipschits 2005: 72–112). This is probably the historical background for the prosperity of the Benjamin plateau during the 7th century b.c.e. Alongside the main sites in the region, such as Tell en-Nasbeh (Zorn 1993: 1099–1101), el-Jib (Pritchard 1962: 162–63), Tell el-Fûl (Lapp 1978: 39–46; Finkelstein 2011: 111–13), Khirbet el-Burj (De Groot and Winberger-Stern 2013: 95–102) and Nebi-Samwil (Magen and Dadon 2000: 62–63), the survey of this region also demonstrated a growth in the rural settlement in this area in the late Iron Age.24 From the administrative point of view, almost a quarter of the late lmlk-stamped jar handles and the concentric-circle-incised handles were found in the Benjamin region, mainly at el-Jib, Tell en-Nasbeh, and Khirbet el-Burj (Lipschits, Sergi, and Koch 2010: 21; Lipschits forthcoming). This vast number of 7th-century b.c.e. stamped handles leads to the conclusion that the Benjamin region was, for the first time, part of the Judahite administration. Furthermore, after Jerusalem and the area around Ramat Raḥel, it can now be said that the area of Benjamin was the third-most-important region of the Judahite administration. The centrality of the region probably resulted from the loss of the Shephelah after the Sennacherib campaign (701 b.c.e.): Judah had to focus its aims on substitute areas that remained within its boundaries, such as Benjamin to the north of Jerusalem and the northern Judean Hills to the south of the city, that evolved during the early 7th century b.c.e. (Lipschits and Gadot 2008: 88–96; Gadot 2015: 3–26). However, it was only for a short time, since during the last phase of the 7th century b.c.e. the area of Benjamin lost its administrative centrality; only 15 rosette-stamped handles (about 7% of the corpus), were unearthed at the Benjamin sites (Koch and Lipschits 2013: 59, 66–67). Since there was no change in the settlement history of this region at the end of the First Temple period and the demographic picture points to continuous prosperity in this area, it seems that the reason may lie elsewhere—in the Shephelah—and one can assume that, because of the renewed Judahite activity in that area, there was less need for agricultural supply from Benjamin. Thus, the region lost its importance to the administrative system, and its new dimensions resembled those of the early lmlk system, before the loss of the Shephelah.25 24.  For the survey’s data, see Magen and Finkelstein 1993. For analyses of the survey finds, see Milevski 1996–97: 7–29; Lipschits 1999: 180–84; 2005: 245–49. 25.  As for the reflection of this process in the biblical historiography, Naʾaman (2009: 116) connected the decrease in the status of Gibeon and Benjamin with Josiah’s reform (2 Kings 23). In his opinion, Josiah cancelled the sanctuary of Gibeon, and the

Rural Economy and Settlement History

247

The main conclusion from the archaeological excavations and surveys is that there was a continuation of settlement in this region during the 6th and 5th centuries b.c.e.—that is, from the late Iron Age to the Babylonian and then to the early Persian periods (Lipschits 1999: 180–84; 2005: 245–49).26 The relative prosperity of this area must be attributed to the political centrality assigned to Mizpah and the economic importance of Gibeon and Moza. It was probably around these settlements that the agricultural hinterland was based and the settlement pattern was shaped. The fact that 30 out of 43 mwṣh stamp impressions found in Tell en-Nasbeh27 were all well dated to the 6th century b.c.e. (Zorn, Yellin, and Hayes 1994: 166–68) and probably part of the local administration of the Babylonian province of Judah (Lipschits forthcoming) should be interpreted as part of the local supply to the governor of the province, whose seat was in Mizpah (Tell en-Nasbeh). That during this same period a different and much larger system of lion-stamped jar handles existed, with 71 stamped handles discovered in Ramat Raḥel, 29 in Jerusalem, and only 19 in the Benjamin area (12 in Nebi-Samwil, 5 at Tell en-Nasbeh, and 2 at Gibeon) is a clear indication that the main and separate administrative system of jar handles continued to exist just as it had in the system that was present before the destruction of Jerusalem. As with the rosette system at the end of the 7th century b.c.e., also in the lion system from the 6th century b.c.e., the area of Benjamin did not play a major role. During the Persian period, Nebi-Samwil became the main administrative place, together with Mizpah, while Gibeon and Khirbet el-Burj lost the major role they had in the system of the stamped jar handles during the 7th century b.c.e. local population was labeled “Gibeonite” in the royal historiography (Joshua 9) and therefore foreigners. 26.  It may be that by the 6th century b.c.e. the settlement in Benjamin had withdrawn to the core of the region, the narrow zone along either side of the watershed, while settlements in the northern and eastern zones almost entirely ceased to exist. This assumption accords with existing information about the complete cessation of settlement that took place throughout the Jordan Valley and the western littoral of the Dead Sea. It would seem that this, too, was related to the collapse of the economic, military, and political system of the Kingdom of Judah. Most of the surviving settlements in Benjamin were concentrated along the mountain ridge and the upper mountain slopes (in the central and western Benjamin region), around the main economic and administrative centers that continued to exist in the region. On this subject, see the description and model presented by Magen and Finkelstein (1993: 20–21) and the assessment of the surveyors as cited there (Magen and Finkelstein 1993: 138). 27. Four more mwṣh-stamped handles were discovered in Gibeon, 4 in Jerusalem, 2 in Jericho, 1 each in Ramat Raḥel and Tsubah, and the origin of another stamped handle is unknown. See Lipschits forthcoming.

248

Oded Lipschits

The Jerusalem District and the District of Beth-haccherem The land in the immediate environment of Jerusalem has a very low carrying capacity, the result of a steep and rocky terrain with poor soil coverage and very limited water sources. The result of these conditions was that the agricultural settlement connected to the city was not located in its immediate sphere but in an area concentrated around the Rephaim and Soreq Valleys with their many tributaries, about 2–4 km to the north, west, and southwest of the city (Gadot 2015: 3–26, with additional bibliography). Naḥal (= Wadi) Rephaim begins just west of the Old City, and from there it flows southward until it curves to the southwest just below Ramat Raḥel, creating the widest valley in the region. The mouth of Naḥal Soreq is located 4 km north of Jerusalem, and from there it flows westward, making a wide curve at the ancient site of Moza and continuing southward. In these valleys as well as in the rocky terrain between are rich alluvial soil and many small springs. This area was used for growing grain while the slopes of the ravines were usually used for cultivating vines (Gadot 2015: 3–26, with additional bibliography). It is not surprising that these periods of mounting human activities along the Rephaim and the Soreq Valleys correspond well to the periods of Jerusalem at its zenith: Middle Bronze II and Iron IIB–C (Gadot 2015). By focusing not only on surveys but especially on the many salvage excavations around Jerusalem, Gadot demonstrated that most of the sites in this area were settled for the first time during the 7th and not the 8th century b.c.e., as was previously suggested. He showed that, as against 11 sites that were built and existed in the 8th century, 62 sites existed in the 7th century b.c.e. It is important to note, however, that the growth in the number of sites is not an indication of dramatic growth in population but, rather, of specialized agricultural activities (Gadot 2015).28 The change in the 7th-century b.c.e. settlement pattern is mainly in the number of sites related to agricultural activities.29 It was the first time in the history of the Jerusalem periphery—to the northwest, west, southwest, and south—that so many different rural sites existed.

28.  All 11 of the sites in existence during the 8th century b.c.e. continued well into the 7th century; two new forts were added to the existing one in Giloh, and to the villages of el-Burj and Ras el ʿAmud. New villages were added at Er-Ras and possibly at Manhat and Binyanei Haummah. 29. Thirty-two of the new sites are single-building farmsteads, as compared with only one single farmhouse built in the 8th century b.c.e. Ten more sites were isolated agricultural installations (Gadot, 2015: 3–26, and table 1 there).

Rural Economy and Settlement History

249

It is not sufficient to connect this dramatic growth in the rural settlement around Jerusalem to the growth of the city itself, which had already begun to grow during the 8th century b.c.e. In concurrence with the opinions of Greenberg and Cinamon (2006), Faust (2007; 2013), Gadot (2015), and others, I consider it reasonable that this growth was partly the outcome of the 701 Assyrian campaign and the loss of the Shephelah. However, as stated above, Lipschits (2011a: 58–63; 2011b: 198–201; see also Lipschits and Gadot 2008: 88–96) also connected this growth to the development of the royal estates in the hill country (during the period after Judah became an Assyrian vassal kingdom) and the need to pay taxes to the Assyrian Empire, especially after the loss of the Shephelah. The growth of rural settlements to the west and south of Jerusalem should thus be associated with the existence of Ramat Raḥel and the development of the system of stamped jar handles during this same period, as part of the system that was developed in Judah to acquire more agricultural products, especially wine and oil, which could then be exchanged for metals (especially silver) that were paid as taxes to the ruling empire (Lipschits forthcoming). The transition of the rural settlement around Jerusalem from the late Iron Age through the Babylonian period and into the Persian period is marked by a smaller decline in the number of settlements than previously perceived (Lipschits 2005: 250–58, 261–71; Gadot 2015: 3–26; and see further below). Based on these observations, and, as mentioned above, according to the historical understanding of the processes in this region, the historical and settlement roots of the Persian period Jerusalem District together with the Beth-haccherem District were the outcome of the development and dividing of the geographical unit known in the First Temple period as “the environs of Jerusalem.” During the 7th century b.c.e., there was a well-defined growth in the number of rural settlements around Jerusalem, especially in the Rephaim and Soreq Valleys, not in the immediate sphere of the city. After the destruction of Jerusalem by the Babylonians, when the main destruction was in the city and in its close periphery, “the environs of Jerusalem” were divided into two different administrative units. The city and its close periphery ceased to exist, and it became (possibly already during the Babylonian period but certainly during the Persian period) “the district of Jerusalem.” It was cut off from its agricultural surroundings to the west and south of the city; these were still in existence but as a separate region that was administratively connected to Ramat Raḥel. This area became the District of Beth-haccherem (the ancient name of Ramat Raḥel), which flourished in the coming centuries as a rural area, just like the area to the north of the city (the area of Benjamin/District of Mizpah; Lipschits 2005: 118–26, 152–53).

250

Oded Lipschits

The importance and central status of the area around Jerusalem to the system of stamped jars is clear. About 65% of the stamped handles from the late lmlk system dated to the early 7th century b.c.e. were discovered in this area.30 About 70% of the concentric-circle incisions dated to the middle of the 7th century b.c.e. were discovered in this region.31 In the late 7th and early 6th century b.c.e., when there was slow restoration in the settlement of the Shephelah, the number of rosette-stamp impressions discovered in the area around Jerusalem went back to about 65%.32 After Jerusalem was destroyed and the territory of Judah was reduced mainly to the northern hill country, the number of stamped jar handles discovered mainly at Ramat Raḥel (but also in Jerusalem and the area around it) totaled 80% of the lion system33 and 87% of the early yhwd system.34 These statistics underscore the importance of the rural settlement and the administrative role of the area around Jerusalem during the Persian period after it was divided into two districts—the Jerusalem District and the Beth-haccherem District. The Beth-zur District The area to the south of the Beth-haccherem District was never an important part of the economy and administration of the Kingdom (and then the province) of Judah. It was the least populated area in the hill country (Ofer 1993: 1.110–11). Its economic potential was much more limited, when the large areas on the desert fringe were used for pastoral economy, and the higher areas facing the central hills and the western slopes were used mainly for growing grain by dry farming (Finkelstein 1986: 121–26; Ofer 1993: 1.111–13). The conclusion is that the economy of the central and southern hill country was based mainly on pasture and dry farming. The economic output of the vineyard areas and olive orchards was limited, much more so than in the areas of the two districts to the north. 30.  Of 486 stamped handles found in and around Jerusalem, 319 were found in the area around Jerusalem, 180 in Jerusalem, and 133 at Ramat Raḥel. See Lipschits, Sergi, and Koch 2011: 30–35, and table 2; Lipschits forthcoming. 31.  Of 282 incised handles found in and around Jerusalem, 197 were found in the area around Jerusalem, 146 were found in Jerusalem, and 40 at Ramat Raḥel. See Lipschits, Sergi, and Koch 2011: 31, 35, and table 3; Lipschits forthcoming. 32.  Of 247 rosette-stamped handles found in and around Jerusalem, 155 were found in the area around Jerusalem, 87 of them in Jerusalem, and 57 at Ramat Raḥel. See Koch and Lipschits 2013: 57–58; Lipschits forthcoming. 33. Of 130 lion-stamped handles found in and around Jerusalem, 104 were found in the area around Jerusalem, 29 in Jerusalem, and 71 at Ramat Raḥel. See Lipschits 2010: 17–19; forthcoming. 34.  Of the 128 early yhwd-stamped handles found in the area in and around Jerusalem, 111 were in the area around Jerusalem, 17 in Jerusalem, and 90 in Ramat Raḥel. See Lipschits and Vanderhooft 2011: 11–15, 19; Lipschits forthcoming.

Rural Economy and Settlement History

251

The limited agricultural and economic potential of this area may explain why it is nearly absent from the stamped-jar system. Only 2 early lmlk-stamped handles from the late 8th century b.c.e. were found in this region (one of them in Hebron), and only 14 late lmlk-stamped handles from the early 7th century b.c.e. (6 in Hebron and 8 in Beth-zur), and 2 incised handles from the middle of the 7th century b.c.e. were discovered in this region (Lipschits, Sergi, and Koch 2011: 30–35).35 From the late 7th and the early 6th century b.c.e., 6 rosette-stamped handles were discovered in this area (5 of them in Beth-zur; see Koch and Lipschits 2013: 59) and from all the other systems of stamped handles, dated to the 6th century b.c.e. and on, not even 1 stamped handle was discovered. This is the background against which we begin to understand the settlement history of this region. Even though the archaeological data from the excavations and surveys are minor and limited,36 they seem to bolster the premise of a substantial difference between the settlement and demographic processes that transpired at the beginning of the 7th century b.c.e. in the southern Judean highlands and those that transpired to the north (Lipschits 2005: 224–32, with additional bibliography). These facts seem to indicate a constant increase in the number of settlements and in the overall settled area in the entire mountain region until the late 8th century b.c.e. (Ofer 1993: 2.121–26; and see the summary by Finkelstein 1994: 174–75). Apparently the entire region was dealt a severe blow during Sennacherib’s campaign. In contrast to the rapid recovery of the region north of Hebron (and primarily in the northern Judean highlands), the area south of Hebron was slow to mend; it recovered only partially during the 7th century b.c.e., and settlement in all this area remained sparse.37 The major sites excavated in this area did not yield any tangible evidence of an early 6th-century b.c.e. destruction, but at least some of them indicated a settlement gap throughout the 6th century b.c.e. and even in the early Persian period.38 The results of surveys show what was 35.  Six more unidentified lmlk-stamped handles were discovered in Hebron, and 3 in Beth-zur (Lipschits, Sergi, and Koch 2011: 30–35). 36.  For a summary of the settlement picture in the Judean highlands at the end of the Iron Age and the Persian period, see Ofer 1993: 4.16–18; 1998: 46–48; Lipschits 1997: 291–99; 2005: 224–32. 37. See Ofer 1993: 2.125–34, 138–42; 4.13–18; 1998: 46–48; however, see also the critique by Finkelstein 1994: 174–75. 38.  At Hebron (Tell er-Rumeida), the circumstances of the demise of the settlement at the end of the Iron Age were not clarified, but it seems that the tell was abandoned in the 6th century b.c.e., and there was a settlement gap throughout the Persian period (Ofer 1989: 91; 1993: 2.59–60, 132; NEAEHL 2.478). There is no evidence that the settlement had migrated in the Persian period to its new site in

252

Oded Lipschits

apparently marked settlement dwindling farther south, from Bethlehem to Beth-zur (Lipschits 2005: 224–32). From the geopolitical and historical perspectives, it is likely that, after the destruction of Jerusalem, the status of the southern hills of Judah and the southern Shephelah was not defined, and these areas remained sparsely populated and relatively desolate. At this stage, a growing distinction was created between the area south of Hebron and Maresha, which remained stricken and relatively uninhabited, and the area north of them, which was less affected and where a large share of the settlements carried on as they had before. As mentioned above, the beginning of the separation process between these two regions took place during the 701 b.c.e. Assyrian campaign, when settlement in the southern hills of Judah and the Shephelah dwindled noticeably and was never rehabilitated. In the Judean highlands, a clear distinction was created between the areas north and south of Hebron. North of Hebron, the settlement had returned to its previous dimensions by the mid-7th century b.c.e. South of Hebron, the process of rebuilding was slower, and the size of the settlement never reverted to what it had been before the Assyrian campaign (Ofer 1993: 4.51–61; Lipschits 2005: 224–32). In the Shephelah, the settled area was concentrated at the foot of the mountain, and there was a marked ebbing of the population that steadily persisted through the western and southern parts of the region (Lipschits 2005: 218–23, with additional bibliography). The “collapse of systems” in the Beer-sheba and Arad Valleys probably began during the Babylonian siege of Jerusalem and ended with the fall of the city (Lipschits 2005: 228–32). Following the destruction, pastoral groups—especially Edomites and Arabs—from the south began to invade the Negev. In a process that was partially an explosive and partially a gradual and quiet incursion, these groups began to advance northward toward the Shephelah and the Hebron highlands. This mithe valley at the foot of the tell, and it seems that the settlement activity there began only in the Hellenistic period. At Tell Rabud (Debir), a settlement stratum was uncovered that was dated to the 7th and the beginning of the 6th century b.c.e. (Kochavi 1973: 55–57). This stratum may have been associated with traces of a fire, but there is no tangible proof of destruction, and the circumstances of its abandonment are not clear. Nonetheless, it seems that there was a settlement gap at the site that lasted until the early Persian period (Kochavi 1973: 55, 59, 61; and see also NEAEHL 4.1440). Remnants of a settlement from the end of the 7th and beginning of the 6th century b.c.e. were excavated at Beth-zur (Sellers et al. 1968: 2–13; and see also Albright 1943: 8 and n. 9; Funk 1958: 14; Ofer 1993: 70–72). Here, too, there was no tangible evidence of a destruction at the beginning of the 6th century b.c.e. (Funk 1958: 14; NEAEHL 1.197), but many scholars believe that there was a settlement gap that lasted until the beginning of the 5th century b.c.e. (Sellers and Albright 1931: 9; Sellers 1933: 43). Stern agrees with this view (E. Stern 1982: 36–38; 2001: 437–38), as does Funk (1992: 1.197).

Rural Economy and Settlement History

253

gration may be linked to the noticeable devastation of the settlements as well as the decreased population in the Negev, the southern Judean highlands, and the southern and central Shephelah.39 The population flow toward the north was curbed only in the late Persian period. At some stage during that period, the boundaries of the provinces in the region were redefined, apparently on the basis of the marked borders that had previously existed, with pinpoint adjustments to fit the demographic situation. Thus, the southern and southwestern borders of the provinces of Yehud were shaped. This border continued to exist until the Hellenistic period, with the beginning of Hasmonean expansion. The information available on the regions occupied by the Hasmoneans, or those granted to them, supports the information on the borders that existed until this period. Archaeological evidence shows settlement in the central mountain region, indicating that a new settlement prototype, different from that during the Iron Age, had developed. One can assume that in the mid-5th century b.c.e. the southern boundary of Yehud had already been charted, and the settlement in this area had been established. The Keilah District The Shephelah is the best area in Judah for cultivating olive trees, and it was traditionally the center of olive-oil production, as indicated from the history of this region in the 8th and 7th centuries b.c.e. This is why the majority of the lmlk jar handles of the early types, dated to the late 8th century b.c.e., were found in the Shephelah, mostly at Lachish (Lipschits, Sergi, and Koch 2011: 30–34, with additional bibliography). However, it seems that this area had never recovered and never regained its importance after the 701 Assyrian blow. Only 20 late lmlk-stamped handles were found in the Shephelah, 11 of them in Tell Socoh, which was probably the main production center of the lmlk jars and which was still in existence and producing jars in the early 7th century b.c.e. (Lipschits, Sergi, and Koch 2011: 30–34; Lipschits and Amit 2011: 182). The number of incised handles with concentric circles, dated to the middle of the 7th century b.c.e. is even lower, and includes only 8 handles (4 of them were discovered at Lachish and 2 at Azekah; and see Lipschits, Sergi, and Koch 2011: 31, 35). However, a change occurred in the late 7th century b.c.e., when the Shephelah sites yielded 51 stamped handles, which is about 21.5% of the corpus.40 This distribution indicates that, in addition to the 39.  Based on literary sources contemporary with the Persian period as well as later sources, one should assume that at least part of the Judean population remained to inhabit these areas (Lipschits 2005: 149–80, 228–32). 40.  Of the 51 stamped handles, 24 were retrieved at Lachish (10.5% of the corpus), 10 at Azekah (4%), 6 at Tell Batash (3%), 5 at Socoh, and 6 at other sites in that

254

Oded Lipschits

prominent position of Jerusalem and Ramat Raḥel in this system, the Judean Shephelah again became an important part of it, and Lachish once again found itself the center of the Shephelah (Ussishkin 2004: 90– 92; Naʾaman 2007a: 232). Apparently, Lachish Level II was founded in the same period when the rosette system was introduced, but this city was just a faint shadow of its predecessor, and the number of stamped handles is only one more indication of its status and the status of the Shephelah in the system of stamped handles (Koch and Lipschits 2013: 59). This change in the status of the Shephelah and its return to the stamped-jar system was short lived. The heavy destruction of all the major cities in this area is a clear archaeological fact (Lipschits 2005: 218–23), and from the surveys conducted in the region it seems that, during the 6th century b.c.e., there was a great settlement vacuum in the Shephelah, and an entirely new settlement pattern arose in the region during the Persian period. In any case, the area was no longer part of the provincial administrative system of the stamped jar handles. Only a few random yhwd-stamped jar handles were discovered in this region, mostly at sites that were beyond the borders of the province, and especially at Gezer (2 early, 1 middle, and 5 late yhwd-stamped handles; and see Lipschits and Vanderhooft 2011: 11–22).41 As the only stronghold that the province of Yehud had in the Shephelah during the Persian period, this district was in proximity to the flourishing coastal area. The important roads leading from Judah westward also passed through this area, crossing the important border towns, including Azekah. The functions of the Keilah District probably made it one of the more strategically important areas in the province, even though its economic potential was far from fully utilized and though its population was very small. In addition, the pottery production center in Socoh was abandoned at this point, and the center of production of the Yehud jars moved to the environs of Jerusalem (Lipschits and Vanderhooft 2011: 59–61). Apparently the province of Yehud could not exploit the economic potential of this region because of the lack of human resources or because of the strong neighbors who resided to the west (Ashdod), to the north (Samaria), and to the south (expanding Idumea). Beyond all this, it seems that the weakness of the Keilah District is a clear sign of the traditional settlement and economy of Judah, region. See Koch and Lipschits 2013: 59. 41.  On the geopolitical processes that shaped the borders and created the provincial system of the late Persian and Hellenistic periods in this region, see above, as well as in Lipschits, Shalom, Shatil, and Gadot 2014.

Rural Economy and Settlement History

255

which continued to preserve the 7th-century b.c.e. model of the post-Sennacherib period and could not restore the glorious days of the Shephelah after the heavy destructions and the appropriation of the southern part of this region.

The Size and Role of the Districts in the Persian Period A total of 238 sites have been found that were located in the various districts of Judah during the Persian period, nearly all of them small and rural.42 Seventy-six of these (32% of the total number) were located in Benjamin, in the District of Mizpah; 62 (26% of the total number), in Jerusalem and its surroundings (the District of Jerusalem and the District of Beth-haccherem; 49 (20.5%) in the northern Shephelah; and 51 (21.5%) in the Judean hills. The total number of sites and the number of sites in each district reflect the catastrophe set in motion by the Babylonian campaign in the early 6th century b.c.e. However, of this number, only 28% are new sites that were built in a new place in the Persian period, while 72% are carry-overs from the Iron Age. The rate of continuity in the different regions is quite similar—about 70–75%. Although it is possible that some sites were reestablished after a settlement gap in the 6th century b.c.e., this figure strengthens the argument toward continuity of the rural settlement all across Judah after the Babylonian destruction despite the destruction of Jerusalem and the military and urban system of Judah, the exile of the elite to Babylon, and the transforming of the kingdom into a province (Lipschits 2011b, with additional bibliography). The conclusion from these statistics, as well as from the above discussion is that the rural economy and the district system in the Persian period were the outcome of a long process of development in Judah, especially during the time it became an Assyrian vassal kingdom in the late 8th century b.c.e. The main events that influenced the rural economy and the history of the Persian period districts were the 701 b.c.e. Assyrian campaign and the 586 Babylonian destruction of Jerusalem. The weakening of the southern and southwestern areas of the kingdom in 701 b.c.e., especially the southern Shephelah, the Negev area, and the southern Judean Hills, and the final excision of these areas from the Babylonian and Persian province of Yehud shaped the drastic changes that influenced especially the southern and southwestern districts— Beth-zur and Keilah. 42.  These numbers were collected already in Lipschits 1997; an updated database of the Persian and early Hellenistic periods is being gathered as part of the Nitsan Shalom research for his M.A. thesis at Tel Aviv University.

256

Oded Lipschits

Conclusions and Hypothesis about the Rural Economy, District System, and Role and Location of the Persian Governor In light of the data discussed above, I suggest that during the Persian period each district in the Yehud province had its own center and an economic and administrative position and role in the overall provincial system. At the center of the Beth-zur District was a military fort that bordered Idumea. The region had low economic potential, and its primary role was probably to guard the southern zone against the nomadic tribes that had begun to establish themselves in this region and who by the end of the Persian and the early Hellenistic periods became part of the new province of Idumea. The Keilah District also probably served a protective role, controlling the main roads leading south and west from the province, and facing the border to the west, toward the province of Ashdod and toward the area of Maresha in the south. At the end of the Persian era and during the early Hellenistic period, this area became the center of the Idumean province (Lipschits, Shalom, Shatil, and Gadot 2014: 146–49). In the years following the crisis of the Babylonian destruction, there was stability and continuity in the rural settlement and economy in the central areas of the province. Apparently, with no dramatic external events, this rural settlement could survive and maintain its existence. This stability was also in the interest of the ruling empires, which preserved the local system of collecting agricultural products, especially wine and oil, in one main administrative center—Ramat Raḥel—that existed throughout the entire period that Judah was under the hegemony of foreign empires. During the entire Persian period, Jerusalem was only a small city, nothing more than a temple with a few hundreds priests and other temple servants living around it (Lipschits 2011b: 189–90, with additional bibliography). This is why all of the agricultural sites and their locations in the central and northern hill country are an indication that the source and reason for them were not the city or the temple but the administrative and economic system that survived in Judah, with its center at Ramat Raḥel. Three different districts existed in this area in the Persian period, with a specific role for each: Jerusalem was the cultic center of the province, where the temple and the center of the district around it were located. Mizpah became the political center of the province after the destruction of Jerusalem and already in the Babylonian period became the seat of the local governor, as well as the center of the district around it. Ramat Raḥel continued to function as the main administrative collection center for agricultural produce, and after the destruction of Je-

Rural Economy and Settlement History

257

rusalem also became the center of the district that included most of the rural settlements to the west and south of Jerusalem. Each site— Jerusalem, Mizpah, and Beth-haccherem—had its specific function, and each site also functioned as the center of its district. However, interesting changes occurred during the Persian period in this system of three centers and three districts with three different functions: Jerusalem’s status strengthened as the central and only cult place, and Mizpah lost its status as the seat of the governor. Ramat Raḥel’s status and position changed as well: besides being the collection center for agricultural produce, it also became the seat of the local governor. Ramat Raḥel was built under Assyrian rule as an administrative center and possibly also as the place where the Assyrian qepu could inspect the vassal kingdom and supervise its loyalty and the payment of taxes. Its magnificence, location, unique architectural items, and many stamped jar handles are a clear indication of the role of this site during the 7th and early 6th century b.c.e. (Lipschits, Gadot, Arubas, and Oeming 2014: 30–44). The site continued to function in the same way and with no change or gap under Babylonian rule. It seems that, during this period, when Judah became a province instead of a kingdom, the duality between the imperial inspection and presence (at Ramat Raḥel) and the local leadership (the king) and priesthood (in Jerusalem) became a kind of threefold system: Ramat Raḥel continued to symbolize the imperial presence and functioned as the main collection center for agricultural products; the local leadership, now the governor, was seated in Mizpah;43 and the main cult place was probably Jerusalem, even though it may be that during the 6th century b.c.e. the Jerusalem monopoly was not very strong, and other cult places existed (Bethel? Gibeon? Mizpah? Shechem?). The two different administrative systems—the local system of the governor in Mizpah and the imperial system that continued to function at Ramat Raḥel—are reflected in the duality of the mwṣh-stamped jar handles, most of them discovered in Mizpah, probably marking agricultural products from the governor’s estate in mwṣh; and the lion-stamped handles, most of them discovered in Ramat Raḥel, probably serving as part of the province’s taxes, continuing the same system and method as they had previously, in the 7th century b.c.e. The archaeological data from the renewed excavations at Ramat Raḥel demonstrate that the edifice that stood on the top of the hill in the center of the Beth-haccherem District reached the zenith of its 43. See Blenkinsopp 1988: 42 and n. 48; Lemaire 1990: 39–40; 2003: 292; Lipschits 2006: 34–37.

258

Oded Lipschits

activity during the 5th–4th centuries b.c.e. (Lipschits, Gadot, Arubas, and Oeming 2014: 77–88; Lipschits, Gadot, and Langgut 2012). It was the only structure in the province of Yehud with royal architecture and grandeur and was surrounded by a well-watered imperial garden, with unique flora that originated in the Persian homeland and surrounding areas. The structure unquestionably symbolized the power and affluence of the local rulers who represented the Persian Empire. It seems that, with this combination of magnificent building and luxurious garden, there is no reason to doubt that during the Persian period Ramat Raḥel served as the edifice of the imperial governor in the province. Only during the Persian period, when Jerusalem renewed its status as the lone cultic site, the location of the temple, and the seat of the priests did Mizpah lose its status as the seat of the local governor, it was partially deserted, and Ramat Raḥel became both the seat of the governor and the main administrative center for collecting taxes. The District of Mizpah was the demographic center of Judah during the 6th century and at least until the middle of the 5th century b.c.e., with about one-third of all sites in Judah (76 of 238). The system of mwṣh-stamped jar handles and the absence of the lion- and yhwdstamped jars from this region indicate that the agricultural economy of this area was aimed at maintaining the governor in Mizpah. In light of this, one can understand the relative prosperity of the region and the establishment of many new rural settlements in it. The gradual impoverishment of the settlement in Benjamin took place in the 5th century b.c.e. (Lipschits 1997: 17–31; 1999: 182–85), probably after the period when Jerusalem based itself as the central and only cultic center in Judah, and the seat of the governor moved to Ramat Raḥel. These changes undermined the economic base of the settlements in the region, resulting in a marked demographic decline. It is likely that the decline in the Benjamin region holds the explanation to the strengthening of the settlement in the northeastern part of the Shephelah (the area between Lod-Ono and Modiʾin) during the Persian period (Lipschits 1997b: 8–32). The political stability, particularly the enhanced economic activity in the coastal area during the Persian period attracted many inhabitants of the hill country to settle nearby. This was in addition to the agricultural potential of the area and its proximity to the major coastal trade routes and the major roads to the Benjamin region and to the area of Jerusalem. Aside from this, one cannot preclude the possibility that this migration had ideological motivation against the new ruling elite in Jerusalem, which forced its religious, ritual, social, and national views on the residents of the province. This, in turn, could have propelled some of the inhabitants of Benjamin to migrate beyond the administrative limits of the province and to settle close to the border.

Rural Economy and Settlement History

259

This is the key to understanding 1 Chr 8:12–13, according to which the families of Elpaal and Beriah, whose origins were in Benjamin, “built Ono and Lod” and “were heads of the fathers of the inhabitants of Aijalon.” Historical sources from the Persian period contain several accounts that lend support to this demographic phenomenon. Particularly striking are the marital connections of “the people of Israel” with the foreign women who came from among the “people of the lands” (Ezra 9:1–2, 12–15; 10:2–3, 10–15; Neh 10:31). This also provides a framework for understanding the proposition made by Sanballat, the Horonite, and Geshem, the Arab, to meet with Nehemiah “in one of the villages in the plain of Ono” (Neh 6:2), where a Judean settlement whose political allegiance was not clear was located. This is apparently the source of the Jewish population that lived in this area during the Hellenistic period (Lipschits 1997b: 10–11, 28–31).

Bibliography Abel, F. M. 1933–38 Géographie de la Palestine. Paris: Gabalda. Aharoni, Y. 1962 The Land of Israel in Biblical Times: A Historical Geography. Jerusalem: Mosad Bialik. [Hebrew] Albright, W. F. 1943 The Excavation of Tell Beit Mirsim, vol. 3: The Iron Age. AASOR 21–22. New Haven, CT: American Schools of Oriental Research. Alt, A. 1934 Die Rolle Samarias bei der Entstehung des Judentums. Pp.  5–28 in Festschrift Otto Procksch zum 60. Geburtstag, ed. A. Deichert and J. Hinrichs. Leipzig. Repr. KS 2 (1953): 316–37. Avi-Yonah, M. 1984 Historical Geography of Palestine From the End of the Babylonian Exile up to the Arab Conquest. Jerusalem: Mosad Bialik. 4th ed. [Hebrew; original pub. date, 1949] Barkay, G. 1985 Northern and Western Jerusalem in the End of the Iron Age. Ph.D. dissertation. University of Tel Aviv. [Hebrew] Blenkinsopp, J. 1988 Ezra, Nehemiah. OTL. Philadelphia: Westminster. Bocher, E., and Lipschits, O. 2013 The Corpus of yršlm Stamp Impressions: The Final Link. TA 40: 99–116. Bowman, R. A. 1954 Ezra and Nehemiah. Pp. 551–819 in vol. 3 of IB. Carter, C. E. 1994 The Province of Yehud in the Post-Exilic Period: Soundings in the Site Distribution and Demography. Pp.  106–45 in Second Temple

260

1999

Oded Lipschits Studies, vol.  2: Temple and Community in the Persian Period, ed. T.  C. Eskenazi and K. H. Richards. JSOTSup 175. Sheffield: JSOT Press. The Emergence of Yehud in the Persian Period: A Social and Demographic Study. JSOTSup 294. Sheffield: Sheffield Academic Press.

Cogan, M. 1986 Raising the Walls of Jerusalem (Nehemiah 3:1–32): The View from Dur-Sharrukin. IEJ 56: 84–95. De Groot, A., and Winberger-Stern, M. 2013 Wine, Oil and Gibeonites: Iron Age II–III at Kh. el-Burj, Northern Jerusalem. Pp. 95–102 in New Studies on Jerusalem, vol. 19, ed. E. Baruch and A. Faust. Ramat-Gan: Bar-Ilan University Press. [Hebrew] Demsky, A. 1983 Pelekh in Nehemiah 3. IEJ 33: 242–44. Faust, A. 2007 Settlement Dynamics and Demographic Fluctuations in Judah. Pp. 23–51 in A Time of Change, ed. Y. Levin, London: T. & T. Clark. 2013 Demographic and Territorial Expansion. Pp. 7–38 in New Studies on Jerusalem, ed. E. Baruch and A. Faust. Ramat-Gan: Bar Ilan University. Finkelstein, I. 1986 The Archaeology of the Israelite Settlement. Tel Aviv: Kibbutz Hameuchad / Jerusalem: Israel Exploration Society. [Hebrew] 1994 The Archaeology of the Days of Manasseh. Pp. 169–87 in Scripture and Other Artifacts: Essays on the Bible and Archaeology in Honor of Philip J. King, ed. M. D. Coogan, J. C. Exum, and L. E. Stager. Louisville: Westminster John Knox. 2011 Tell el-Ful Revisited: The Assyrian and Hellenistic Periods. PEQ 143: 106–18. Funk, R. W. 1958 The 1957 Campaign at Beth-Zur. BASOR 150: 8-20. Gadot, Y. 2015 Setting the Chronology Straight: Jerusalem, Ramat Raḥel and the Rural Hinterland in the 8th–4th Centuries bce. TA 42: 3–26. Greenberg, R., and Cinamon, G. 2006 Stamped and Incised Jar Handles from Rogem Ganim and Their Implications for the Political Economy of Jerusalem, Late 8th–Early 4th Centuries bce. TA 33: 229–43. Heltzer, M., and Kochman, M., eds. 1985 Encyclopedia of the World of the Bible: Ezra and Nehemiah. Jerusalem: Revivim. [Hebrew] Hoglund, K. G. 1991 The Achaemenid Context. Pp. 22–53 in Second Temple Studies, vol. 1: Persian Period, ed. P. R. Davies. Sheffield: Sheffield Academic Press. Kallai Z. 1960 The Northern Boundaries of Judah: From the Settlement of the Tribes until the Beginning of the Hasmonaean Period. Jerusalem: Magnes. Kelso, J. L., et al. 1968 The Excavation of Bethel 1934–1960. AASOR 39. Cambridge, MA: American Schools of Oriental Research.

Rural Economy and Settlement History Klein, S. 1939

261

The Land of Judah from the Days of the Return from Babylon to the Sealing of the Talmud. Tel Aviv: Devir. [Hebrew]

Kloner, A. 2000 Survey of Jerusalem: The Southern Sector. Jerusalem: Israel Antiquities Authority. [Hebrew and English] Kloner, A., and Stern, I. 2007 Idumea in the Late Persian Period (Fourth Century b.c.e.). Pp. 139–44 in Judah and the Judeans in the Fourth Century b.c.e., ed. O. Lipschits, G. N. Knoppers, and R. Albertz. Winona Lake, IN: Eisenbrauns. Koch, I., and Lipschits, O. 2013 The Rosette Stamped Jar Handle System and the Kingdom of Judah at the End of the First Temple Period. ZDPV 129: 55–78. Kochavi, M. 1973 Khirbet Rabud: Ancient Debir. Pp. 49–76 in Excavations and Studies: Essays in Honour of Professor Shmuel Yeivin, ed. Y. Aharoni. Tel Aviv: Institute of Archaeology, Tel Aviv University, and Carta [Hebrew]. English version: Khirbet Rabud = Debir. TA 1 (1974): 1–3. Lapp, N. L., ed. 1978 The Third Campaign at Tell el-Ful: The Excavations of 1964. AASOR 45. Cambridge, MA: American Schools of Oriental Research. Lapp, P. W. 1965 Tell el-Full. BA 28: 2–10. Lemaire, A. 1990 Populations et territoires de Palestine à l’époque perse. Transeuphraténe 3: 31–74. Lipschits, O. 1997a The “Yehud” Province under Babylonian Rule (586–539 b.c.e.): Historic Reality and Historiographic Conceptions. Ph.D. dissertation. Tel Aviv University. [Hebrew] 1997b The Origins of the Jewish Population of Modiʾin and Its Vicinity. Cathedra 85: 7–32. [Hebrew] 1999 The History of the Benjaminite Region under Babylonian Rule. TA 26: 155–90. 2001 Judah, Jerusalem and the Temple (586–539 b.c.e.). Transeuphraténe 22: 129–42. 2003 Demographic Changes in Judah between the Seventh and the Fifth Centuries b.c.e. Pp.  323–79 in Judah and the Judeans in the Neo-Babylonian Period, ed. O. Lipschits and J. Blenkinsopp. Winona Lake, IN: Eisenbrauns. 2004 The Rural Settlement in Judah in the Sixth Century b.c.e.: A Rejoinder. PEQ 136: 99–107. 2005 The Fall and Rise of Jerusalem. Winona Lake, IN: Eisenbrauns. 2006 Achaemenid Imperial Policy, Settlement Processes in Palestine, and the Status of Jerusalem in the Middle of the Fifth Century b.c.e. Pp.  19–52 in Judah and the Judeans in the Persian Period, O.  Lipschits and M. Oeming. Winona Lake, IN: Eisenbrauns.

262 2007

Oded Lipschits

The Babylonian Period in Judah: In Search of the Half Full Cup—A Response to the Panel Discussion: In Conversation with Oded Lipschits, The Fall and Rise of Jerusalem. JHebS 7: 39–47. 2009 Persian Period Finds from Jerusalem: Facts and Interpretations. JHebS 9: 2–30. 2010 First Thought Following the Corpus of the Lion Stamp Impressions. Pp. 17–19 in New Studies in the Research of the Lion Stamp Impressions from Judah, ed. O. Lipschits and I. Koch. Tel Aviv: Tel Aviv University Press. 2011a Shedding New Light on the Dark Years of the “Exilic Period”: New Studies, Further Elucidation, and Some Questions Regarding the Archaeology of Judah as an “Empty Land.” Pp. 57–90 in Interpreting Exile: Interdisciplinary Studies of Displacement and Deportation in Biblical and Modern Contexts, ed. B. Kelle, F. R. Ames, and J. L. Wright. AIL. Atlanta: Society of Biblical Literature. 2011b Persian Period Judah: A New Perspective. Pp. 187–212 in Texts, Contexts and Readings in Postexilic Literature: Explorations into Historiography and Identity Negotiation in Hebrew Bible and Related Texts, ed. L. Jonker. FAT 2/53. Tübingen: Mohr Siebeck. 2012 Nehemiah 3: Sources, Composition and Purpose. Pp. 73–99 in A New Perspective on the Ezra–Nehemiah Story and History, Literature and Interpretation, ed. I. Kalimi. Winona Lake, IN: Eisenbrauns. forthcoming  Judah under Assyrian, Babylonian and Persian Rule: A New Look at Its History and Administration in Light of the Stamped Jar Handles. [Hebrew] Lipschits, O. 1997a The “Yehud” Province under Babylonian Rule (586–539 b.c.e.): Historic Reality and Historiographic Conceptions. Ph.D. dissertation. Tel Aviv University. [Hebrew] Lipschits, O., and Amit, D. 2011 18 Stamped Jar Handles Not Published So Far. Pp. 179–98 in New Studies in Jerusalem 17, ed. E. Baruch, and A. Faust. Ramat-Gan: Bar Ilan University. [Hebrew, with English summary on pp. 54*–55*] Lipschits, O.; Gadot, Y.; Arubas, B.; and Oeming, M. 2011 Palace and Village, Paradise and Oblivion: Unraveling the Riddles of Ramat Raḥel. NEA 74: 2–49. 2014 What Are the Stones Whispering? 3000 Years of Forgotten History at Ramat Raḥel. Jerusalem: Yad Ben Zvi. [Hebrew] Lipschits, O.; Gadot, Y.; and Freud, L. forthcoming  Ramat Raḥel III: Final Publication of Aharoni’s Excavations. In The Emery and Claire Yass Publications in Archaeology. Monograph Series of the Institute of Archaeology of Tel Aviv University. Tel Aviv: Tel Aviv University Press. Lipschits, O.; Gadot, Y.; and Langgut, D. 2012 The Riddle of Ramat Raḥel: The Archaeology of a Royal Persian Period Edifice. Transeuphraténe 41: 57–79.

Rural Economy and Settlement History

263

Lipschits, O., and Naʾaman, N. 2011 From “Baal Perazim” to “Beth-Haccherem”: On the Ancient Name of Ramat-Raḥel. Beth-Miqra 56: 65–86. [Hebrew] Lipschits, O.; Sergi, O.; and Koch, I. 2010 Royal Judahite Jar Handles: Reconsidering the Chronology of the lmlk Stamp Impressions. TA 37: 3–32. Lipschits, O.; Shalom, N.; Shatil, N.; and Gadot, Y. 2014 Judah in the “Long Third Century”: An Archaeological Perspective. Pp. 134–52 in New Studies in the Archaeology of Jerusalem and Its Region, Collected Papers, vol. 8, ed. G. D. Stiebel, O. Peleg-Barkat, D. Ben-Ami, and Y. Gadot. Jerusalem: Israel Antiquities Authority. [Hebrew] Lipschits, O., and Vanderhooft, D. 2011 Yehud Stamp Impressions: A Corpus of Inscribed Stamp Impressions from the Persian and Hellenistic Periods in Judah. Winona Lake, IN: Eisenbrauns. Magen, Y., and Dadon, M. 2003 Nebi Samwil (Montjoie). Pp.  123–38 in One Land, Many Cultures: Archaeological Studies in Honour of Stanislao Loffreda, ed. G. C. Bottini, L. di Segni, and L. D. Chrupcala. Jerusalem: Franciscan Printing Press. Magen, Y., and Finkelstein, I. 1993 Archaeological Survey in the Hill Country of Benjamin. Jerusalem: Israel Antiquities Authority. [Hebrew] Milevski, I. 1996–97  Settlement Patterns in Northern Judah during the Achaemenid Period, According to the Hill Country of Benjamin and Jerusalem Surveys. Bulletin of the Anglo-Israel Archaeological Society 15: 7–29. Naʾaman, N. 1991 The Kingdom of Judah under Josiah. TA 18: 3–71. 1995 The Historical Background of the Philistine Attack on Ahaz in 2 Chronicles 28:18. Pp. 11–26 in Dor le-Dor: Studies in Honor of Joshua Efron, ed. A. Kasher and A. Oppenheimer. Jerusalem: Mosad Bialik. [Hebrew] 2007a Josiah and the Kingdom of Judah. Pp. 189–247 in Good Kings and Bad Kings: The Kingdom of Judah in the Seventh Century bc, ed. L. L. Grabbe. London: T. & T. Clark. 2007b When and How Did Jerusalem Become a Great City? The Rise of Jerusalem as Judah’s Premier City in the Eighth–Seventh Centuries bce. BASOR 347: 21–56. 2009 The Sanctuary of the Gibeonites Revisited. JANER 9: 101–24. Ofer, A. 1993 The Highland of Judah during the Biblical Period. Ph.D. dissertation. Tel Aviv University. [Hebrew] 1998 The Judean Hill in the Biblical Period. Qadmoniot 115: 40–52. [Hebrew] Pritchard, J. B. 1961 The Water System of Gibeon. Philadelphia: University Museum, University of Pennsylvania.

264 1964

Oded Lipschits

Gibeon: Where the Sun Stood Still. Princeton, NJ: Princeton University Press. Schwartz, J. 1988 The History of Lod during the Persian Period. Cathedra 49: 3–12. [Hebrew] Sellers, O. R. 1933 The Citadel of Beth-Zur. Philadelphia: Westminster. 1968 The 1957 Excavation at Beth-Zur. AASOR 38. Cambridge, MA: American Schools of Oriental Research. Sellers, O. R., and Albright, W. F. 1931 The First Campaign of Excavation of Beth Zur. BASOR 43: 2–13. Stern, E., ed. 1982 Material Culture of the Land of the Bible in the Persian Period, 538–332 b.c. Warminster: Aris & Phillips. 2001 Archaeology of the Land of the Bible, vol. 2: The Assyrian, Babylonian, and Persian Periods, 732–332 bce. New York: Doubleday. Soden, W. von 1965 Akkadisches Handwörterbuch, vol. 2: M–S. Wiesbaden: Harrassowitz. Ussishkin, D. 2004 A Synopsis of the Stratigraphical, Chronological and Historical Issues. Pp. 50–119 in The Renewed Archaeological Excavations at Lachish (1973–1994). Sonia and Marco Nadler Institute of Archaeology Monograph Series 22/1. Tel Aviv: Tel Aviv University Press. Weinfeld, M. 2000 Pelekh in Nehemiah 3. Pp.  249–50 in Studies in Historical Geography and Biblical Historiography Presented to Zecharia Kallai, ed. G. Galil and M. Weinfeld. VTSup 81. Leiden: Brill. Williamson, H. G. M. 1985 Ezra, Nehemiah. WBC 16. Waco, TX: Word. Zorn, J. R. 1993 Tell en-Naṣbeh: A Re-evaluation of the Architecture and Stratigraphy of the Early Bronze Age, Iron Age and Latter Periods. Ph.D. dissertation. University of California, Berkeley. 1997 Mizpah: Newly Discovered Stratum Reveals Judah’s Other Capital. BAR 23: 29–38, 66. Zorn, J. R.; Yellin, J.; and Hayes, J. 1994 The M(W)ṢH Stamp Impressions and the Neo-Babylonian Period. IEJ 44: 161–83.

Index of Authors Abbott, D.  12 Abel, F. M.  242 Abu-Rabia, A.  218 Agnes, L. W.  217 Agut-Labordère, D.  184 Aharoni, Y.  179, 213, 221, 222, 223, 240, 241 Aḥituv, S.  180, 181, 192 Ahmed, A. M.  218 Aimé-Giron, M. N.  183 Albertz, R.  154, 155 Albright, W. F.  222, 252 Alt, A.  213, 222, 244 Altmann, P.  8, 18, 109 Amin, S.  31 Amit, D.  253 Aristotle  123, 133, 134, 135 Arrian 170 Arubas, B.  239, 240, 243, 257, 258 Assante, J.  128 Assis, E.  81 Athenaeus 170 Atkinson, K. M. T.  178 Atkinson, K. T.  178 Auld, A. G.  83 Avigad, N.  213 Avi-Yonah, M.  241, 242, 243 Axelrod, R.  5 Ayalon, E.  221 Balcer, J. M.  139 Bang, P. F.  34, 42, 78, 107, 108 Barag, D.  214 Barkay, G.  244 Bartlett, J. R.  222 Batten, L. W.  113 Bauer, H.  181, 193 Bawden, G.  214 Beck, P.  213 Bedford, P. R.  106, 109 Beit-Arieh, I.  213, 214, 221, 222, 223, 226 Ben-Arieh, S.  213 Ben-David, J.  212 Ben ʿElī, Y.  27, 30 Ben-Shlomo, D.  215 Ben Zvi, E.  vii, viii, 17, 39, 41, 44, 62, 67, 83, 126, 223 Berman, M.  218 Bernick-Greenberg, H.  213 Betlyon, J. W.  89, 94

Bienkowski, P.  211, 216 Biran, A.  221, 222 Blanton, R. E.  12, 13, 16 Blenkinsopp, J.  110, 114, 154, 155, 222, 223, 240, 241, 242, 257 Block, D. I.  60 Bocher, E.  239 Boda, M. J.  70 Boer, R.  5, 13, 18, 78, 79, 80, 81, 86, 89, 91, 92, 95, 97, 98 Bongenaar, A. C. V. M.  144 Borowski, O.  62 Bowman, R. A.  240 Braswell, G. E.  6, 7, 8, 16 Briant, P.  55, 56, 65, 136, 138, 143, 152, 167, 169, 172, 178, 182, 184, 185, 186, 187, 189, 191, 194, 195, 196, 197 Brosius, M.  168, 171 Bruno, J.  55 Burckhardt, J. L.  211 Camp, C. V.  27, 30 Cardasçia, G.  152, 154, 156, 160 Carter, C. E.  79, 88, 125, 237, 240, 241, 244, 245 Cartledge, P.  4 Chaney, M. L.  79 Cinamon, G.  249 Clay, A. T.  193 Clines, D. J.  154, 155 Coale, A.  133 Cogan, M.  63, 219, 240 Cohen, R.  55, 222 Cohen-Weinberger, A.  213 Contenau, G.  189 Cook, J.  29 Corner, S.  36 Cowley, A. E.  185, 187 Crook, M. B.  28 Curtius, Quintus  170 Dadon, M.  246 Dalby, A.  55 Dandamaev, M. A.  151, 152, 178, 185, 186, 187, 188, 189, 192, 193 Davies, J. K.  16 Davis, E. F.  4 Davis, J.  9, 11 Dayan, T.  211, 212 Dearman, A.  15

265

266

Index of Authors

De Groot, A.  246 Demsky, A.  240, 241 Dillard, R. B.  52, 57 Diodorus Siculus  65, 170, 175 Dirksen, P. B.  84 Donbaz, V.  176 Dorsey, D. A.  66 Driel, G. van  152, 156 Driver, G. R.  188 Drori, I.  212 Dušek, J.  176 Dyck, J. E.  57

Graf, D. F.  139, 217 Grahame, M.  217, 220 Granovetter, M.  9, 12, 16 Graves, M. W.  220 Greenberg, R.  249 Gropp, D. M.  176 Guillaume, P.  6, 9, 19, 105, 125, 131, 141, 153, 154, 155, 159, 161, 176 Guil, S.  62 Gunneweg, A. H. J.  112, 117, 154, 155, 159 Gunneweg, J.  213

Edelman, D. V.  8, 10, 19, 53, 104, 113, 124, 126, 177, 195 Edens, C.  214 Edwards, A. T.  132 Eisenstadt, S. N.  12 Elayi, J.  8, 53 Elkayam, M.  218 Emberling, G.  220 Ephʿal, I.  8, 176, 177, 179, 186, 188, 192, 194, 195, 214, 215 Eskenazi, T. C.  13, 116 Evans, D. J.  4

Hallo, W. W.  128 Hallock, R. T.  176, 177, 191, 197 Halstead, P.  14 Hausmann, J.  28 Hayes, J.  247 Heltzer, M.  242 Henkelman, W. F. M.  137, 168, 170, 176, 184, 197, 198 Herodotus  10, 56, 65, 125, 138, 139, 167, 171 Herzog, Z.  210, 214, 221, 223 Hesiod  129, 130, 131, 132, 133, 134, 135, 167 Hesse, B.  212 Hilprecht, H. V.  193 Hinz, W.  177, 178 Hirth, K. G.  15 Hodder, I.  220 Hodkinson, S.  36 Hoftijzer, J.  156 Hoglund, K. G.  7, 13, 88, 153, 237 Hopkins, K.  103 Horden, P.  34 Horsley, R. A.  77, 79, 92 Houston, W. J.  107 Hunt, R. C.  42 Hurowitz, V. A.  27, 28, 35

Fales, F. M.  216 Falk, M.  30 Fantalkin, A.  223 Fargher, L. E.  12, 13, 16 Faust, A.  7, 112, 124, 125, 152, 210, 223, 237, 244, 249 Fick, G. W.  4 Finkelstein, I.  124, 209, 210, 212, 214, 246, 247, 251 Finley, M. I.  4, 5, 6, 12, 15, 103 Fischer, I.  27, 28 Fox, M. V.  27, 28, 29, 30, 32, 46 Frankel, R.  64 Freud, L.  213, 240 Fried, L. S.  10, 11, 19, 94, 114 Frood, E.  33 Funk, R. W.  252 Fusfeld, D. B.  106 Gadot, Y.  237, 238, 239, 240, 243, 244, 246, 248, 249, 254, 256, 257, 258 Gallaga, A.  12 Gal-Peer, I.  218 Garraty, C. P.  3, 4, 9, 10 Geus, C. H. J. de  175 Gilbert, M.  30 Godley, A. D.  138 Goh, S. T. S.  30 Gottwald, N. K.  79 Grabbe, L. L.  88, 95

Ikeda, Y.  65 Ikeguchi, M.  78 Ilan, D.  132 Isaac, B.  217 Iserlis, M.  213 Issar, A. S.  134 Jäger, S.  134 Japhet, S.  57, 66, 70, 84 Jasmin, M.  212, 214 Jigoulov, V. S.  53 Jobling, D.  79 Jongeling, K.  156 Jonker, L. C.  18, 54 Joyce, P. M.  60

Index of Authors Jursa, M.  105, 108, 109, 110, 117, 126, 144, 152, 162, 185 Justinian 170 Kallai, Z.  241, 242, 243 Kalluveettil, P.  52 Kartveit, M.  81, 85, 86 Katz, H.  210 Kensky, T. F.  36 Kent, R. G.  65 Kessler, J.  105, 111 Kilmer, A. D.  129 Kleber, K.  144 Klein, R. W.  53, 54, 57, 61, 64, 66, 70 Klein, S.  241, 242 Kletter, R.  222 Klinkott, H.  56 Kloner, A.  237 Knauf, E. A.  176 Knoppers, G. N.  vii, viii, 17, 18, 52, 53, 54, 61, 63, 64, 66, 67, 70, 184 Koch, H.  177, 178 Koch, I.  62, 238, 239, 246, 250, 251, 253, 254 Kochavi, M.  222, 252 Kochman, M.  242 Kogan-Zehavi, E.  215 Kolska Horwits, L.  212 Kraeling, E. G.  183, 187 Kuhrt, A.  61, 167, 170 Kuntzmann, R.  70 Lambert, S.  196 Lane, A.  165 Lang, B.  27, 29, 36, 43 Lang, G.  15 Langgut, D.  258 Lapp, N. L.  246 Lawrence, B.  37, 46 Lemaire, A.  52, 175, 176, 177, 186, 187, 188, 192, 194, 222, 257 Lernau, H.  114 Lernau, O.  114 Leuchter, M.  124 Levi, G.  44 Levin, Y.  82, 84, 87, 96, 97 Lewis, D. M.  56 Lewis, N. N.  218 Lichtenstein, M. H.  27 Lincoln, B.  39, 55 Liphschitz, N.  125, 209, 211 Lipschits, O.  7, 8, 14, 20, 52, 62, 89, 90, 93, 125, 126, 180, 181, 222, 237, 238, 239, 240, 242, 243, 244, 245, 246, 247, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259

267

Liverani, M.  52, 215 Llewellyn-Jones, L.  165, 167, 169 Lowry, S. T.  123 Lozachmeur, H.  186 Luke, K.  27, 28 Lukonin, V.  185, 186, 187, 188, 189, 192, 193 Luquet, G. H.  129 Lyons, E. L.  29, 46 MacDonald, N.  114 Machinist, P.  219 Madreiter, I.  168 Magen, Y.  246, 247 Malthus, T. R.  127, 129 Mankowski, P.  10 Mann, M.  11 Marcinkowska, U. M.  38 Marcos, F. S.  44 Marx, E.  211 Masters, B.  218 Mathew, S.  92 Mauss, M.  105 Mayerson, P.  217 Mazar, A.  221 Mazar, B.  222 McCreesh, T. P.  30 McKechnie, P.  134 Meikle, S.  4, 5, 15 Meissner, B.  181, 193 Métral, F.  141 Meyers, C.  35 Milevski, I.  237, 246 Miller, M. C.  55 Miller, M. L.  vii, viii Miller, W.  9 Milner, C. M.  220 Mitchell, L.  56 Mittmann, S.  180 Mommsen, H.  213 Morris, I.  105, 107, 123, 133 Most, G. W.  130 Motro, H.  211, 212 Myers, J. M.  113, 154, 222 Naʾaman, N.  55, 133, 209, 210, 213, 214, 215, 216, 219, 222, 223, 240, 241, 242, 243, 244, 245, 246, 254 Nam, R. S.  51, 60, 63, 64, 105 Narotzky, S.  106 Naveh, J.  156, 158, 159, 176, 177, 178, 179, 180, 183, 186, 187, 188, 189, 190, 192, 194, 195, 213 Neale, W. C.  133 Negev, A.  212 Nelson, S.  130

268

Index of Authors

Netzer, E.  221 Niehr, H.  187 Noonan, B. J.  113 North, D. C.  103, 104, 105, 106, 107, 108, 110, 114 Noth, M.  66, 222 Notley, S.  54, 60 Novick, T.  27, 37 Nussbaum, G. B.  56 Nwaoru, E. O.  27, 28 Oded, B.  143 Oeming, M.  83, 85, 86, 92, 239, 240, 243, 257, 258 Ofer, A.  241, 244, 245, 250, 251, 252 Ohrenstein, R. A.  136 Ornan, T.  214, 219 O’Shea, J. M.  14, 220 Otzen, B.  214 Parker, B. J.  215 Parkinson, R. B.  134 Parpola, S.  54, 214, 215, 219 Peersmann, J.  215 Peirce, L. P.  168 Piontek, J.  135 Plato 133 Plutarch  170, 171 Pohlmann, K.-F.  60 Polanyi, K.  3, 4, 11, 105, 106, 107, 123 Pomeroy, S. B.  27 Porten, B.  175, 176, 177, 178, 183, 184, 187, 190, 191, 194, 195 Porter, M.  54 Posener, G.  65 Postgate, J. N.  189, 190, 195, 216, 219 Pratico, G. D.  213 Pritchard, J. B.  246 Procopius  216, 217 Purcell, N.  34 Pury, A. de  55 Rainey, A.  54, 60, 214 Ralbag (Rabbi Levi ben Gershon)  30 Redford, D. B.  55, 184 Redmount, C. A.  65 Reich, R.  126 Reinmuth, T.  112, 113, 154, 155 Reisner, G. A.  179 Renz, J.  180 Ries, G.  152, 160 Riley, W.  70 Rofé, A.  27, 28 Römer, T. C.  55 Root, M. C.  55 Rop, J.  56 Rosenfeld, A.  129

Rosenthal, F.  156 Rostovtzeff, M.  4 Roth, M. T.  155 Rowton, M.  224 Roy, J.  56 Rubin, R.  209, 210 Rudolph, W.  66, 154, 155 Ruzicka, S.  125, 138, 139, 140 Saadia Gaon  30 Sahlins, M.  9 Saller, R.  107 Sancisi-Weerdenburg, H.  55 Sanders, J. T.  134 Sapin, J.  175 Sapir-Hen, L.  212 Sass, B.  213 Sasson, A.  211 Sasson, I.  27, 30 Schaper, J.  95, 109 Scheidel, W.  107, 110, 133, 166, 167, 172 Schellenberg, A.  128 Schenker, A.  142 Schnegg, C.  168 Schunck, K.-D.  111 Schwartz, J.  242 Schweitzer, S. J.  58, 70 Scott, J. C.  11 Segal, J. B.  178, 180, 181, 186 Seidl, E.  193 Sellers, O. R.  252 Sergi, O.  239, 246, 250, 251, 253, 254 Shack, W. A.  220 Shahîd, I.  217 Shaked, S.  156, 158, 159, 176, 178, 187, 188, 189, 190, 191 Shalom, N.  237, 254, 255, 256 Shatil, N.  237, 254, 256 Shaus, A.  62 Shereshevski, J.  212 Silver, M.  105 Singer-Avitz, L.  124, 209, 210, 211, 213, 214 Sivan, D.  27 Smith, A.  4, 9 Smith, C. A.  6, 7 Smith, D. A.  12 Smith, R. L.  8, 11 Snyman, G. F.  87 Soden, W. von  240 Sparks, J. T.  83, 86 Stanish, C.  5 Steen, E. van der  211 Steinkeller, P.  33 Stern, E.  8, 213, 219, 221, 222, 241, 242, 252 Stern, I.  237

Index of Authors Stolper, M. W.  54, 125, 126, 152, 160, 176 Stone, E. C.  33 Stone, O.  165, 166, 170 Strabo 171 Szijártó, I.  44 Szlos, M. B.  27 Tadmor, H.  208, 214, 215 Tal, O.  89 Talshir, Z.  27 Tandy, D. W.  133 Tcherov, E.  212 Thareani, Y.  207, 210, 211, 213, 214, 220, 221, 222 Thareani-Sussely, Y.  208, 209, 210, 212, 214, 219, 223, 224 Throntveit, M. A.  58 Tidwell, N. L.  66 Tiňo, J.  58, 70 Trundle, M.  56 Ucko, P.  129 Ussishkin, D.  254 Valler, S.  27, 30 Van Beek, G. W.  215 Vančata, V.  135 Van de Mieerop, M.  105 Vanderhooft, D. S.  52, 89, 239, 250, 254 Vandorpe, K.  33 Van Leeuwen, R. C.  39 Veenhof, K. R.  42 Velde, C. W. M. van de  212 Vleeming, S. P.  182 Vogt, H. C. M.  114 Vos, R. J.  4 Waegeman, M.  27, 29, 43 Wagner-Tsukamoto, S.  4 Waisel, Y.  209, 211 Walser, G.  139 Walton, J. H.  127 Wapnish, P.  212 Watanabe, K.  215 Weber, M.  110 Weinfeld, M.  240

269

Weiss, E.  112, 210 Weiss, R.  66 Welles, C. B.  196 Welten, P.  65, 222 West, G. O.  79, 92 Westbrook, R.  36, 46, 52, 55 White, L., Jr.  128 White, R.  129 Wiesehöfer, J.  19, 55, 169, 172 Wiessner, P.  220 Willi, T.  81 Williams, R. T.  134 Williamson, H. G. M.  57, 65, 112, 154, 155, 240 Wilson, I. D.  34 Winberger-Stern, M.  246 Wittenberg, G.  127 Wolters, A.  27, 29, 37 Wood, B. G.  124 Wright, J. L.  111, 112, 113, 115, 117, 154, 155 Wright, J. R.  30 Wright, K. P.  4 Wunsch, C.  34, 42, 143, 162 Würthwein, E.  63 Xenophon  9, 27, 29, 43, 47, 56, 136, 151 Yair, A.  210 Yamauchi, E.  10 Yardeni, A.  175, 176, 177, 178, 183, 184, 190, 191, 194, 195 Yellin, J.  247 Yilibuw, D.  28 Yoder, C. R.  27, 29, 35, 43 Yona, S.  27 Young, T. C., Jr.  55 Zaccagnini, C.  55 Zahor, M.  134 Zevit, Z.  124 Ziche, H. G.  78 Ziffer, I.  213 Zimhoni, O.  221 Zorn, J. R.  246, 247

Index of Scripture Genesis 1 127 1:22 127 1:28  127, 128 6:1–8 127 9 132 9:1 128 9:2 127 9:7 128 17:2 131 17:20 131 18:6 179 22:17 131 23:9 6 26:4 131 26:24 131 27:5 112 27:33 112 28:3 131 35:11 131

Deuteronomy (cont.) 12:10–12 57 13:17 131 17:14–20 57

Exodus 3:9 154 13:13 137 20 112 23:11 142 23:23 132 23:32–33 132 32:13 131 34:20 137

Judges 3:19 65 4:11 65 14:5 129

Leviticus 17:13 112 25 36 25:3–7 140 25:4  141, 142 25:5 142 25:6–7 142 25:21–22 142 26 131 26:9–10 131 27 109

2 Samuel 5:11–12 52 18:8–10 125 20:15 66 24:18 7 24:18–24 7

Numbers 35:2 83 Deuteronomy 5 112 6:3 131 7:13 131

Joshua 1:8 34 9:5 112 9:14 112 9:17 243 15:21–32 213 15:60 243 18:26 243 18:28 243 21  82, 83, 84, 91, 92 21:1–3 82 21:5–40 82 21:8  84, 92 21:9 84 21:10–19  83, 84

1 Samuel 8:11–17 161 17:36 129 25:18 179

1 Kings 1–2 58 2:1–10 58 2:11–12 58 2:13–46 58 3:1–3 58 3:4–15 58 3:12–13 69 3:16–28 58 4:1–19 58 4:20 58 5:1  53, 57, 58

270

1 Kings (cont.) 5:2 61 5:2–3 58 5:4  53, 57, 58 5:5 58 5:6 58 5:7–8 58 5:9–14 58 5:15  59, 60 5:15–23 60 5:16  59, 60 5:16–20 59 5:17–19 59 5:20  59, 60 5:21  59, 62 5:22–25 59 5:25  59, 60, 61, 62 5:26  52, 59 5:27 59 5:28 59 5:29–30 59 5:31–32 59 7:13 59 7:13–14 60 7:26 61 7:38 61 9:10 63 9:11  62, 63 9:11–13 63 9:11–14 51 9:12–13 63 9:13  52, 64 9:14  63, 64 9:16 68 9:26 65 9:26–28 66 9:27 65 9:28  64, 65 10:1–10 66 10:10 64 10:11  64, 66, 67 10:11–12 66 10:12  66, 67 10:13 66 10:15 66 10:21 68 10:22 68 10:23 69 10:24–25 69 10:26 58

Index of Scripture 1 Kings (cont.) 10:27 58 10:28–29 58 18:32 179 20:32–33 52 21:3 35 2 Kings 2:24 129 6:25 179 7:1 179 7:16 179 7:18 179 9:27 65 19:32 66 23 246 25 124 Isaiah 5:10 61 9:17 125 10:17 125 10:28–32 243 13:12 64 17:6 192 23 34 34:12 114 37:33 66 Jeremiah 6:6 66 13:18–19 222 17 115 17:20 115 17:21 115 17:21–22 115 17:26  115, 244 21 115 23:3 131 27:20 114 29:6 131 30:20 131 32:7–15 243 32:44 244 33:13 244 39:6 114 40 124 40:10–12 135 51:23 114 51:28 114 51:57 114 Ezekiel 4:2 66 17:4 34 17:17 66

Ezekiel (cont.) 21:27 66 23:6 114 23:12 114 23:23 114 26–28 113 26:8 66 27  34, 113 27:2 113 27:14 113 27:15 113 27:17  113, 117 27:20 113 27:24 113 35:1–36 222 36:10 131 36:11 131 37:26 131 45:11  61, 181 45:14  179, 181 Amos 1:9 52 8:5  112, 116 Obadiah 11–14 222 Nahum 3:16 34 Psalms 45:10 64 107:38 131 128 134 128:1–3 130 132:15 112 137 222 138 131 Job 7:2 190 22:24 64 38:16 64 38:41 112 42:13 131 Proverbs 1–9  29, 35, 39 3:15 30 5:15–19 36 8:11 30 8:22–36 30 30:10–31  29, 33 31  17, 30 31:1–9  27, 35

271 Proverbs (cont.) 31:10  30, 36 31:10–31  17, 27, 28, 29, 30, 35, 39, 40, 43, 47 31:11  31, 36 31:13 34 31:15  33, 34 31:16  33, 34, 43 31:17  37, 47 31:18 34 31:19 34 31:20 31 31:21 33 31:22 33 31:23 31 31:25 33 31:25–26 30 31:26 34 31:28 43 31:29  37, 38 31:30 39 31:31  32, 37 Ruth 3:11 30 Song of Songs 3:10 66 Daniel 11:15 66 Ezra 1–8 104 2:25 243 3 113 3:7  53, 116 4:3 156 4:13  10, 155, 188 4:20  10, 188 7:7 141 7:10 34 7:24  10, 155, 156, 188 9:1–2 259 9:2 114 9:12–15 259 10:2–3 259 10:10–15 259 Nehemiah 2:7 140 2:16 114 3  240, 244 3:2 241 3:5 241 3:7 241

272 Nehemiah (cont.) 3:9 240 3:12  240, 241 3:13 241 3:14 240 3:14–18 240 3:15 240 3:16 241 3:17 241 3:18 241 3:19 241 3:22 241 3:27 241 4:8 114 4:13 114 5  11, 12, 19, 43, 104, 153, 160 5:1 154 5:1–5 153 5:1–13 116 5:2 154 5:3 154 5:4  117, 155, 158, 159, 160, 161, 189 5:5  11, 134, 161 5:6–13 162 5:7  112, 114 5:11 162 6:2 259 6:17 114 6:17–19 114 7:5 114 7:29 243 8:1 104 9:1–2 104 10:1 116 10:31 259 10:32  116, 161 10:32–40 112 10:33–40 104 10:38 105 12:28–29 244 13  116, 117 13:7 114 13:10–13 111 13:11 114 13:15  111, 112, 114, 115, 116, 117 13:15–16 104 13:15–22  8, 18, 103, 111 13:16  8, 53, 111, 112, 113, 114, 116 13:16–17 116 13:17  115, 116 13:17–18  111, 114, 115 13:18  111, 112, 114

Index of Scripture Nehemiah (cont.) 13:19–20 111 13:19–22  115, 116 13:20  113, 116 13:20–21 113 13:22  111, 112, 116 13:24 136 1 Chronicles 1–9  18, 77, 81, 84, 86 1:1–27 81 1:28–2:2 81 2:1–2 82 2:3–4:23  81, 83 2:23 82 3:4–5  81, 90 4:41 82 4:43 82 5:10 82 5:19 82 5:27–41 82 5:27–6:66  81, 82, 84, 97 5:36 81 5:41 81 6  83, 86, 91 6:1–15 82 6:1–81 82 6:16–17 92 6:16–34 82 6:35–38 82 6:39  83, 84 6:39–66  82, 84, 91 6:40 92 6:40–45  83, 84 6:46 84 6:46–48 84 6:47 84 6:48 84 6:49  84, 92 6:50  84, 92 6:51 84 6:51–55 84 6:52 92 6:56–61 84 6:62–66 84 7:28 82 8:1–9:1 81 8:12–13 259 8:28  81, 90 8:32  81, 90 9:1–34  81, 90, 91 9:3  81, 90 9:19 92 9:23 92 9:34 90 9:38  81, 90

1 Chronicles (cont.) 13:1 58 14:1 52 14:1–2 52 17:11 58 17:11–14 54 17:14 70 22  58, 93 22–29 58 22:4 52 22:7–17 59 22:8–9 59 22:9  54, 64 22:17–19 59 23–27  90, 93 23:1 58 28  58, 93 28:2–8 59 28:5  54, 70 28:6  54, 58, 70 28:9–10 59 28:10 54 28:20–21 59 29 93 29:1 54 29:1–5 58 29:4 64 29:6–9 93 29:11 70 29:20 59 29:23 70 29:23–25  58, 59 29:25  57, 70 2 Chronicles 1:1 58 1:2–13 58 1:12–13  57, 69 1:14 58 1:14–17 57 1:15 58 1:16–17 58 1:18  57, 58, 59 2:1  57, 59 2:2–3 59 2:2–9 57 2:2–15 60 2:4–5 59 2:6 59 2:6–8 60 2:7 66 2:7–8 59 2:9  59, 61 2:10 179 2:10–11  59, 62 2:12–13 59

Index of Scripture 2 Chronicles (cont.) 2:13–14  52, 62 2:14–15 59 2:16 59 2:17 59 6:10 58 7:17–18 58 8:1 63 8:2 63 8:3  63, 64 8:17  64, 65

2 Chronicles (cont.) 8:17–18 66 8:18  64, 65 9:1–9 66 9:8 70 9:10  66, 67 9:10–11 66 9:11  66, 67 9:12 66 9:18 66 9:20  67, 68

273 2 Chronicles (cont.) 9:21  67, 68 9:22 69 9:23–24 69 9:25 58 9:26  53, 58, 64 13:8 70 24:4 93 27:5  61, 179 31 93 34:9 94 36:22–23  70, 96

Deuterocanonical Literature Sirach 19:1 190 22:3 131

Sirach (cont.) 25:22 41 37:11 190

Sirach (cont.) 38:24–39:11 134 42:11–12 134 50:10 192

Index of Other Ancient Sources Persian Period Tablets, Ostraca, and Papyri Arad Ostraca 1–3 179 6–8 179 10 179 12–13 179 15–16 179 24–25 179 28 179 1–4 180 8 180 11 180 18 180 25 180 31–34 180 42 180 46 180 60–61 180 76 180 79 180 83 180 2 181 Aramaic Documents from Bactria (Naveh and Shaked 2012) A1  158, 188 A2 190 A4 190 A6  190, 191 A8 159 C1  178, 187 C3 178 C5:5 187 D2:2 187 Beer-sheba Ostraca 1–6 179 13 179 27–30 179 32 179 45 179 47 179 50 179 Elephantine Papyri AP (Aramaic Papyri) 1:3–5 186 7:2 186 16:2 186 11 185 24 183

Elephantine Papyri (cont.) AP (cont.) 45:8 183 62–63 180 65–66 180 70 180 79 180 81 180 TAD (Textbook of Aramaic Documents from Ancient Egypt) A.5.2:2 186 A6.11  157, 158 A6.13  159, 160 B3.3 32 B3.8 32 B3.13 182 B4.2 185 B5.1:3–5 186 B.7.2:2 186 C.3.28 180 C 3.14  184 C 3.18  184 C 3.25–28  184 D7.46:2 183 Idumean Ostraca E/N (Ephʿal and Naveh 1996) 1:3 186 3:2 186 3:3 188 4:2 186 5:3 188 6:2 188 7:2 188 11:2 186 13:3 179 15:2 192 25:2 186 26:2 188 30:1 188 38:2 192 47 192 48:2 188 49:3 192 52:2 188 62:3 179 69:1 186

274

Idumean Ostraca (cont.) E/N (cont.) 72:2 186 79:2 186 80:2 188 85:1? 186 87 191 92 192 95 191 98  194, 95 98:2 186 99:2 179 118 191 120:2 186 124:2 192 133:1 186 137:1 186 150:1 186 154:3 186 168 195 185–95 186 271:3 186 L1 (Lemaire 1996) 21:3 186 42:2 188 58:3 186 123:1 186 128:2 188 148:1 186 L2 (Lemaire 2002) 2:3 186 4:1 186 7:3? 188 9:3 188 44:2 186 70:2 186 110:2 186 176:1 186 198:1 188 212:1 186 213:2 192 225:2 192 230 188 238:3 186 245:3 186 259 187 283  186, 188 285:2 186

Index of Other Ancient Sources Idumean Ostraca (cont.) L2 (cont.) 292:2 186 300:2 188 310:1 186 365:10 186 Porten and Yardeni 2006 (table 3a) 1 (= ISAP 210 = EN 95) 191 2 (= ISAP 109)  191 3 (= ISAP 259)  191 5 (= ISAP 424)  191 14 (= ISAP 1909 = EN 118) 191 17 (= ISAP 1881 = EN 87) 191 18 (= ISAP 1243)  191 20 (= ISAP 460)  191 25 (= ISAP 495)  191 29 (= ISAP 443)  191 31 (= ISAP 209)  191 39 (= ISAP 452)  191

Idumean Ostraca (cont.) PY 2006 table 3a (cont.) 45–51 (= ISAP 1600, 816, 1641, 1660, 1659, 221)  191 52 (= ISAP 1114)  191 ISAP (Aramaic Papyri) 109 191 209–10 191 221 191 259 191 424 191 443 191 452 191 460 191 495 191 816 191 1114 191 1243 191 1600 191 1640–41 191 1659 191 1881 191

275

Idumean Ostraca (cont.) ISAP (Aram. P.) (cont.) 1909 191 900–901 194 Persepolis Fortification Archive 5 168 278 197 352 197 363 197 364 197 588 197 1216 170 1943 191 1948 191 Saqqara Aramaic Papyri 18 186 24 186 24.8 181 31 186 69a 181 Saqqara Ostraca 180

Greco-Roman Sources Inscriptional

IG II2 207:5–6  196 Welles 3:80–85 196

Literary

Aristotle Politics 2.6 133 Arrian of Nicomedia Anabasis 7.4.4–8 170 Athenaeus Deipnosophists 12.538b–530a 170 Curtius Historiae Alexandri Magni 10.3.11–12 170 Diodorus Bibliotheca historica 1.33.9–10 65 17.107.6 170 Herodotus Histories 1.56–57 65 1.136 171 2.158–59 65 2.164 125 2.168 125

Herodotus, Hist. (cont.) 3.63–66 139 3.89 56 3.91 10 4.44 65 6.41 167 7.1 138 Justin Epitoma Historiarum Philippicarum 12.10.9 170 Plato Laws 5.740d 133 6.784 b–e  133 Plutarch Alex. 69.1  171, 172 70.2 170 Procopius History of the Wars 1.19.7–16 216 Strabo Geography 15.3.17 171 Xenophon Anabasis 1.1.5 56 1.5.7 136 1.9.20–28 56

Xenophon (cont.) Cyropaedia 7.5.69 9 8.6.4 151 8.6.5 151 Oeconomicus  27, 29, 41, 43 3.10–15 29 6.9 29 7.1–41 29 8.11–17 29 10.10–13 29 Xenophon (cont.) Hellenica 3.1.27–28 151 Hesiod Works and Days 129, 130, 132 228 130 376 129 380 130 376–81 130 405 167 698–700 130

276

Index of Other Ancient Sources

Rabbinic/Judaic Sources Mishnah m. Moʿed Qaṭ. 1:4 186 m. Peʾah 8:5 192 m. Šeb. 2:1 186

Midrash Proverbs (Buber) 31.10 30 Pesiq. Rab. 31 30

Ralbag Commentary on Proverbs Proverbs 31  30 Zohar 3.42.b 30

Additional Sources British Museum Tablets (BM) 42352+ 185 Cuneiform Texts (CT) 57 22  109

Darius, Suez c Inscription (DZc) §3.7–12 65

Hammurabi Laws of Hammurabi (LH) §§49–52 154