Approaches to Literary Readings of Ancient Jewish Writings Studia Semitica Neerlandica

Editor-in-Chief Prof. dr. K.A.D. Smelik

Editorial Board Prof. dr. P.C. Beentjes, Prof. dr. W.J. van Bekkum, Dr. W.C. Delsman, Prof. dr. H. Gzella, Prof. dr. J. Hoftijzer, Prof. Dr. W. Th. van Peursen, Prof. dr. J. Van Steenbergen, Prof. dr. E. Talstra, Prof. dr. M. Tanret

volume 62

The titles published in this series are listed at brill.com/ssn Approaches to Literary Readings of Ancient Jewish Writings

Edited by Klaas Smelik and Karolien Vermeulen

Leiden • boston 2014 Library of Congress Cataloging-in-Publication Data

Approaches to literary readings of ancient Jewish writings / edited by Klaas Smelik and Karolien Vermeulen. pages cm. — (Studia Semitica Neerlandica ; volume 62) “The essays in this volume were presented at the “Ancient Jewish Texts and the ‘Literary’” Conference organised by the institute of Jewish Studies (university of Antwerp) and the research unit Hebrew and Judaic Studies at Ghent university in 2012 (March 14–15)”—Preface. Includes index. ISBN 978-90-04-25819-8 (hardback : alk. paper) — ISBN 978-90-04-25856-3 (e-book) 1. Bible. Old Testament—Criticism, interpretation, etc.—Congresses. 2. Midrash—Criticism, Textual—Congresses. I. Smelik, K. A. D., 1950– editor of compilation. II. Vermeulen, Karolien, editor of compilation. III. Smelik, K. A. D., 1950– editor of compilation. Battle of wits and words: Hushai, Ahithophel, and the Absalom rebellion (2 Samuel 16–17)

BS1171.3.A67 2013 221.6’6—dc23 2013027599

This publication has been typeset in the multilingual “Brill” typeface. With over 5,100 characters covering Latin, IPA, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities. For more information, please see www.brill.com/brill-typeface.

ISSN 0081-6914 ISBN 978-90-04-25819-8 (hardback) ISBN 978-90-04-25856-3 (e-book)

Copyright 2014 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Global Oriental, Hotei Publishing, IDC Publishers and Martinus Nijhoff Publishers.

All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher.

Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change.

This book is printed on acid-free paper. Contents

Preface ...... vii Abbreviations ...... ix

Introduction: Some Thoughts on Ancient Jewish Texts and the ‘Literary’ ...... 1 Karolien Vermeulen

Part One Hebrew Bible

“Literary” Craft and Performative Power in the Ancient Near East: The Hebrew Bible in Context ...... 19 Scott B. Noegel

Selecting and Analyzing Metaphors in the Hebrew Bible: Cognitive Linguistics and the Literary ...... 39 Johan de Joode and Hanneke van Loon

Literary Theory and Composition History of the Torah: The Sea Crossing (Exod 14:1–31) as a Test Case ...... 53 Angela Roskop Erisman

Eyewitness Accounts in the Book of Samuel? A Reappraisal ...... 77 Klaas Smelik

A Battle of Wits and Words: Hushai, Ahithophel and the Absalom Rebellion (2 Samuel 16–17) ...... 99 Robert P. Gordon

דבר סתר The Intentional Use of Polysemy: A Case Study of (Judg 3:19) ...... 115 Karolien Vermeulen vi contents

Bridging the Gap between Linguistics and Literary Studies of Ancient Biblical and Jewish Texts: A Proposal Exemplified by a Study of Gen 9:8–17 ...... 137 Ellen van Wolde

The ‘Literary’ in Trickster Narratives: A Comparison of Gen 27:1–39; 29:15–30; 31:19–55; Joshua 9 and 2 Sam 3:26–27 ...... 167 Anne-Laure Zwilling

Part two Other Ancient Jewish Writings

“Friends Hearken to Your Voice”: Rabbinic Interpretations of the Song of Songs ...... 183 Tamar Kadari

Between Tradition and Innovation: Seder Eliyahu’s Literary Strategies in the Context of Late Midrash ...... 211 Lennart Lehmhaus

Medieval Rationalist Exegesis and Intertextuality in Samuel Ibn Tibbon’s Ma’amar Yiqqawu ha-Mayim ...... 243 Rebecca Kneller-Rowe

Index of Primary Sources ...... 269 Preface

The essays in this volume were presented at the “Ancient Jewish Texts and the ‘Literary’” Conference organised by the Institute of Jewish Studies (University of Antwerp) and the research unit Hebrew and Judaic Studies at Ghent University in 2012 (March 14–15). The conference explored the literary dimension of ancient Jewish texts, gathering scholars from different academic fields (rabbinic studies, biblical studies, literary studies, and philosophy) and drawing on different methodologies (literary criticism, rhetorical criticism, cognitive linguistics, historical criticism, and reception history) in order to establish the state of the arts in the field of ancient Jewish text studies. Participants discussed the relevance, accuracy, potential—and possible alternatives to the literary approach developed in the 1960s and 1970s and predominant since the 1980s in addressing textual features. We are grateful for the generous support of the Institute of Jewish Stud- ies and her director, prof. Vivian Liska, who made this meeting possible. We also like to thank the other members of the Institute who assisted in various ways and in particular Jan Morrens, the administrative coordina- tor of the Institute. Additional thanks go to the University of Antwerp and Ghent University for providing the more general setting for the conference and the consequent book volume as well as to the several speakers at the conference who were willing to rework their contributions for the current volume. We are indebted to Irene van Rossum and Stephanie Paalvast of Brill for accepting our manuscript and for their help in producing the text.

Ghent, 17th May, 2013 Klaas Smelik and Karolien Vermeulen

ABBREVIATIONS

AB Anchor Bible ABRL Anchor Bible Reference Library ACEBT Amsterdamse Cahiers voor Exegese en Bijbelse Theologie ACEBT.SS Amsterdamse Cahiers voor Exegese en Bijbelse Theologie Supplement Series AnBib Analecta Biblica AJSR Association for Jewish Studies Review AMD Ancient Magic and Divination ANEP J.B. Pritchard, Ancient Near Eastern in Pictures relating to the Old Testament ANET J.B. Pritchard, Ancient Near Eastern Texts relating to the Old Testament AOAT Alter Orient und Altes Testament AOS American Oriental Series ARM Archives royales de Mari AS Anatolian Studies AThD Acta Theologica Danica Bibl Biblica BiKi Bibel und Kirche BDB F. Brown, S.R. Driver & C.A. Briggs, A Hebrew and English Lexicon of the Old Testament. Oxford, 1907 BEThL Bibliotheca Ephemeridum Theologicarum Lovaniensis BIS Biblical Interpretation Series BJS Brown Judaic Studies BKAT Biblisches Kommentar zum Alten Testament BM British Museum Collection BLS Bible and Literature Series BSPS Boston Studies in the Philosophy of Science BSS Biblical Seminar Series BZAW Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft CAT Commentaire de l’Ancien Testament CBC Cambridge Bible Commentary CB.OT Coniectanea Biblica. Old Testament Series CBQ Catholic Biblical Quarterly CHAN Culture and History of the Ancient Near East Series x abbreviations

DCH The Dictionary of Classical Hebrew DJD Discoveries in the Judaean Desert EThL Ephemerides Theologicae Lovanienses ESV English Standard Version FAT Forschungen zur Alten Testament FRLANT Forschungen zur Religion und Literatur des Alten und Neuen Testament GKC W. Gesenius, E. Kautzsch & A.E. Cowley, Hebrew Grammar HALOT Koehler, L., W. Baumgartner, and J.J. Stamm, The Hebrew and Aramaic Lexicon of the Old Testament. Translated and edited under the supervision of M.E. Richardson. 5 vols. Leiden, 1994–2000 Herv. Teol. Stud. Hervormde Teologiese Studies HTR Harvard Theological Review ICC International Critical Commentary of the Holy Scriptures IJS IJS Studies in Judaica Interp. Interpretation ISBL Indiana Studies in Biblical Literature ITC International Theological Commentary JANES Journal of Ancient Near Eastern Studies JAOS Journal of the American Oriental Society JBL Journal of Biblical Literature JBQ Jewish Bible Quarterly JCS Journal of Cuneiform Studies JEA Journal of Egyptian Archaeology JECS Journal of Early Christian Studies JHS Journal of Hebrew Scriptures JJS Journal of Jewish Studies JJTP Journal of Jewish Thought and Philosophy JLT Journal of Literature and Theology JNES Journal of Near Eastern Studies JNSL Journal of Northwest Semitic Languages JR Journal of Religion JS Jewish Studies JSHL Jerusalem Studies in Hebrew Literature JSIJ Jewish Studies: An Internet Journal JSJ Journal for the Study of Judaism in the Persian, Hel- lenistic, and Roman Periods abbreviations xi

JSOT Journal for the Study of the Old Testament JSOT.SS Journal for the Study of the Old Testament, Supplement Series JSQ Jewish Studies Quarterly JThS Journal of Theological Studies KJV King James Version KTU M. Dietrich, O. Loretz & J. Sammartin, Die Keilalphabetischen Texte aus Ugarit LHB/OTS Library of Hebrew Bible/Old Testament Studies MMA Monographs in Mediterranean Archaeology MT Masoretic Text NABU Nouvelles assyriologiques brèves et utilitaires NCB New Clarendon Bible NIDOTTE New International Dictionary of Old Testament Theology and Exegesis NIV New International Version NJPS The New Jewish Publication Society of America Tanach NRSV New Revised Standard Version NRT La nouvelle revue théologique OBO orbis Biblicus et Orientalis OIS oriental Institute Seminars Or Orientalia OT Oral Tradition OTL old Testament Library OTS oudtestamentische Studiën P & P Past and Present Proof Prooftexts: A Journal of Jewish Literary History Psychol Rev Psychological Review PT Poetics Today RB Revue Biblique RBJ Review of Rabbinic Judaism RCL Review of Cognitive Linguistics REB Revised English Bible RGRW Religions in the Graeco-Roman World RHPR Revue d’histoire et de philosophie religieuses SAACT State Archives of Assyria Cuneiform Texts SAAS State Archives of Assyria Studies SAK Studien zur altägyptischen Kultur SAOC Studies in Ancient Oriental Civilization ScrHie Scripta Hierosolymitana xii abbreviations

SFEG Schriften der Finnischen Exegetischen Gesellschaft SSN Studia Semitica Neerlandica SBS Stuttgarter Bibelstudien SBT Studies in Biblical Theology Semeia Semeia Studies Tarb. Tarbiz TECC Textos y Estudios “Cardenal Cisneros” de la Biblia Políglota Matritense THAT Theologisches Handwörterbuch zum Alten Testament ThB Theologische Bücherei ThSt Theologische Studien ThWAT Theologisches Wörterbuch zum Alten Testament TSMJ Texts and Studies in Medieval and Early Modern Judaism UF Ugarit-Forschungen VT Vetus Testamentum VT.S Vetus Testamentum. Supplements WBC Word Biblical Commentary ZA Zeitschrift für Assyriologie und vorderasiatische Archäologie ZAW Zeitschrift für die alttestamentliche Wissenschaft Zion Zion Introduction: Some Thoughts on Ancient Jewish Texts and the ‘Literary’

Karolien Vermeulen

Abstract

Since the late 1970s and early 1980s, the literary approach has considerably defined the academic landscape of biblical studies. As an answer to the historical-critical and theological studies of the Hebrew Bible, it offered a fresh look on an old text, focusing on its qualities as a piece of literature. Whereas the Bible text received much attention from literary scholars, the approach did not make its way immediately to studies of other ancient Jewish texts, such as the study of midrash, liturgical texts, or early commentaries. This article presents some observations regarding the relationship between literary analysis on the one hand and ancient Jewish text studies on the other hand.

In 2007, the academic journal Prooftexts issued a special volume entitled “Before and After the Art of Biblical Narrative.”1 In this volume, articles are brought together that analyze the past, present, and future of the literary approach to ancient Jewish texts as it was given shape in biblical studies by Robert Alter in the early 1980s.2 Some of the contributions are rather critical, accusing the method of having a hidden theological agenda3 and

1 “Before and After the Art of Biblical Narrative,” Proof 27,2 (2007). 2 Robert Alter, The Art of Biblical Narrative (New York: Basic Books, 1981). Other works and scholars did do preliminary work on the literary approach in biblical studies; how ever, it is Alter who achieves a major breakthrough for the method’s general acceptance in the field of biblical studies. In a very accessible and novel like way, he makes a literary reading of the Bible available to a broad audience (both academic and non-academic). Preliminary steps can be found in the work of among others, Erich Auerbach (Mimesis: Dargestellte Wirklichkeit in der abendländischen Literatur [Bern: A. Francke 1946]); Martin Buber & Franz Rosenzweig (Die Schrift und ihre Verdeutschung [Köln/Olten: Jakob Hegner, 1954–1962]); Meir Weiss, (The Bible from Within: The Method of Total Interpretation [Jerusa lem: Magness Press, 1962] [in Hebrew]); Roland Barthes (“La lute avec l’ange: Analyse tex tuelle de Genèse 32:23–22,” in Analyse structurale et exégèse biblique: Essais d’interprétation, ed. Roland Barthes et al. [Neuchâtel: Delachaux et Niestlé, 1971]); Jan Fokkelman (Narrative Art in Genesis [Assen: Van Gorcum, 1975]); Shimon Bar-Efrat (Narrative Art in the Bible [Tel Aviv: Sifriyat po’alim, 1979] [in Hebrew]). See also the early issues of the journal Semeia, founded in 1972 by the Society of Biblical Literature. 3 Mara Benjamin, “The Tacit Agenda of a Literary Approach to the Bible,” Proof 27,2 (2007): 254–274. 2 karolien vermeulen a rigid conservatism not in sync with the evolution in the broader field of humanities.4 Others take a more positive stance, demonstrating the validity of the approach by applying it to both biblical texts and later texts inspired by the Bible.5

For Starters: The Literary Approach According to Weitzman

By way of introduction, let us consider the following quotes by Steven Weitzman, who wrote the introduction of the Prooftexts volume. He opens his piece by stating, “however newfangled it seemed in the 1980s, the ‘literary approach’ to the Bible, the attempt to understand it as a work of aesthetic and not just religious or historical value, is as old as most other methods of biblical study.”6 He continues that the declining interest in this way of reading the Bible will probably not come as a surprise to readers familiar with how the practice of literary interpretation has changed in the last twenty to thirty years, an era that saw the crystallization of deconstruction and other subversive reading strategies that sought to emphasize the elusiveness, ruptures, and self-contradictions of literary language or of the self in its relationship to language; the impact of reception theory that shifted the focus from the text and its capacity to shape interpretation to readers and their importation of meaning into the text; the rising dominance of a cultural studies paradigm that expanded interpretation’s purview from certain canonical texts privileged by their aesthetic merits to a broader array of discourse and modes of signification; ideologically committed perspectives for which literary analysis became a vehicle for critique; and other trends conventionally labeled as “theory.”7 He concludes his article with the question, In which ways does the literary approach to the Bible remain an engaging and defensible project? Can it again play the role it did in the hands of an Alter, not just feeding off what scholars in other disciplines are doing with

4 Steven Weitzman, “Before and After the Art of Biblical Narrative,” Proof 27,2 (2007): 191–210. 5 Jonathan Sheehan, “The Poetics and Politics of Theodicy,” Proof 27,2 (2007): 211–232; Ilana Pardes, “Job’s Leviathan: Between Melville and Alter,” Ibidem: 233–253; Menakhem Perry, “Counter-Stories in the Bible: Rebekah and Her Bridegroom, Abraham’s Servant,” Ibidem: 275–323; Robert Kawashima, “Comparative Literature and Biblical Studies: The Case of Allusion,” Ibidem: 324–344; Chaya Halberstam, “The Art of Biblical Law,” Ibidem: 345–364. 6 Weitzman, “Before and After,” 192. 7 Ibidem, 195–196. introduction: thoughts on ancient jewish texts 3

texts or mimicking their interpretive moves, but helping to shape the future of literary study?8

The Literary Approach is ‘Old’ The first issue Weitzman touches upon is that the literary approach is an ‘old’ method. As Alter himself already indicated in his The Art of Biblical Literature, the literary approach to the Bible existed long before literary scholars came across the text. In midrash, ample attention has been paid to the narratives, the characters, the choice of words and formulation.9 Actually, the writers of midrash might be considered literary scholars avant-la-lettre, were it not for their different starting point as well as the different goals they had in mind. Obviously, their analyses occurred in a religious setting and they aimed at clarifying the text and assisting the audience in their interpretation. Their work did not originate from an interest in the stories as being stories or the joy that reading a well- composed work brought about for its reader. But in essence, the method of both the rabbi and the literary scholar is the same: close reading of a text they consider to be a unity and not a collection of historically different snippets.10 Thus, Alter and with him other literary scholars did not do something new in terms of method (Alter never claimed this either). Yet, simultaneously, they did something radically different because they put the Bible’s nature as a story before its status as a religious and historical document. The larger framework in which they were applying their

8 Ibidem, 207. למה נברא העולם בב' ]. . .[ ולמה לא באל"ף :In M. Gen. Rab. I.10, the following can be read 9 “Why has the world been created with the letter bet? [. . .] Why not the letter ’aleph?” To this comment on the language, various answers are formulated of which some interpret רבי יונה בשם ר' לוי אמר למה נברא העולם :the form of the language, rather than its content בב' אלא מה ב' זה סתום מכל צדדיו ופתוח מלפניו כך אין לך רשות לומר מה למטה מה למעלה ,Rabbi Jona in the name of Rabbi Levi said“ ,מה לפנים מה לאחור אלא מיום שנברא העולם Why has the world been created with the letter bet? Because like the bet is closed on all its sides and opens but to the front, likewise you are not required to ask ‘What is there underneath, what is there on top, what is there in front, what is there after’ but [you only ask] from the day that the world was created.” Another nice example can be found in Gen. מטרונה אחת שאלה את רבי יוסי אמרה לו למה בגניבה אמר לה משל אם הפקיד :Rab. 17.7 אדם לידך אונקיא של כסף בחשאי וחזרת ליה ליטרא של זהב בפרהסיא זו גניבה אמרה לו למה A matron“ ,במטמוניות ואמר לה בתחלה בראה לו וראה אותה מלאה רירין ודם הפליגה ממנו asked Rabbi José, ‘Why did God steal a rib from Adam?’ He said to her, ‘An example: If one were to take away from your hand an ounce of silver in silence, and give you in return a pound of gold publicly, that would be stealing.’ She said, ‘Why secretly?’ And R. José said to her, ‘He started creating her in front of him, but he [i.e., Adam] saw her full of secretions and blood and he removed himself from her.’” 10 Alter, The Art of Biblical Narrative, 11–12. 4 karolien vermeulen method of close reading differed substantially from that of the rabbis or the church fathers.11 Moreover, literary analyses of the Hebrew Bible may have existed before biblical literary criticism came along, but these never made it to the academy or a broad audience of both believers and non- believers. And that is exactly what Alter initiated: academic scholarship on the Hebrew Bible as literature and a societal awareness of the text as being also a good novel (in addition to or instead of its religious and historical roles).

The Literary Approach is Old-Fashioned Weitzman describes the declining interest in the literary approach as being not fit for modern times. Despite this rather pessimistic evaluation, he does recognize that Alter’s work was read and continues to be read on a broad scale and that it inspires scholars up until today to conduct valuable research based on a literary perspective.12 Moreover, a look at recent publications in some of the leading journals of the field, among which Vetus Testamentum, Journal of Biblical Literature, Journal for the Study of the Old Testament, Zeitschrift für Alttestamentliche Wissenschaft, Journal of Hebrew Scriptures, Journal for the Study of Judaism in the Persian, Hel- lenistic and Roman Periods, Association of Jewish Studies Review, Journal of Jewish Studies, and even Prooftexts itself, reveals that 30 years after The Art a substantial amount of the research published in these journals (at an average of one out of five) draws upon the literary paradigm in one way or another.13 Taking into account the theoretical and methodological inflation of the last decades with emerging frameworks such as feminism, post-colonialism, queer theory, psychoanalytic readings, eco criticism, cognitive linguistics and cyber readings—the method still stands, maybe

11 Alter, The Art of Biblical Narrative, 3–33. Weitzman, “Before and After,” 191, 195. Yairah Amit, Reading Biblical Narratives: Literary Criticism and the Hebrew Bible (Minneapolis, MN: Augsburg Fortress, 2001), 9–14. As an example, compare the observations on Gen 9:6 by Robert Alter on the one hand and the comments to be found in Midrash Genesis Rabbah on the other hand. Alter discusses the ingenious structure and the exploration of language in order to get the message of the verse across: “As many analysts of the Hebrew have noted, there is an emphatic play on dam, ‘blood,’ and ’adam, ‘human,’ and the chias tic word order of the Hebrew formally mirrors the idea of measure for measure: shofekh [spills] dam [blood] ha’adam [of the human], ba’adam [by the human] damo [his blood] yishafekh [will be spilled] (= A B C C’ B’ A’)” (Robert Alter, Genesis: Translation and Com mentary [London/New York: WW Norton & Company, 1996], 38). In M. Gen. Rab. 34.14, the discussion centers on the different crimes committed as well as on the interpretation of murder as diminishing God’s image. No comments are made on the style of the text. 12 Weitzman, “Before and After,” 207. 13 For an overview, see the appended table. introduction: thoughts on ancient jewish texts 5 precisely because of its conservatism, or as Weitzman says, its “history.”14 Just like historical-critical and historical research, literary criticism became an established value in biblical studies. In addition, while the literary approach for biblical text study claimed its place in the 1980s, other ancient Jewish texts became objects of literary research more recently. In midrashic and talmudic studies the literary paradigm started to replace a historical-critical one in the 1990s and 2000s. Geoffrey Hartman and Sannford Buddick’s Midrash and Literature (1986) and Daniel Boyarin’s Intertextuality and the Reading of Midrash (1991) are two seminal works in this context, in addition to the work of Jonah Fraenkel on haggadic texts and their narrative art. In recent times, scholars such as Charlotte Elisheva Fonrobert, Galit Hasan-Rokem, Jeffrey Rubenstein and Steven Fraade among others gave rabbinic studies a more literary outlook.15 Their work is sometimes interdisciplinary, sometimes not, but in all cases “leaning literary, reading rabbinics.”16 If we would look beyond the scope of ancient times, we can see that the method just found its way to the research of Kabbalistic texts.17 To counter Weitzman’s prophecy of a silent death of literary criticism, it can be argued that the study of the Bible as literature is just the tip of the iceberg, opening up to the literary study of an array of Jewish texts composed in conjunction with the biblical text.

14 Weitzman, “Before and After,” 202. 15 Geoffrey Hartman & Sannford Buddick, Midrash and Literature (New Haven, CT: Yale University Press, 1986); Daniel Boyarin, Intertextuality and the Reading of Midrash (Bloomington, IN: Indiana University Press, 1991). Jonah Fraenkel’s articles have been col lected in the volume Sipur ha-agadah, ahdut shel tokhen ve-tsurah: Kovets mehkarim (Tel Aviv: Hakibbutz Hameuchad, 2001). For recent contributions, see among others, Charlotte Elisheva Fonrobert & Martin Jaffee, eds., The Cambridge Companion to the Talmud and Rabbinic Literature (Cambridge/New York: Cambridge University Press, 2007); Galit Hasan- Rokem, Tales of the Neighborhood: Jewish Narrative Dialogues in Late Antiquity (Berkeley, CA: University of California Press, 2003); Joshua Levinson, Jacob Elbaum & Galit Hasan- Rokem, eds., Higayon L’Yona: New Aspects in the Study of Midrash, Aggadah and Piyut in Honor of Professor Yona Fraenkel (Jerusalem: Hebrew University Magnes Press, 2006); Jeffrey Rubenstein, Talmudic Stories: Narrative Art, Composition, and Culture (Baltimore: Johns Hopkins University Press, 1999); Steven Fraade, Legal Fictions: Studies of Law and Narrative in the Discursive Worlds of Ancient Jewish Sectarians and Sages (Leiden/Boston: Brill, 2011), Moshe Simon-Shoshan, Stories of the Law: Narrative Discourse and the Con struction of Authority in the Mishnah (Oxford/New York: Oxford University Press, 2012). See also Joshua Levinson “Literary Approaches to Midrash,” in Current Trends in the Study of Midrash (ed. Carol Bakhos; Leiden/Boston: Brill, 2006), 189–226. 16 Burton Visotsky, “Leaning Literary, Reading Rabbinics,” Proof 28 (2008): 85–99 (86). 17 Shaul Magid, “Lurianic Kabbalah and Its Literary Form—Myth, Fiction, History,” Proof 29,3 (2009): 362–397; Eitan Fishbane, “The Scent of the Rose: Drama, Fiction, and Narrative Form in the Zohar,” ibidem: 324–361; Don Seeman & Shaul Magid, “Mystical Poetics: The Jewish Mystical Text as Literature,” ibidem (2009): 317–323. 6 karolien vermeulen

The Future of the Literary Approach Weitzman also wonders about the future of biblical literary criticism. Is it still able to add something to the scholarly discussion or has it brought forth all there is to offer? Can it continue as it stands or should it reinvent itself to stay relevant and valuable for current and future generations of scholars? The previously discussed issues partly answer those questions. The method embedded in literary criticism is as old as the Hebrew Bible text itself and it has been used ever since. Given the research it has generated, it has proven its validity as well as its flexibility and multivalency in order to be of use within different academic frameworks. Furthermore, ancient Jewish texts cover more than the Hebrew Bible, and many of these other texts benefit from a literary analysis that sheds new light on them and their internal functioning. Recent publications show that literary criticism can reinvent itself and that it now takes up considerable space within the broader academic field of ancient Jewish text study. It is true that there are many rivals in the field, but that may be rather an advantage than a disadvantage because it will force the literary scholar to question his own framework and consider its suitability in light of other approaches. Therefore, the method’s challenges do not have to do with its relevance but with issues that have existed from the beginning: the compatibility of the modern and the ancient, the dichotomy between the literary and the historical, and the position towards and the interaction with other methodological frameworks.

The Challenges Faced by the Literary Approach

The Ancient and the Modern One of the main points of critique towards the literary approach is its anachronism. How can one analyze ancient texts by means of a concept that did not exist at that time? It is a problem that all early texts share, yet it has not kept us from calling Gilgamesh a wonderful story and Homer the author of a piece of world literature. However, ancient Jewish texts incorporate a fair amount of religious texts that are important for world religions even now. Nobody will make accusations of blasphemy when someone describes Hesiod’s Theogonia as great literature even though it relates a story of gods and creation. It is a different story when the same happens with Genesis 1. Gradual secularization has taken off some of the introduction: thoughts on ancient jewish texts 7 pressure, and the work of Alter and his colleagues has definitely helped. Moreover, as Burke Long notes, Alter’s reading is not really literary, i.e., “engaging in a resistive criticism;”18 it is an inspired reading that “avoids and delays the questions that must be asked and dealt with forthrightly.”19 According to Mara Benjamin, Alter’s approach has a hidden agenda that makes it in essence disguised theology.20 The study of non-biblical but ancient Jewish texts revives the same difficulties. Moshe Simon-Shoshan attempts to solve this issue by interpreting literariness as “a socially constructed category that reflects a text’s place within a given culture rather than the essential qualities of the text.”21 Also Joshua Levinson addresses the issue in his contribution in Current Trends in the Study of Midrash, seeing the literary as an “open category,” a reference to a discipline of which “the contours [. . .] are constantly changing.”22 With these corrections, the literary label is conceived as less problematic or less in conflict with the other labels the texts have carried through tradition. The literary approach thus will have to find a way to address the questions that arise from the gap between the text’s original context and that of the present day scholar using his modern methods of analysis.

The Literary and the Historical Another element that needs attention is the relationship as well as distinction between the literary and the historical. Exoneration of the Bible was one of the main grounds for approaching it through literary analysis. In a post-war era when far fewer people were interested and believed in the religious message of the biblical text, literary analysis retrieved its authority by giving it status as a piece of world literature.23 In entering the

18 Burke Long, “The ‘New’ Biblical Poetics of Alter and Sternberg,” JSOT 51 (1991): 71–84 (84). 19 Long, “The ‘New’ Biblical Poetics,” 83. 20 Benjamin, “The Tacit Agenda,” 254–274. 21 Simon-Shoshan, Stories of the Law, 7. 22 Levinson, “Literary Approaches to Midrash,” 190. 23 I have mentioned the work of Robert Alter before (note 2), but various other works can be mentioned that have contributed to the Bible’s status as literature. The follow ing list is indicative rather than exhaustive with emphasis on works that establish the status rather than those that maintain it afterwards: Northrop Freye, The Great Code: The Bible and Literature (New York: Hartcourt Brace Jovanovich, 1982); Adele Berlin, Poetics and Interpretation of Biblical Narrative (Sheffield: Almond Press, 1983); Meir Sternberg, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading (Bloomington, 8 karolien vermeulen realm of the literary, a realm associated with fiction and with a disinter- est in history, the approach presented itself almost automatically as ahistorical. It formed a diametrical opposition with the dominant paradigm of historical criticism.24 However, reading the Bible as literature does not require making abstraction of its historical context and evolution. As Ben- jamin rightly remarks, “can it regain its status as The Book if its authors had a religious vision, or must this religious vision address us and call upon us to respond to it; must we, perhaps, even share the ancient vision to plumb this book’s meaning?”25 With regards to history, we can likewise wonder whether we may ignore the Bible’s context and historical value in the reading process or whether every reading includes and implies attention to the historical and religious context. Similarly, we may look at the interaction of the literary with other methods: should we aim for an ‘absolute literary reading’? In other words, the question is not only whether we can separate between the literary and the historical, but also whether that is necessary after all. Each of the above considerations applies to the study of other ancient Jewish texts. There as well, the literary paradigm has often been interpreted as excluding any historical concerns or interests. Yet, some more literary orientated approaches, as the one to be found in Jacob Neusner’s work, nuance this diametrical thinking on the position of the literary: The story is something other than history. Those who read this material as history misread the purpose of the storyteller. Yet those who would abandon the historical dimension in interpreting these stories, who take up the structuralist position on interpreting them and treat them as utterly ahistorical, also err [. . .] To state matters simply: if we do not know who told a story, to whom, and for what purpose, if we cannot account for social context, we

IN: Indiana University Press, 1985); John Gabel & Charles Wheeler, The Bible as Literature: An Introduction (New York: Oxford University Press, 1986); Robert Alter & Frank Kermode, eds., The Literary Guide to the Hebrew Bible (Cambridge, MA: Belknap Press of Harvard University Press, 1987); David Norton, A History of Bible as Literature (Cambridge/New York: Cambridge University Press, 1993); Martin Kessler, ed., Voices from Amsterdam (Atlanta, GA: Scholars Press, 1994). 24 Historical criticism, also called Literaturkritik or literary criticism, should not to be confused with the later literary criticism as it is discussed here. The former shows an inter est in the text’s origin and transmission history, whereas the latter considers the final stage of the text as its starting point. Although the later form of literary criticism did not call for a rejection of the historical as such, it was often felt to be necessary precisely to distinguish itself from the predominant paradigm (David Gunn & Danna Nolan Fewell, Narrative in the Hebrew Bible [New York: Oxford University Press, 1993], 7–12; Amit, Reading Biblical Narratives, 10–14). 25 Benjamin, “The Tacit Agenda,” 270. introduction: thoughts on ancient jewish texts 9

do not yet fully understand that story. Structure without context, that is, the social and economic, material context defined by concrete history, is insuf- ficient either for description or for explanation.26

The Literary and Other Methodological Frameworks At the outset of the literary study of biblical text, it was important for scholars to distance themselves from historical and religious readings. The former resulted from a more academic debate in which the dominant paradigm was questioned, while the latter had more to do with the context outside academia. Meanwhile, the plain literary analysis, although still popular, has been replaced more and more by encounters of the literary with other frameworks. Especially successful are crossovers between literature and linguistics, literature and historicism, and literature and cultural studies. These combinations arise from a growing awareness that ancient Jewish texts fulfill several roles simultaneously, which they might have had originally but have definitely increased during their transmission over time. As the goal of research is to get a better understanding of these texts in all of their aspects (literary, legal, religious, ideological, conceptual, etc.) and not to see how well they fit the prerequisites of a modern scholarly theory, the different roles should be taken into consideration when seeking that insight. The introduction of cognitive linguistics, for example, has shed new light on the use of metaphor in Bible text by freeing the feature from its ornamental role as defined by literary and rhetorical critical studies in the past.27 Unfortunately, the awakening of one method has led to rejection of another, maintaining that metaphor is not a literary feature but a conceptual one. Steps are now taken to bridge that gap by drawing on more than one method.28

26 Jacob Neusner, Neusner on Judaism: Volume 1: History (Aldershot: Ashgate, 2004), 258. Notice that Neusner makes this observation in the history volume of the three volume work that also comprises a book on literature and one on religion and theology. 27 See among others Kurt Feyaerts, The Bible through Metaphor and Translation: A Cognitive Semantic Perspective (Oxford/New York: Peter Lang, 2003); Pierre Van Hecke, ed., Metaphor in the Hebrew Bible (Leuven: Peeters, 2005); Elizabeth Hayes, The Pragmat ics of Perception and Cognition in MT Jeremiah 1:1–6:30: A Cognitive Linguistics Approach (Berlin/New York: de Gruyter, 2008); Job Jindo, Biblical Metaphor Reconsidered: A Cognitive Approach to Poetic Prophecy in Jeremiah 1–24 (Winona Lake, IN: Eisenbrauns, 2010); Antje Labahn, ed., Conceptual Metaphor and the Hebrew Bible (Piscataway, NJ: Gorgias, 2013). 28 For the merger of literature and linguistics, see the current contribution by Ellen Van Wolde, as well as the one by Johan de Joode and Hanneke Van Loon. Outside biblical stud ies, the work of Gerard Steen shows an overt attempt to bridge the gap between cognitive linguistic and literary approaches to metaphor (Gerard Steen, Understanding Metaphor in 10 karolien vermeulen

The Current Volume and the Literary Approach

The contributions in this volume take different stances in the discussion on the relationship between ancient Jewish texts and literary criticism. They look for the literary in biblical as well as non-biblical texts and demonstrate how such an approach can enhance our understanding of the texts. Sometimes the literary is addressed as a means of answering questions other than the literary such as philosophical or historical inquiries. At other times, the literary is the goal, showing us the qualities of texts that have been considered anything but literary over time. The literary paradigm is itself challenged by combining or juxtaposing it with other methods, such as cognitive linguistics, historicism, and comparative (cultural) studies. In all cases, the literary is questioned either as method or as textual characteristic and forms the outset of contributions that—placed together—give an overview of the ‘after of the Art of Biblical and other Ancient Jewish Texts.’ For reasons of convenience the contributions are grouped according to the type of text they discuss: biblical or non-biblical. It is evident, however, that most authors use both, substantiating the biblical argument by non-biblical material and vice versa. The distinction therefore should be mainly taken as a division based on the main corpus discussed in the text rather than as a rigorous categorization of the material into either the biblical or non-biblical camp.

Part One: Hebrew Bible The first part opens with an essay by Scott Noegel (University of Washing- ton) entitled “‘Literary’ Craft and Performative Power in the Ancient Near East: The Hebrew Bible in Context” in which he recognizes the importance and role of literary approaches to biblical texts in terms of acknowledg- ment of the presence of literary features. Yet, he points out the method’s main shortcoming: its inability to provide much insight into the function of the biblical devices. According to him, these are “performative devices of perceived power” rather than “stylistic embellishments.”29 To demonstrate this, he conducts a comparative analysis of the social context of

Literature: An Empirical Approach [London/New York: Longman, 1994]; idem, “The Con temporary Theory of Metaphor—Now New and Improved,” RCL 9,1 [2011]: 26–64). 29 Scott Noegel, “‘Literary’ Craft and Performative Power in the Ancient Near East: The Hebrew Bible in Context,” 21. introduction: thoughts on ancient jewish texts 11 both the biblical text and other ancient Near Eastern texts carrying similar devices. Attention is paid to the connection between literary text production and its cosmological grounds as well as to the ritual setting in which the text was created. A case study of one particular ‘literary’ device, biblical wordplay, exemplifies the observations. Noegel’s approach addresses both the issue of the literary versus the historical and the fruitful collabo- ration of different methods, including the literary, yet in a less prominent role when it comes to interpreting biblical textual devices. Johan de Joode and Hanneke Van Loon (KU Leuven) focus on the nature of literary metaphors in the Hebrew Bible, drawing on the work of cognitive stylistician Gerard Steen on the one hand and pragmatician Adrian Pilkington on the other. Illustrated with metaphors in Job 6:14–21, they propose a way to determine and consequently analyze literary metaphors as opposed to non-literary metaphors as studied in cognitive metaphor theory (CMT). The advantage of their approach is that they see the integration of authorial intent as well as the reader’s response. But they also see possible danger in the assumption that only so-called deliberate, i.e., literary, metaphors should reveal the author’s intent. In addition, they acknowledge the possible overrepresentation of the modern reader’s view, which may bias the findings. De Joode and Van Loon offer an attempt to answer literary questions, such as the role of the author and the reader, within the cognitive paradigm. Interestingly, they seem to remediate the latter’s shortcomings rather than the former’s, by pointing at the lack of ‘deliberateness’ in cognitive linguistic approaches. Implicitly, the authors affirm the literary paradigm as the more suitable frame for analyzing literary metaphors. In the contribution by Angela Roskop (Hebrew Union College-Jewish Institute of Religion) on Exodus 14, a dialogue of the literary with the historical-critical approach is created. The author proposes relevance theory and reception theory as suitable alternative to traditio-historical and source-critical analyses of the passage. Roskop suggests that some elements, such as a complaint motif or a response not to fear, are picked from a cultural repertoire and that they are blended and played out in the text rather than that the use of different motives may indicate different sources only. As such, she acknowledges the various motives recognized by historical and source critics while explaining them, drawing on a more literary oriented approach. The essay shows, as has been argued before, that text analysis of the Hebrew Bible may benefit from an inclusive approach instead of the typical either-or that characterizes the methodological cacophony of the last decades. 12 karolien vermeulen

Similarly, the essay by Klaas Smelik (Ghent University) on the (im)possibility of eyewitness accounts in the book of Samuel centers on a problem that has been addressed before in a mainly historical frame: the nature of the Succession Narrative as eyewitness account. Smelik first gives an overview of the various discussions on the categorization of the passage as a literary unit next to others, such as the Ark narrative, and on its classification as eyewitness account. Since historical-critical and source- critical methods are not able to satisfactorily answer any of the issues raised, Smelik turns to a literary approach that understands the text of 2 Samuel 6 as a unity and the genre as engaged literature. However, in doing so, he does not reject the historical context, but aims for a historical- literary reading of the passage, incorporating the findings of both fields in order to enrich the understanding of the story. The article argues in favor of an approach that overcomes the tension between the literary and the historical; it actually considers the tension to be far less problematic as often assumed. Robert Gordon (University of Cambridge) conducts in his contribution “Battles of Wits and Words: Hushai, Ahithophel and the Absalom Rebellion (2 Samuel 16–17)” a close reading of a much debated passage in 2 Samuel. Paying attention to the construction of the story as well as its exact wording, Gordon concludes that not the divine but the human is played out in the story. The many metaphors, wordplays, and repetitions are aimed at delaying the story and the council. Whereas other approaches have difficulties explaining the double council and the frustration of Ahithophel’s advice, a literary reading can justify the actions since the frustration is considered intended, and the contrast between Husai’s name (“quick”) and his intervention (to slow things down) is interpreted as one of many subtle hints to the audience. Gordon’s contribution shows the strength and usefulness of the literary approach: it is able to answer standing questions in a way other methods cannot and while the author does not reject other readings of the passage, he convincingly shows the validity of a literary analysis. Karolien Vermeulen (Ghent University & University of Antwerp) discusses the Ehud-Eglon episode with its double polysemy, aiming to remediate the symptomatic approach in which the story has been analyzed in the past. Instead of considering the language afterwards, as an additional element of proof, the style of the text is used as starting point of the research, a search for the effect of the wordplay on the story and its characters but also on the interpreter. The first part, focusing on the narrative itself and the functionality of the double polysemy, shows that the introduction: thoughts on ancient jewish texts 13 polysemy device creates opportunities previously locked by the story’s setting: an opportunity to kill the king, but also the possibility to get out unharmed if the plan fails. In such respect, Ehud’s words turn out to be always ‘true,’ i.e., functional within the narrative and benefiting Ehud, the one who uses the tool of language. In the second part, these findings are confronted with the various interpretations formulated over time that consistently limit the phrase’s workability. The author suggests that, whereas each of the interpretations has valuable things to say, they fail in grasping the wordplay’s effect. The contribution merges a narratological interest with cognitive-stylistic observations and reception history, as such breaking through the narrative world focus of literary criticism. Ellen Van Wolde (Radboud Universiteit Nijmegen) shows by means of a case study of the (rain)bow in Gen 9:8–17 how and why to bridge the gap between literary and linguistic approaches in ancient text study. Drawing on the notion of embedded meaning and cognitive domains, Van Wolde over the level of syntactical ענן and קשת moves from the word level of constructions in clauses and sentences to the composite meaning of the text of Genesis 9. The linguistic analysis is supplemented with intertextual, iconographic, and meteorological material, all to support her argument -in the flood narrative most likely was “a warrior’s bow” defin קשת that ing divine power and a “willingness to transfer [t]his power [. . .] to those living on earth.”30 Van Wolde’s approach finds itself at the crossroads of literary and linguistic studies with special attention paid to the extra- textual context of the biblical passage. The essay illustrates the benefits of a combined approach that incorporates the ancient historical setting of the text, which brings her to a new interpretation of a well-known text. The first part of this volume is concluded by Anne-Laure Zwilling (CNRS/ University of Strasbourg) and her piece on “literary tricks.” Zwilling analyses so-called stories of deception in which tricks are part of the plot. She explores the relationship between the trick as narratological topos and the literary form it takes in the text. Starting with the well-known stories of the trickster Jacob in Genesis 27 and 29, the author points out the many differences in the narrative format between the two stories, this despite their thematic similarities. She interprets these differences as a means to make Jacob’s perspective the reader’s. Yet, other stories of tricksterism, such as Joshua 9 or 2 Samuel 3, show different patterns. Whereas Zwilling

30 Ellen Van Wolde, “Bridging the Gap between Linguistics and Literary Studies of Ancient Biblical and Jewish Texts: A Proposal Exemplified by a Study of Gen 9:8–17,” 125. 14 karolien vermeulen concludes that there is “no definite literary design for the rendering of the trick,” she does recognize the trick as a whole as a literary device, “aiming at bringing the reader to take sides.”31 Her contribution shows that the biblical text can be called literary at various levels and that the lack of literariness on one level does not need to exclude its presence at other levels.

Part Two: Other Ancient Jewish Texts In the second part of this volume, Tamar Kadari (Schechter Institute of Jewish Studies) discusses the rabbinic interpretation of the Song of Songs by means of the literary category of poetry. She distances herself from the traditional and rigid categorization of the commentary as being derash, peshat or mysticism to turn to the category of poetry offered by literary studies. According to Kadari, the central questions of the commentary concur with those of reading poetry: how many speakers are there, who is speaking and when and where are they speaking? By using a literary category as analytical tool, she offers insight in the different aspects present in rabbinic interpretation. She shows that the literary can serve as a way to understand rabbinic commentary in all its aspects; it does not replace any of the approaches the rabbis used themselves. The contribution illustrates the added value of the literary and its usefulness in fathoming rabbinic commentary. The essay by Lennart Lehmhaus (Martin-Luther-Universität Halle- Wittenberg) centers on the literary strategies of Seder Eliyahu. The author opens with the observation that the literary turn in biblical and rabbinic text study brought about both advantages and disadvantages. By means of a case study of Seder Eliyahu Zuta and Rabba, Lehmhaus aims at exploring the former and remediating the latter in order to shed light on the “complex interaction with and modification of biblical and rabbinic traditions, hermeneutics, and rhetoric.”32 He points out some of the work’s innovative literary strategies and how they relate to the earlier traditions of the Bible and rabbinic works. A strategy such as the Leitwortstil, later on called “word-motif” by Robert Alter, turns out to also incorporate a linguistic strategy in which a semantic domain is fully explored by the

31 Anne-Laure Zwilling, “The ‘Literary’ in Trickster Narratives: A Comparison of Gen 27:1–39; 29:15–30; 31:19–55; Joshua 9 and 2 Sam 3:26–27,” 162. 32 Lennart Lehmhaus, “Between Tradition and Innovation: Seder Eliyahu’s Literary Strategies in the Context of Late Midrash,” 192. introduction: thoughts on ancient jewish texts 15 writers. This results in so-called “philological midrash,” itself realized by means of literary techniques, although Lehmhaus uses the term with certain caution. His contribution rather favors the in-between position that would generate the neologism ‘literary-philological midrash’ and is an illustration of an inclusive approach that draws both on the literary and the linguistic paradigm. The text under study reveals a similar tendency, being both adoptive and adaptive. A last contribution by Rebecca Kneller (Michlalah Jerusalem College) focuses on a later commentary, that of Samuel-Ibn Tibbon and its intertextual approach to biblical exegesis. Ibn Tibbon’s approach turns out to be both old and new, reconnecting with the midrash idea of filling gaps by means of referring to earlier passages in Scripture as well as adher- ing to the rationalist and logo-centric approach of medieval Spanish and French exegesis that aims at decoding metaphors and allegory. The common denominator in Ibn Tibbon’s work, and in Kneller’s analysis, is the concept of intertextuality that allows getting insight in the biblical exegesis of Ibn Tibbon as well as into his own use of intertextuality. The contribution draws twice on the literary, once through the method of Kneller herself and once through the approach of the author she is studying. The article ties into the discussion of the newness of the literary approach as touched upon before. 16 karolien vermeulen

Appendix—Table of Publications with a Literary Approach (2007–2012) Journal Number Number Number Number Number Number Total 2007 2008 2009 2010 2011 2012 VT 2 of 7 3 of 9 2 of 9 4 of 12 2 of 11 2 of 9 27% 3 of 6 0 of 10 5 of 14 3 of 10 1 of 10 3 of 8 1 of 7 3 of 9 5 of 11 2 of 15 4 of 13 3 of 9 0 of 8 4 of 18 0 of 7 6 of 17 6 of 10 4 of 10 JBL 1 of 5 0 of 6 3 of 12 3 of 6 2 of 7 4 of 10 29% 0 of 6 3 of 7 2 of 6 1 of 7 3 of 7 3 of 11 3 of 5 1 of 6 0 of 8 1 of 6 2 of 6 1 of 11 5 of 9 0 of 6 6 of 8 1 of 4 1 of 4 4 of 10 JSOT 2 of 6 2 of 7 1 of 6 4 of 7 1 of 7 2 of 6 32% 0 of 7 0 of 6 2 of 6 2 of 6 0 of 5 6 of 6 0 of 7 2 of 7 2 of 7 3 of 6 4 of 6 2 of 7 0 of 7 2 of 6 3 of 7 4 of 8 2 of 5 4 of 7 ZAW 0 of 8 1 of 6 1 of 8 1 of 10 1 of 8 2 of 9 13% 1 of 10 1 of 8 1 of 9 1 of 9 0 of 9 0 of 5 0 of 8 1 of 7 0 of 6 2 of 9 0 of 9 4 of 9 2 of 12 1 of 8 2 of 9 2 of 7 1 of 8 2 of 10 JHS 3 of 16 4 of 25 5 of 26 2 of 19 6 of 15 3 of 18 19% JSJ 0 of 5 2 of 4 0 of 3 0 of 3 1 of 3 1 of 5 20% 0 of 4 2 of 7 2 of 5 2 of 4 1 of 5 1 of 5 0 of 3 0 of 3 0 of 3 1 of 2 1 of 4 2 of 4 1 of 4 0 of 4 1 of 3 0 of 4 1 of 4 1 of 7 AJSR 1 of 6 0 of 6 4 of 5 2 of 4 0 of 7 2 of 6 30% 0 of 4 6 of 9 0 of 5 2 of 5 4 of 11 1 of 6 JJS 5 of 10 2 of 8 0 of 9 2 of 8 2 of 7 1 of 7 23% 3 of 9 2 of 8 1 of 8 2 of 9 0 of 7 2 of 7 Proof 0 of 5 1 of 4 0 of 5 0 of 5 0 of 4 0 of 5 16% 8 of 8 0 of 5 1 of 6 1 of 5 0 of 5 1 of 4 1 of 6 0 of 6 0 of 6 23% Part One

Hebrew Bible

“Literary” Craft and Performative Power in the Ancient Near East: The Hebrew Bible in Context

Scott B. Noegel

Abstract

Biblical scholars have done a great deal to advance our appreciation of the literary sophistication of biblical texts. Biblical commentaries and a host of other publications now regularly draw our attention to a multitude of textual devices that operate on micro- and macro-levels such as punning, allusion, inclusio, chiasm, repetition, and ring structure, to name just a few. While such research has at times emphasized literary similarities at the expense of cultural differences, it nonetheless has removed the study of the Hebrew Bible from its relative literary isolation. The combined influence of these factors has allowed scholars to speak about biblical texts on par with Homer, Beowulf, Shakespeare, and other great works of literature. In this essay, I shall argue that while such literary and comparative approaches have helped us to appreciate more fully the repertoire of technical devices employed by the Israelite writers, in the main, they have not helped us to understand their functions. In particular, I shall argue that a comparative examination of the social context for textual production in the ancient Near East suggests that we should think of biblical literary devices not as stylistic embellishments, but as performative devices of perceived power.

ּכְ ֹ֣ ב ד אֱֹ֭להִ ים ֣ רהַסְּתֵ ּדָבָ ֑ ר The glory of God is to conceal a word Prov 25:2 A number of biblical scholars have done a great deal to advance our appreciation of the literary sophistication of biblical texts. Biblical commentaries and a host of other publications now regularly draw our attention to a multitude of clever literary devices, tropes, and figures of speech. The recognition of such devices has allowed scholars to speak about Israelite literary skill on par with the artistry of Homer, Shakespeare, and other great works of literature. What makes the devices especially fascinating is that many of them are not unique to the Hebrew Bible, but appear also in texts from elsewhere in the ancient Near East, such as Mesopotamia, Egypt, and Ugarit. Not only are some metaphors and similes shared, but even compositional 20 scott b. noegel devices such as parallelism, chiasm, repetition, and ring structure appear to have transcended cultural and geographic boundaries.1 This tells us that these devices were learned techniques; wide-spread oral and textual conventions that were passed on within scribal communities. They enjoyed centuries of consistent use and they were employed with a purpose. But what exactly were these devices meant to achieve? I should like to answer this question, in part, by examining Israelite literary craft in the light of comparative evidence from Mesopotamia, Egypt, and Ugarit. I divide my essay into three parts. In the first, I shall argue that we obtain insight into ancient Israelite literary craft by recognizing the cosmological underpinnings that inform the production of literary texts throughout the Near East. Foremost among them is an ontological understanding of words and script as potentially powerful. This conception, I maintain, permits us to understand the Bible’s literary devices not merely as embellishments of style and rhetoric, but as performative devices of perceived power. In the second, I shall argue that the social context for the production of literary texts in the Near East is one closely associated with learned ritual professionals, such as diviners and priests, whose preoccupation with divine power and cosmic order further complicates our notion of literariness. In the third and final portion, I shall build upon my previous observations and consider as a case study the literary device known as “word play.” Drawing upon comparative evidence, I shall argue that the Israelite literati viewed word play as a powerful technology capable of manipulating reality.

Ontological Conception of Words

I begin with the cosmological underpinnings. It is well known that the literati of the ancient Near East regarded words, whether written or spoken, to be inherently, and at least potentially powerful.2 Georges Contenau has

1 See, e.g., Wilfred G.E. Watson, Classical Hebrew Poetry: A Guide to its Techniques (JSOT. SS 26; Sheffield: JSOT Press, 1984); Yitzhak Avishur, Stylistic Studies of Word-Pairs in Biblical and Ancient Semitic Literatures (AOAT 210; Kevelaer: Verlag Butzon & Bercker; Neukirchen- Vluyn: Neukirchener Verlag, 1984), to cite just two important works among many. 2 For a discussion of how the writing systems operate under, and are influenced by different cosmologies, see Scott B. Noegel, “‘Sign, Sign, Everywhere a Sign’”: Script, Power, and Interpretation in the Ancient Near East,” in Divination and Interpretation of Signs in “ literary” craft & performative power in the ancient near east 21 characterized the illocutionary power of words and texts in Mesopotamia as follows: Since to know and pronounce the name of an object instantly endowed it with reality, and created power over it, and since the degree of knowledge and consequently of power was strengthened by the tone of voice in which the name was uttered, writing, which was a permanent record of the name, naturally contributed to this power, as did both drawing and sculpture, since both were a means of asserting knowledge of the object and consequently of exercising over it the power which knowledge gave.3 A classic demonstration of this concept appears in the opening lines of the Babylonian creation account, in which the author describes the primordial world of pre-existence as one that has not yet been put into words. 1. when the heavens above had not yet been termed enūma eliš lā nabû šamāmu 2. nor the earth below called by name šapliš ammatum šuma lā zakrat Enuma Elish I 1–2 Since spoken and written words were understood as capable of wielding divine power, the elite group of literate scribes naturally guarded this power carefully and considered it a secret that was not for the uniniti- ated. Indeed, in their own words, writing was nothing less than the “secret of scribes and gods.”4 It is probably for this reason that we lack native terms for the many textual devices they employed. At the upper end of the scribal elite were the scholars, who referred to themselves as integral links in a chain of transmission originating from the gods. Some scribal masters are even said to have transmitted knowledge directly from the mouth of their patron god Ea, the god of wisdom, magic, music, and the crafts.5

the Ancient World, ed. Amar Annus (OIS 6; Chicago, IL: Oriental Institute of the University of Chicago, 2010), 143–162. 3 George Contenau, Everyday Life in Babylon and Assyria (London: Edward Arnold, 1955), 164. 4 On this topic, see Alan Lenzi, Secrecy and the Gods: Secret Knowledge in Ancient Mesopotamia and Biblical Israel (SAAS 19; Helsinki: The Neo-Assyrian Text Corpus Project, 2008). 5 See Piotr Michalowski, “Sailing to Babylon: Reading the Dark Side of the Moon,” in The Study of the Ancient Near East in the Twenty-First Century: The William Foxwell Albright Centennial Conference, ed. Jerrold S. Cooper & Glenn M. Schwartz (Winona Lake, IN: Eisen brauns, 1996), 177–193. 22 scott b. noegel

Moreover, in so far as writing was perceived as a technology of power, it also served to establish the cosmic order. Several cuneiform texts underscore this by referring to the art of writing as “the (cosmic) bond of everything” (markas kullat).6 The use of writing for keeping the cosmic order is perhaps best embodied in the Mesopotamian conception of divine ledgers or “Tablets of Life” on which gods were believed to inscribe the destinies of individuals.7 The preoccupation with performative power and maintaining cosmic order explains the format and organization of the hundreds of divinatory compendia and lexical lists that the scribal elites produced. According to Mogens Trolle Larson, these texts represent an effort to “present a systematic and ordered picture of the world.”8 Joan Goodnick Westenholz similarly registered the cosmological import of such texts, as she noted, “On the intellectual level, knowing the organization of the world made it possible to affect the universe by magical means.”9 A similar preoccupation with divine efficacy and cosmic control under- girds the Egyptian conception of text. As David O’Connor points out, art and writing were [. . .] ritually and magically empowered to literally transform contexts (temples, tombs, palaces and others) into cosmically charged settings that reflect the belief that the activities carried out in them were effective beyond the human realm.10 Moreover, as in Mesopotamia, the Egyptian priests believed that speech and writing could manipulate the universe, and thus, establish cosmic order. As David Frankfurter states:

6 Ake W. Sjöberg, “In Praise of the Scribal Art,” JCS 24 (1972): 126–131. 7 Shalom M. Paul, “Heavenly Tablets and the Book of Life,” JANES 5 (1973): 345–353. 8 Mogens Trolle Larsen, “The Mesopotamian Lukewarm Mind: Reflections on Science, Divination, and Literacy,” in Language, Literature, and History: Philological and Historical Studies Presented to Erica Reiner, ed. Francesca Rochberg-Halton (New Haven, CT: Ameri can Oriental Society, 1987), 209–212. 9 Joan Goodnick Westenholz, “Thoughts on Esoteric Knowledge and Secret Lore,” in Intellectual Life of the Ancient Near East: Papers Presented at the 43rd rencontre Assyri ologique internationale, Prague, July 1–5, 1996, ed. Jiri Prosecky (Prague: Academy of Sci ences of the Czech Republic, Oriental Institute, 1998), 451–462 (453). 10 David O’Connor, “Ancient Egypt: Egyptological and Anthropological Perspectives,” in Anthropology and Egyptology: A Developing Dialogue, ed. Judith Lustig (MMA 8; Sheffield: Sheffield Academic Press, 1997), 13–24 (17–18). “ literary” craft & performative power in the ancient near east 23

[. . .] Egyptian letters were the chief technology of a hierocratic scribal elite who preserved and enacted rituals—and by extension the cosmic order itself—through the written word.11 The Egyptians also referred to the hieroglyphic script as mdw ntr, literally “the words of the gods,” and the scribal art was to them an occupation without equal. The ibis-headed god Thoth, who is credited with the invention of writing, is said to be “excellent of magic” (mnḥ ḥk¡) and “Lord of hieroglyphs” (nb mdw ntr). He sometimes is depicted writing the hieroglyphic feather sign representing maat (m¡ʿt), a word that stands for the cosmic force of equilibrium by which kings keep their thrones and justice prevails.12 The link between writing and maat underscores how integral the scribal elites perceived their art for maintaining the cosmic order. An illuminating demonstration of the power that written words possessed can be seen in the Ptolemaic Tale of Setne Khamwas, in which a revived mummy describes how written spells were made effective even for the illiterate. I read another formula for writing [. . .] though I cannot write. I was speaking with regard to my elder brother Na-nefer-ka-ptah, who is a good scribe and a very wise man. He caused that a new sheet of papyrus be brought before him. He wrote down every word that was on the papyrus, completely. He burned it with fire; he dissolved it with water. He recognized that it had dissolved; he drank it and he knew that which was in it. Setne I col. 4/1–413 A belief in the power of words is well attested also in the literary texts found at the Canaanite seaport of Ugarit. A textbook example appears in the Epic of Baal, in which we find the dually-named craftsman god Kothar- wa-Hasis ritually empowering the weapons he created so that storm-god Baal might defeat Yamm, the god of the Sea. Of specific import is that he empowers his weapons by naming them. Kothar fashioned two maces, and he pronounced their names: You, your name is “Driver” (ygrš)

11 David Frankfurter, “The Magic of Writing and the Writing of Magic: The Power of the Word in Egyptian and Greek Traditions,” Helios 21 (1994): 189–221 (192). 12 On the concept of m¡ʿt, see Emily Teeter, The Presentation of Maat: Ritual and Legiti mation in Ancient Egypt (SAOC 57; Chicago, IL: Oriental Institute, 1997). 13 Found in Robert Kriech Ritner, The Mechanics of Ancient Egyptian Magical Practice (SAOC 54; Chicago, IL: Oriental Institute of the University of Chicago, 1993), 108. 24 scott b. noegel

Driver, drive (grš) Sea! Drive (grš) Sea from his throne, River, from his seat of dominion. [and over the second mace he proclaims] You, your name is “Expeller” (ʾaymr) Expeller, expell (mr) Sea! Expell (mr) Sea from his throne, River, from his seat of dominion [. . .] KTU 1.2 iv 11–13, 19–20 The two weapons’ functions are embodied in the meanings of their names, Driver and Expeller, and Kothar activates them by transforming their names into verbal commands. As Seth Sanders has described it, “The direct discourse of Kothar is not performative by virtue of social conventions but on a higher level: as sheer self-enacting divine language, Kothar’s speech is performative by cosmic law.”14 Kothar’s ability to empower items through speech is directly connected to his role as a god of magic, music, and technology.15 Elsewhere in the Ugaritic corpus, Kothar-wa-Hasis demonstrates his expertise in metallurgy and architecture. He also is described as a divine mantic. Kothar is your magician kṯr ḥbrk And Hasis your diviner wḫss dʿtk KTU 1.6 vi 49–50 Kothar’s use of language as a vehicle of performative force is again representative of a wide-spread understanding of words. Since the land of Israel served throughout its history as a conduit for cultural influences from Canaan and the dominant Mesopotamian and Egyptian powers around them, we should expect to find a similar conception in Israelite texts. According to Isaac Rabinowitz, this is precisely the case, as he remarks: [in ancient Israel] words were not merely presumed to have the properties of material objects, but might be thought of as foci or concentrations of dynamic power. They were plainly regarded as not only movable but mobile, not only susceptible to being acted upon, but capable of acting upon other

14 Seth Sanders, “Performative Utterances and Divine Language in Ugaritic,” JNES 63 (2004): 161–181 (174). 15 On the connection between word manipulation, technology, and magic, see Scott B. Noegel, Nocturnal Ciphers: The Allusive Language of Dreams in the Ancient Near East (AOS, 89; New Haven, CT: American Oriental Society, 2007), 36–45. “ literary” craft & performative power in the ancient near east 25

entities in ways not confined to communication, of producing and enacting effects, conditions, circumstances and states.16 The conceptual link between a word and a physical object is reflected which means “word” and ,דבר ,most clearly in the Hebrew word dābār also “thing, object.”17 The understanding that spoken words also could embody power is represented best by yhwh’s creation of the universe in the first chapter of Genesis in which the narrator tells us: “And God said, ‘Let there be light’” (Gen 1:3). Like the Mesopotamians and Egyptians, the Israelites also attributed a cosmologically powerful role to writing. One could cite many proof texts, such as the role that divine writing plays in issuing the Ten Com- mandments (Exod 31:18), or yhwh’s heavenly text in which he keeps the names of the sinless (Exod 32:32–33), or the priestly curses that must be written on a scroll, dissolved in water, and imbibed by a wife to test her for unfaithfulness (Num 5:23–24), or of course, the many prophecies that yhwh orders his prophets to put into writing (e.g., Jer 36:18, 36:27–28). Perhaps one of the best demonstrations appears in Numbers 11 in which we hear how yhwh gave a portion of Moses’ spirit to seventy leading Israelites so they could help bear the people’s burdens. In this story, the names of the seventy men are written on a list at the Tent of Meeting, outside the camp. As the text tells us: Now two men stayed behind in the camp, one named Eldad, the second Medad; but as they were among those written (on the list), the spirit rested upon them even though they had not gone out to the Tent; so they were prophetically possessed within the camp. Thereupon a lad ran and told Moses, and said, “Eldad and Medad are prophesying within the camp” (Num 11:26–27). This text illustrates that the written names of the seventy men alone suf- ficed to bring on the spirit of prophesy. The expectation was that prophesying would occur close to the Tent of Meeting and not within the camp.18 Such references could be multiplied, but these should suffice to demonstrate that speaking and writing in the ancient Near East could be perceived as acts of cosmological power. This conception of words would appear to be a necessary starting point for understanding the perceived

16 Isaac Rabinowitz, A Witness Forever: Ancient Israel’s Perception of Literature and the Resultant Hebrew Bible (Bethesda, MD: CDL Press, 1993), 16. 17 Discussed by Rabinowitz, A Witness Forever, 9–12. 18 Rabinowitz, A Witness Forever, 34. 26 scott b. noegel nature of language, writing, and text in the Hebrew Bible. Nevertheless, it is seldom integrated into studies of scribal culture or text production, and even more rarely into studies on literary craft. Moreover, it is no accident that the illocutionary power of words was closely affiliated with gods of technology. You will recall, that in Mesopo- tamia it was associated with Ea, in Egypt, with Thoth, and at Ugarit, with Kothar-wa-Hasis—each wielding skill in magic and crafts. The connection to these gods offers additional evidence that the ancients conceived of literary craft as a technology that could influence reality. I contend that such a context must also inform our understanding of its devices.

Social Context for the Production of Literary Texts

The social context for the production of literary texts throughout the ancient Near East also was closely associated with performative ritual and mantic praxis. In Mesopotamia, the central role that mantic professionals played in the production and dissemination of literary texts should call into question how we understand Mesopotamian “literature” and its devices. At least from about 1500 bce and onwards, a master scholar, whether serving as an author in the modern sense or merely as a redactor, would have been responsible for transmitting a wide variety of texts including the full gamut of divinatory lore, including omens, medical texts, incantations, prayers, and a variety of myths. He would have been perceived as having transmitted oral and textual traditions directly from the divine. Therefore, to understand the function of literary devices in Mesopotamia we must foreground this context. Perhaps no Mesopotamian text best demonstrates the overlap between the worlds of mantic practice and literary production as the famous Epic of Gilgamesh. Of particular interest is that the epic was redacted and expanded by a kalû-priest named Sîn-lēqi-uninnī.19 Our knowledge of this person is unfortunately scant. Nevertheless, we know by his profession that he represented the pinnacle of Mesopotamian intellectual and eso-

19 Paul-Alain Beaulieu, “The Descendants of Sîn-lēqi-unninni,” in Assyriologica et Semit ica: Festschrift für Joachim Oelsner anlässlich seines 65. Geburtstages am 18. Februar 1997, ed. Joachim Marzahn & Hans Neumann (AOAT, 252; Münster: Ugarit Verlag, 2000), 1–16. Wilfred G. Lambert, “A Catalogue of Texts and Authors,” JCS 16 (1962): 59–77, originally suggested the reading mašmaššu, though this appears less likely in the light of the work by Beaulieu. “ literary” craft & performative power in the ancient near east 27 teric thought, encompassing mastery in astronomy, astrology, mathemat- ics, mythical history, magic, divination, and ritual. The kalû-priests were the true polymaths of ancient Mesopotamia. They were a highly interdisciplinary lot whose many areas of expertise were subsumed under the label of “wisdom.” As I have written elsewhere, when read through the lens of the kalû-priestly profession, the Epic of Gilgamesh betrays signs of that craft’s erudition and ideology.20 Given this background, it is difficult to think of the epic as literature qua literature and devoid of performative detail and function. Indeed, Assyriologists have long observed a close relationship between Mesopotamian ritual and mythological re-enactments, as well as the use of mythological references in ritual and divinatory texts. More recently, they have begun to question the degree to which Mesopotamian literary and mythological traditions acquired a performative function.21 Since mantic interests played a central role in the creation, transmission, and redaction of literary traditions, it is plausible to think that when the scribal masters transmitted their mythological lore, the recitation of its contents consti tuted a demonstration, if not activation, of its performative power. We derive added insight into the performative dimension of ancient literary texts by turning to the work of historians of religion, who have begun to ask similar questions about the performative power of narration. A representative example is Frankfurter’s study of Egyptian historiolae. A historiola is a brief tale built into a magical text that provides a mythic precedence in order to make a charm magically effective. Frankfurter explains how this works in his definition of “narrating power.” [Narrating power]: a ‘power’ intrinsic to any narrative, any story, uttered in a ritual context, and the idea that the mere recounting of certain stories situates or directs their ‘narrative’ power into this world. Egyptologists have long been familiar with this concept of narrative, since ancient Egyptian ritual traditionally involved the recitation of mythic narratives as a kind

20 Noegel, Nocturnal Ciphers, 57–82. 21 See, e.g., Jean van Dijk, “Une incantation accompagnant la naissance de l’homme,” Or 42 (1973): 502–507 (505); Walter Burkert, “Literarische Texte und funktionaler Mythos Ištar und Atrahasis,” in Funktionen und Leistungen der Mythos: Drei altorientalische Bei spiele, ed. Jan Assmann, Walter Burkert & Fritz Stolz (OBO 48; Göttingen: Vandenhoeck and Ruprecht, 1982), 63–82; Ulla Jeyes, Old Babylonian Extispicy: Omen Texts in the British Museum (Istanbul: Nederlands Historisch-Archaeologisch Instituut te Istanbul, 1989), 186, n. 8; Alasdair Livingstone, “The Magic of Time,” in Mesopotamian Magic: Textual, Historical, and Interpretive Perspectives, ed. Tzvi Abusch & Karel van der Toorn (AMD 1; Groningen: Styx, 1998), 131–137. 28 scott b. noegel

of instrumental praxis, but also because Egyptians have a highly nuanced sense of the power of the spoken word.22 Though Frankfurter’s texts functioned in different social and cultural matrices, much of his definition is applicable to the Mesopotamian mythological corpus. Consider again, for example, the Babylonian creation story Enuma Elish, whose tablets narrate the creation of the cosmos and provide a hierarchy of beings. This narration too contains a citation of events and acts performed in cosmogonic time, and thus, we might understand it as activating its performative power by establishing precedence and paradigm. Perhaps, we should think of it as becoming dynamically real within the ritual context of reciting it. In Egypt, the social context for the production of literary texts also was located in ritual, for advanced literacy was exclusively the provenance of the priesthood. There is little doubt that the recitation of Egyptian texts that we today classify as “literary” bore cosmological import. Obvious examples, such as the Pyramid and Coffin Texts, and the Book of the Dead, aimed to secure for the deceased a safe place in the afterlife. When reciting these texts, Egyptian priests believed they were not merely describing the journey, they were making it happen; they were participating in cosmogonic time. The same may be said of many other texts that one now finds in antholo- gies of Egyptian literature. For example, the story of the Tale of Two Brothers has been understood as a complex cosmic allegory whose characters per- sonify deities.23 The Tale of the Shipwrecked Sailor is widely regarded as a hallmark of Egyptian literature. Nevertheless, John Baines has argued that we read it as a didactic piece that spoke directly to the esoteric ideology of the solar cult.24 Even its circular compositional form, he suggests, embodies and enacts the cyclical nature of the cosmos. The autobiographical fiction known as the Tale of Sinuhe is similarly considered one of the fin- est pieces of Egyptian literature. Yet, it too has been understood as an allegory with a cosmic dimension. The journey undertaken by its main character, from Egypt to Canaan and back, is reminiscent in theme and

22 David Frankfurter, “Narrating Power: The Theory and Practice of the Magical Histo riola in Ritual Spells,” in Ancient Magic and Ritual Power, ed. Marvin Meyer & Paul Mirecki (RGRW 129; Leiden: E. J. Brill, 1995), 457–458. 23 On the diverse interpretations this text has received, see Susan T. Hollis, The Ancient Egyptian “Tale of Two Brothers”: A Mythological, Religious, Literary, and Historico-Political Study (Oakville, CT: Bannerstone Press, 2008). 24 John Baines, “Interpreting the Story of the Shipwrecked Sailor,” JEA 76 (1990): 55–72. “ literary” craft & performative power in the ancient near east 29 language of cultic texts that ritualize death and entrance to the afterlife.25 Indeed, there are few Egyptian literary texts that have not been mined for their cultic and cosmological meanings, because the social context for text production in Egypt was the priesthood. At Ugarit, the evidence for the production and dissemination of literary texts also adheres to this pattern. The literary texts discovered there, such as the epics of Baal, Aqhat, Keret, and the Rephaim were all located in the private house of the high priest, between the Baal and Dagan temples.26 In fact, a colophon appended to the end of the Baal epic tells us that the text was written by a scribe who is described as follows: [The] scribe is Ilimilku from Shuban, spr. ilmlk. šbny disciple of Attenu the diviner, lmd. atn. prln chief of the priests, rb.khnm chief of the shepherds, rb.nqdm (and) secretary of Niqmaddu, ṯʿy. nqmd the king of Ugarit mlk.ugrt KTU 1.6 54–58 The colophon leaves little doubt that, just as in Mesopotamia, diviners with priestly functions played central roles in the creation and transmission of literary texts.27 Indeed, as Olof Pedersén has shown, from roughly 1500–300 bce the great majority of Mesopotamian and Ugaritic texts that we might today label as literary, such as prayers, laments, myths, wisdom texts, and epics, and those we might label scholarly, such as lexical and grammatical lists, astrological reports, omens, and medical texts—were found in temple libraries or private archives owned by priests and diviners.28 Texts of a more mundane type, such as letters, treaties, loan and tax dockets,

25 On the multiple meanings conveyed in Sinuhe, see John Baines, “Interpreting Sinuhe,” JEA 68 (1982): 31–44. 26 See Olof Pedersén, Archives and Libraries in the Ancient Near East, 1500–300 B.C. (Bethesda, MD: CDL Press, 1998), 73–74. 27 See also Wilfred van Soldt, “Atn prln, ‘Attā/ēnu the Diviner,’” UF 21 (1989): 365–368. 28 Pedersén, Archives and Libraries in the Ancient Near East, 1500–300 B.C., 273–282. 30 scott b. noegel chronicles, and lists of employees and supplies, were typically found in palace archives or those of private officials. Unfortunately, much less is known about the social and cultural background for the production of literary texts in ancient Israel. While positing a role for mantics in the creation of the Bible’s prophetic texts seems clear, we are on less sure footing positing such a background for the production of narrative and other non-prophetic texts. Nevertheless, since biblical texts contain the same devices found in other Near Eastern literary texts, and since they evince a conception of speech and writing on par with Mesopotamian, Egyptian, and Canaanite dogma, it is plausible to think that the priesthood played some role in the production of non-prophetic literary texts, even if such a role was state-sponsored.29 This conclusion adds further support to Karel Van der Toorn’s view that we should see the temple in Jerusalem as the locus of the scribal elite whose primary period of collection, copying, and transmission occurred roughly between the years 500 and 200 bce. As he concludes: The affiliation of the scribal school [. . .] to the temple is especially important in view of the role of the school as a center of text production. In the ancient Near East, the men who taught others to read texts were also the men who wrote texts themselves. All over the Near East, schools were not merely centers of text transmission but also of text composition. While the temple scribes in Israel were responsible for teaching the scribal craft, they were also the ones who created the bulk of the biblical literature.30 Thus far, I have argued that in order to understand ancient Israelite literary craft we must recognize the importance of two factors usually not integrated into the study of biblical literature. The first is a conception of words and texts as powerful technologies capable of manipulating reality. The second is a well-evidenced social context that places literary production in the hands of mantics, diviners, and/or priests, who viewed themselves as receivers, transmitters, and vehicles of the divine word.

The Problem with “Literary” Devices: The Case of “Word Play”

With these factors in mind, I should like to look anew at one of the Hebrew Bible’s literary devices. Specifically, I shall focus on the device commonly

29 See Noegel, Nocturnal Ciphers, 177–182, 278. 30 Karel van der Toorn, Scribal Culture and the Making of the Hebrew Bible (Cambridge, MA: Harvard University Press, 2007), 89. “ literary” craft & performative power in the ancient near east 31 referred to as “punning” or “word play,” of which there are essentially two basic types: polysemy and paronomasia.31 “Polysemy” refers to devices that involve multiple meanings in a single context, whereas “paronomasia” refers to devices involving sound that function across word divisions and involve a dissimilarity in meaning. Word play has long been observed in biblical texts. The early rabbis lāšōn nōpēl ʿal ,lāšōn “language falling לָ שֹׁון נֹופֵ ל עַ ל לָ שֹׂון referred to it as upon language” (Gen. Rab. 18:6, 31:8). Though the topic still lacks comprehensive study, it is fair to say that scholarship on the topic, with few exceptions, has understood the device as one of style and rhetorical flare. In his now classic study authored in 1893, Immanuel Casanowicz characterized paronomasia as follows:32 Paronomasia in the Old Testament is, like all other embellishments of speech, an element of higher style, that is, of the poetical and prophetical diction. In the historical books, except in the poetical passages embodied in them and the plays on the etymology of proper names, cases in which it occurs are few and far between. It is everywhere merely a casual, not an organic, element of diction. Hebrew poetical style hardly differs from the rhetorical; both have in common all the peculiarities which distinguish them from the lower style.33 Though scholars of more recent vintage typically avoid imposing a higher or lower value on biblical word play, the perception of its function as a device of style and rhetoric remains the dominant view. Compare this with Raymond Faulker’s discussion of the device in Egyp- tian texts. Under a rubric entitled “The Magical Power of Word Play,” Faulkner states: Since words were a major category of images for the Egyptians, manipulating the sounds or the signs of a word was thought to affect the object it represented. The goal of such word play, or paronomasia, was far more than the creation of incantations with mysterious sounds. [. . .] Word play was an important method of linking the earthly realm with the world of the gods.

31 On this subject, see Scott B. Noegel, “Polysemy,” in Encyclopedia of Hebrew Language and Linguistics, ed. Geoffrey Khan (Leiden: Brill, 2013), in press; “Paronomasia,” in Ency clopedia of Hebrew Language and Linguistics, in press; Puns and Pundits: Wordplay in the Hebrew Bible and Ancient Near Eastern Literature (Bethesda, MD: CDL Press, 2000); Noctur nal Ciphers, 9–11; “Word Play” in Ancient Near Eastern Texts [forthcoming]. 32 Immanuel M. Casanowicz, “Paronomasia in the Old Testament,” JBL 12 (1893): 105–167. The article derives from work found in Immanuel M. Casanowicz, Paronomasia in the Old Testament (PhD diss., Johns Hopkins University, 1892/Boston, 1894). 33 Casanowicz, “Paronomasia in the Old Testament,” 120. 32 scott b. noegel

Egyptian puns seem heavy-handed, but they were not intended to amuse. They created an alternate focus which could deflect the power in a word.34 Antonio Loprieno similarly sees word play in Egyptian as a performative device, one that was perceived as capable of manipulating the cosmos while serving as a method of “scientific classification of the world and its entities.”35 See, for example, the following Egyptian dream omen.36 Here we read that if one dreams of [. . .] cutting up a bull with his own hand; good, (it means that) his (own) opponent will be killed. [. . .] ḥr sf.t jḥ m ḏr.t=f; nfr, sm¡ p[¡] y=f jry-n-ʿḥ¡(.t) r. 4:16 Here the bovine hieroglyphic sign (i.e., ⁄) is read as jḥ “bull.” Neverthe- less, it forms a lexical association with the word sm¡ meaning “killed,” because the latter word also can mean “wild bull,” and even takes as its determinative the bovine glyph (i.e., ⁄ or ¡). See similarly the Egyptian Coffin Texts in which the Egyptian Lord of All proclaims: ™w rmt m rmwt ™rt=™ “I made humankind from tears” (Spell 1130, § 465a). Here the word for “humankind” (rmṯ) resounds in the word for “tears” (rmwt). The paronomasia forges a creative link, and thus binds the essences of the divine substance of creation and the created mortals, while it also classifies the place of humankind within the cosmos.37 The understanding of word play among scholars of Mesopotamia is very similar, as Sheldon Greaves’ comment illustrates: [. . .] word play was thought to play an active role in magic by taking advantage of the linkage that was thought to exist between the word for an object and the object itself. In practical terms this means that if the magician can use a verb or an object in the incantation that puns with the object or condition he or she is trying to alter, the association creates a link to that object that will achieve the desired result.38

34 Raymond O. Faulkner, et al., The Egyptian Book of the Dead: The Book of Going Forth by Day (San Francisco, CA: Chronicle Books, 1994), 147. 35 Antonio Loprieno, “Puns and Word Play in Ancient Egyptian,” in Puns and Pundits, 3–22 (13). 36 See Scott B. Noegel & Kasia Szpakowska, “‘Word Play’ in the Ramesside Dream Man ual,” SAK 35 (2007): 93–212; text on 200. from the dust of the (האדם ,See similarly the creation of “the man” (hā-ʾādām 37 .in Gen 2:7 (האדמה ,ground” (hā-ʾădāmāh“ 38 Sheldon W. Greaves, “Ominous Homophony and Portentous Puns in Akkadian Omens,” in Puns and Pundits, 103–116 (113). “ literary” craft & performative power in the ancient near east 33

Two Babylonian omens illustrate this well. The first is a dream omen.39 If a man dreams that he is traveling to Idran (id-ra-an); (it means) he will free himself from a crime (Á-ra-an). K. 2582 rev ii, x + 21 Here the cuneiform sign id also can hold the value Á, which permits the diviner to read the toponym Idran as the word aran, meaning ‘crime.’ The second omen derives from the practice of extispicy, the reading of abnormal features on the liver of an animal.40 When (the) lobe is like the grapheme (named) kaškaš, (then) (the storm god) Adad will inundate (with rain). Here the name of the cuneiform sign kaškaš suggests to the diviner the word kaškaššu, meaning “all powerful,” which is an epithet used of the storm god Adad. These omen texts illustrate that the use of word play did not mark elevated style, but rather served to make sense of the divine sign.41 Since words were loci of divine power, an ambiguous omen put into words naturally represented a potentially unbridled form of power. Punning interpretations limit that power by restricting the parameters of an omen’s interpretation. The omen cannot now mean anything, but only one thing. Therefore, the employment of word play constitutes an act of power. Moreover, diviners manipulated not just words, but behavior and belief. By deploying the performative puns, they determined an individual’s fate. At some level, therefore, we must see the function of the punning technology as a form of social and cosmic control. Moreover, it is important to realize that acts of divination also constitute acts of divine judgment. Not only do diviners use the word purussû “legal decision” or “verdict” to refer to an omen’s prediction, but as Francesca Rochberg has shown, divinatory texts share in common with legal codes the formula if x, then y.42 Therefore, as I have concluded elsewhere: [. . .] within this performative juridical context, word plays connecting omens to their interpretations constitute vehicles for demonstrating and justifying divine judgment. They are persuasive legal arguments based on

39 Discussed in Noegel, Nocturnal Ciphers, 21. 40 See Noegel, Nocturnal Ciphers, 13. 41 See Noegel, “‘Sign, Sign, Everywhere a Sign’.” 42 Francesca Rochberg, “Empiricism in Babylonian Omen Texts and the Classification of Mesopotamian Divination as Science,” JAOS 119 (1999): 559–569. 34 scott b. noegel

analogy that establish precedent. Insofar as they underscore the tie between the sign and its prediction, they illustrate the principle and process of lex talionis “the law of retribution.” Thus, the punning not only affirms theological and legal principles, it embodies them; and it is more performative than literary, since words index power and since the acoustic impact is only the result of a pun’s function.43 Since the literati who composed the omen compendia also created the literary texts, their word plays in the latter corpus function similarly as indices of transformative power and vehicles of divine judgment. To illustrate further, I turn to the Epic of Gilgamesh, in which we hear the god Ea instructing Utnapishtim, the Babylonian Noah, what to tell the people when they see him building an ark. Since we later learn that this message came in the form of a dream, the god’s message constitutes an omen. According to the text, Ea instructs him to say: He (Ea) shall provide you with abundance (or rain down a torrent) [. . .] eli kâšunu ušaznanakunūši nuḫšamma [. . .] in the morning, cakes (kukkū), [ina šēr] kukkī [. . .] and in the evening, he shall provide (or rain down) a “rain” of wheat (kibātu). ina līlāti ušaznanakunūši šamūtu kibāti Epic of Gilgamesh XI 43–47 For the erudite, there is a hidden message here. The words “cakes” (kukkū) and “wheat” (kibātu) can be read as “darkness” (kukkû) and “heaviness” (kibittu).44 Moreover, the verb zanānu, can mean “provide with food” or “rain down,” and the word nuḫšu, “abundance,” can refer to “agricultural yield” or “flood waters.”45 Thus, in one statement the god of magic and craft conceals his real intentions and relies on Utnapishtim’s wisdom to decode the signs of Ea’s divine missive.46 As the aforecited examples demonstrate, throughout the ancient Near East word play was not understood as a feature of literary style and

43 Noegel, Nocturnal Ciphers, 44–45. On word play as a means of demonstrating lex tal ionis in biblical narrative, see Scott B. Noegel, “Drinking Feasts and Deceptive Feats: Jacob and Laban’s Double Talk,” in Puns and Pundits, 163–179. 44 Carl Frank, “Zu den Wortspielen kukku und kibâti in Gilg. Ep. XI,” ZA 36 (1925): 218. On other omenistic puns in Gilgamesh, see Noegel, Nocturnal Ciphers, 57–83. 45 Scott B. Noegel, “Raining Terror: Another Wordplay Cluster in Gilgamesh Tablet XI (Assyrian Version, ll. 45–47),” NABU 75 (1997): 39–40. 46 On the connection between word play and wisdom, see Noegel, Nocturnal Ciphers, 28–35. “ literary” craft & performative power in the ancient near east 35 rhetorical flare. Such a view is not in keeping with the ontological conception of words and text as vehicles of power or with the social context for the production of literary texts. Indeed, it is more appropriate to think of word play as a technology of power and a vehicle for enacting divine judgment. I submit that a close examination of word play in the Hebrew Bible suggests that the Israelites employed the device to similar ends. Two biblical texts will illustrate my point, though I could cite many more.47 I first turn to the tower of Babel (Gen 11:1–9). 1. now the whole world had one language and a common speech. 2. as men moved eastward, they found a plain in Shinar and settled there. 3. they said to each other, “Come, let us make bricks (nilbǝnā̄h lǝbēnîm, (הַ ּלְבֵ נָ ה ,and bake them.” They used the brick (hal-lebēnāh (נִלְּבְ נָ ה לְבֵ נִ ים for mortar ( הַ חֵ מָ ר ,and tar (ha-ḥēmār ,( לְ אָ בֶ ן ,instead of stone (lǝ-ʾāben .(לַחֹמֶ ר ,la-ḥōmer) ,ourselves a city (נִבְ נֶה־ּלָ נּו ,then they said, “Come, let us build (nibneh lānū .4 with a tower that reaches to the heavens, so that we may make a name for ourselves and not be scattered over the face of the whole earth.” 5. but yhwh came down to see the city and the tower that the sons of .(ּבָ נּו ּבְ נֵ י הָ אָ דָ ם , humankind were building (bānū bǝnē̄ -ādām ha 6. Yhwh said, “If as one people speaking the same language they have begun to do this, then nothing they plan to do will be impossible for them. their language so they ( נָ בְ לָ ה ,Come, let us go down and confuse (nābǝlāh .7 will not understand each other.” 8. so yhwh scattered them from there over all the earth, and they stopped .the city ( לִ בְ ֹנ ת ,building (libnōt -because there yhwh con ,( ּבָ בֶ ל ,that is why it was called Babel (bābel .9 the language of the whole world. From there, yhwh ( ּבָ לַ ל ,fused (bālal scattered them over the face of the whole earth. The passage contains several cases of paronomasia that set the scene for lingual manipulation and inversion. In particular, note the narrator’s statement in verse 3 that the Babylonians built their structure with brick instead of stone and tar instead of mortar. The replacement of one item for another is embodied in word play. In this case the expression “let us ”is contrasted with “stone ( נִ לְ ּבְ נָ ה לְבֵ נִ ים ,make bricks” (i.e., nilbǝnā̄h lǝbēnîm ”replaces “mortar ( הַ ֣ חֵ מָ ר ,and the “tar” (i.e., ha-ḥēmār ( לְ אָ בֶ ן ,i.e., lǝ-ʾāben)

47 I have discussed the two passages elsewhere in a different, though related, context. See Scott B. Noegel, “The Ritual Use of Linguistic and Textual Violence in the Hebrew Bible and Ancient Near East,” in State, Power, and Violence, ed. Margo Kitts, et al. (Vol. 3 of Ritual Dynamics and the Science of Ritual; Wiesbaden: Harrassowitz, 2010), 33–46. 36 scott b. noegel

-Far from being embellishments, these puns char .(לַחֹמֶ ר ,i.e., la-ḥōmer) acterize the materials of construction as inversions of Israelite normative practice, and thus as aberrations of the cosmic order. For their act of hubris in trying to make a name for themselves, the Babylonians receive a divine judgment that is again carried out by means of paronomasia. In verse 7 the performative force of yhwh’s words “let us manipulates the very letters contained in the ( נָ בְ לָ ה ,confuse,” (i.e., nābǝlāh Babylonians’ original pronouncement “let us make bricks” (i.e., nilbǝnā̄h Indeed, the power of God’s lingual transformation .(לְבֵ נִ ים נִ לְ ּבְ נָ ה ,lǝbēnîm is underscored at tale’s end when the narrator explains: “That is why he ,because there yhwh confused (bālal ,( ּבָ בֶ ל ,called its name ‘Babel’ (bābel the language of the entire earth” (Gen 11:9). In effect, God altered ( ּבָ לַ ל the essence and destiny of Babylon simply by manipulating the letters in its name. Since the text’s puns tie Babylon’s deeds directly to yhwh’s judgment, they again embody the juridical and theological concept of lex talionis. Jeremiah’s prophecy against Babylon (Jer 51:34–37) will be my final demonstration of the performative power of puns. Here again we find ourselves in a juridical context as verse 36 makes clear.48 34. nebuchadnezzar king of Babylon has devoured us: He has thrown us into confusion; he has made us an empty jar. he has swallowed us ,(ּתַ ּנִ ין ,like the primordial serpent (tannîn (ּבְלָעָ נּו ,bǝlāʿānū) and filled his stomach with our delicacies, and then he has spewed us out. 35. “May the violence done to us and our children be upon Babylon (bābel, .say the dwellers of Zion ”,( ּבָ בֶ ל says ”,( ּבָ בֶ ל ,May our blood be on those who live in Babylonia (bābel“ Jerusalem. 36. therefore, this is what Yhwh says: (הִ נְנִי־רָ ב אֶת־רִ יבֵ ְך ,See, I will defend your cause (hinǝnî rāb ʾet rîbēk“ and avenge you; i will dry up her sea, and make her fountain run dry.

48 On the rîb, “cause,” “case,” as grounded in legal usage, see Berend Gemser, “The Rîb- or Controversy Pattern in Hebrew Mentality,” in Wisdom in Israel and in the Ancient Near East, ed. Martin Noth & D. Winston Thomas (VT.S 3; Leiden: Brill, 1955), 120–137; Herbert Huffmon, “The Covenant Lawsuit in the Prophets,” JBL 78 (1959): 285–295; James Limburg, ,and the Prophetic Lawsuit Speeches,” JBL 88 (1969): 291–304; Kirsten Nielsen ריב The Root“ Yahweh as Prosecutor and Judge: An Investigation of the Prophetic Lawsuit (Rîb Pattern), (JSOT.SS 9; Sheffield: University of Sheffield Press, 1978). “ literary” craft & performative power in the ancient near east 37

a den of jackals ,(ּגַּלִ ים ,babylon shall become rubble heap (gallîm .37 ”.an object of horror and hissing, without inhabitant ,(ּתַ ּנִ ים ,tannîm) In this prophecy, Jeremiah invokes several forms of polysemy and paronomasia to affect Babylon’s violent reversal of fortunes. His first is the word -in verse 37, a polyseme that can mean “water waves” or “rub ,ּגַּלִ ים ,gallîm ble heap.” Since God has just stated that he will dry up Babylon’s waters, gallîm first suggests the meaning “waves.” It is only when we hear the remainder of the passage and its reference to wasteland that we realize it must mean “rubble heap.” In essence, the prophecy has transformed Baby- lon’s abundant “waters” into “rubble” simply by changing the linguistic context of a single word—the transformation happens in the recitation. -jack“ ,ּתַ ּנִ ים ,A second use of transformative punning is the word tannîm als,” in verse 37. Just previous to this, in verse 34, yhwh had described i.e., “the primordial serpent of chaos.” By altering ,ּתַ ּנִ ין ,the king as a tannîn one consonant, the prophet transformed the serpent of chaos into a lair for jackals. The prophecy continues with performative language that ties “Baby- in verse (ּבְלָעָ נּו ,with the act of swallowing (bǝlāʿānū ( ּבָ בֶ ל ,lon” (i.e., bābel 34. These words are resound later again in verse 44 where yhwh issues ּופָקַדְּתִ י עַל־ּבֵ ל ּבְבָבֶ ל וְהֹצֵאתִ י אֶת־ּבִלְ עֹומִּפִ יו וְ לֹא־יִנְהֲ רּו אֵ לָ יו עֹוד :his verdict I shall punish Bel in Babylon, and I will make“ ,ּגֹויִם ּגַ תם־חֹומַ ּבָבֶ ל נָפָלָ ה him disgorge what he has swallowed. The nations shall no longer stream to him, and the wall of Babylon shall fall.” Here the words “in Babylon” ,and the god “Bel” (bēl ,(ּבִלְ עֹו ,swallowed” (i.e., bilʿō“ ,( ּבְ בָ בֶ ל ,i.e., bǝ-bābel) recall the primordial dragon who had swallowed Jerusalem, and in ,( ּבֵ ל so doing, they connect Babylon’s deeds with its punishment. Note also -wǝ-lōʾ-yinhărū, “they shall no lon ,וְ לֹא־יִנְהֲ רּו ,how the words wǝ-lōʾ-yinhărū ger stream,” remind us of the drying of the sea and fountains in verse 36 in verse 37. The referential language ,ּגַּלִ ים ,and the punful use of gallîm again embodies the very means by which the principle of lex talionis is activated. The reversal of fortunes through the performative power of words reaches a climax in verse 41 by way of an Atbash cipher. An Atbash occurs when one exchanges the first letter of the alphabet with the last, the second with the penultimate, the third with the antepenultimate, and so on.49 The alphabetic reversal results in transforming the name “Babylon” (bābel,

49 For other instances of Atbash in Jeremiah, see Scott B. Noegel, “Atbash in Jeremiah and Its Literary Significance,” Parts 1–3, JBQ 24/2, 24/3, 24/4 (1996): 82–89, 160–166, 247–250. 38 scott b. noegel

a word devoid of meaning, and thus ,( שֵׁ שַׁ ְך ,into “sheshak” (šēšak ( ּבָ בֶ ל without essence or destiny. Yhwh has turned it into a “rubble heap” of letters. In this essay, I have argued that in order to understand ancient Isra- elite literary craft we must recognize the importance of two factors. The first is a conception of words and texts as powerful technologies capable of manipulating reality and activating divine judgment. The second is a social context that places literary production in the hands of mantics, diviners, and/or priests. Though I have used cases of punning to illustrate how textual devices operated beyond the realm of literary and rhetorical flourish, the evidence I have raised naturally calls into question the presumed “literary” functions of other devices often categorized as elements of style. For example, is it possible that similes and metaphors served as performative technologies in order to connect the essence of one thing to another, as they do in ancient Near Eastern incantations? Is it possible to understand repetition as more than just an emphatic feature of persuasion? Might repetition have served, as it does in ancient magic texts, as a formulaic means of legitimating and strengthening the power of the utterance? And what about so-called compositional devices such as chiasm, parallelism, and ring structure? Were they too perceived as having more than a prosodic or organizational purpose? Since they often appear in poems that were intoned and accompanied by music, could they have functioned not merely to express the numinous, but like music itself, to invoke it? While a preoccupation with performative power might not inform every textual device typically deemed literary, I do think we have good reason to con- template it. Selecting and Analyzing Metaphors in the Hebrew Bible: Cognitive Linguistics and the Literary

Johan de Joode and Hanneke van Loon

Abstract

In Cognitive Linguistics, it is customary to stress the ubiquity of conceptual meta phors. In literary studies, however, critics tend to highlight the special character of specific metaphors in texts. Exegetes using the tenets of Conceptual Metaphor Theory have thus far been at odds with the ubiquity of conceptual metaphors in literary texts: do all conceptual metaphors (and they are many) deserve equal attention? Do they all equally affect the text and the reader’s experience? In this article, we study what is needed to discuss literary metaphors in Biblical Hebrew texts based on the theorems of Cognitive Linguistics. We combine Steen’s find ings on the deliberate use of metaphor with those of Pilkington on the range and strength of mappings. As such, we introduce criteria to first select and sub sequently analyze metaphors in the biblical corpus. We advocate for explicitness and illustrate our point with an example from the book of Job (6:14–21).

Introduction

Conceptual Metaphor Theory (henceforth CMT) is undoubtedly one of the most popular—if not the most popular—current theory on metaphor. Anyone interested in metaphor cannot ignore the fairly recent insight that metaphor is first and foremost a conceptual rather than a linguistic phenomenon. As such, cognitive linguists teach us that metaphor, being conceptual, is ubiquitous. It is an established fact that many examples (some of which will follow later in this article) illustrate that point. Exe getes, ourselves included, generally yield the point of the ubiquity of con ceptual metaphor, yet are confused about the impact this theorem has on the analysis of metaphor in literary texts. Has metaphor not always had a special place in literary studies? Should exegetes now deal with each and every conceptual metaphor one can find in biblical poetry or prose? Are we to study all metaphors in a text? Are there conceptual metaphors in a literary text that specifically call for our attention? In this contribution, we propose two distinctive steps for the analysis of metaphor in literary texts, one heuristic and the other analytical. First, we search for a systematic way of selecting certain conceptual metaphors that might be worth the 40 johan de joode and hanneke van loon while for literary critics. Based on Steen, we argue that exegetes can opt to focus on deliberately used metaphors.1 Consequently, our quest leads us to wonder how one can analyze these metaphors. Drawing on Pilkington’s notions of the strength and range of the mapping, we describe the choices a reader has to make when interpreting metaphors.2 Having discussed these theoretical issues, we introduce several general principles that advance the practical study of metaphor in exegesis. Finally, as exam ple is better than precept, we demonstrate our principles in an analysis of the metaphors in Job 6:14–21.

Conceptual Metaphor Theory

Metaphor is a central theme in Cognitive Linguistics. In Metaphors We Live By, George Lakoff and Mark Johnson strongly react against the folk belief that metaphor is found mainly, primarily or most naturally in poetic language. Metaphor, they claim, is not merely a matter of language; it is a matter of thought: [M]etaphor is typically viewed as characteristic of language alone, a matter of words rather than thought or action. For this reason, most people think they can get along perfectly well without metaphor. We have found, on the contrary, that metaphor is pervasive in everyday life, not just in language but in thought and action. Our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature.3 In their iconic book, Lakoff and Johnson introduce the notion of “con ceptual metaphor.” CMT (Conceptual Metaphor Theory) claims we understand one concept or experience in terms of another—often more concrete—concept or experience.4 Conceptual metaphor is thus a cogni tive phenomenon. It involves two conceptual domains: the source domain is the concept that helps to understand a more abstract concept, called the target domain. Our knowledge about warfare, for instance, helps us to conceptualize argumentation and debate. The existence of linguistic

1 Gerard Steen, “The Contemporary Theory of Metaphor—Now New and Improved,” RCL 9,1 (2011): 26–64. 2 Adrian Pilkington, Poetic Effects: A Relevance Theory Perspective (Amsterdam/Philadel phia: John Benjamins, 2000). 3 George Lakoff and Mark Johnson, Metaphors We Live By (Chicago: University of Chi cago Press, 1980), 3. 4 “The essence of metaphor is understanding and experiencing one kind of thing in terms of another.” Ibid., 5. selecting and analyzing metaphors in the hebrew bible 41 expressions like ‘he had to defend himself,’ ‘he attacked her on several points of her presentation,’ and ‘your claims are indefensible’ can be explained by the reality of the conceptual metaphor argument is war.5 Important to know is that we comprehend the target domain by select ing some elements that are part of the source domain whilst ignoring others. Choosing which set of features can be applied to the target domain is called mapping. Mapping is both selective and unidirectional, viz. we choose to map some, but not all elements of the source domain and the mapping goes from source to target domain and not vice versa. For argu ment is war, this can schematically be represented as follows.

SOURCE TARGET Enemies → Discourse partners Weapons → Words Strongholds → Positions Fight → Debate Peace → Agreement War industry ? Time investment? Spies ? Observers? Victory ? Winning the argument?

The above innovations in metaphor theory greatly affect which linguis tic expressions exegetes consider to be metaphorical. Lakoff and Johnson argue convincingly and with many examples that conceptual metaphor is ubiquitous. It is attested in all language use, in all languages and across all genres.6 Nearly all everyday expressions use conceptual metaphor. As such, Lakoff and Johnson break down the hegemony of the literary when it comes to the use of metaphor. The latter is no longer restricted primar ily to literary texts; it is pervasive. Now, how do these findings influence the analysis of metaphors in lit erary texts? Can we methodically make use of CMT in literary analysis? Two issues force us to search for additional theoretical insights. First, we have a long-standing tradition that recognizes the special character

5 Cognitive linguists signal both concepts and conceptual metaphors with small capitals. 6 Cf. Gerard Steen, “The Contemporary Theory of Metaphor,” 50. Steen’s observations confirm the ubiquity of metaphors, as he notes that by average 13,6 percent of natural language is metaphorical. 42 johan de joode and hanneke van loon of metaphor in literary texts. As Elena Semino and Gerard Steen put it, “most scholars seem to agree that the metaphorical expressions typically found in literature are more creative, novel, original, striking, rich, inter esting, complex, difficult, and interpretable than those we are likely to come across in non-literary texts.”7 Second, we have to deal with a very practical point. Considering the distribution of conceptual metaphors, one might wonder whether the literary scholar that pays attention to all conceptual metaphors in a text does not end up with too much data to analyze. Are we to investigate all linguistic expressions of all conceptual metaphors found in a text? Are they all equally significant for the interpre tation of that text? In what follows, we propose a theoretical framework that takes into account these issues as it allows exegetes to first select and subsequently analyze metaphors that are potentially relevant for literary analysis.8 The approach suggested below is but one of many possible modi operandi, yet it might be a particularly fruitful one for scholars interested in the literary.

Deliberately Used Metaphors

Before analyzing a text’s conceptual metaphors, we have to select them. For practical reasons—mainly the pervasiveness of metaphor—exegetes might want to narrow down their corpus of analysis. If they were to inves tigate all conceptual metaphors in a biblical text, this would to lead to a very large data-set. In fact, the time spent studying these data, might not be in proportion to the intended outcome. The present authors argue that, for literary critics, some metaphors in a text might be more reward

7 Elena Semino & Gerard Steen, “Metaphor in Literature,” in The Cambridge Handbook of Metaphor and Thought, ed. Raymond W. Gibbs (New York: Cambridge University Press, 2008), 232–246 (233). 8 In biblical exegesis the findings and hypotheses of Cognitive Linguistics have been used by several exegetes and interpreters have in various ways demonstrated the exegetical relevance of the recognition of the conceptual nature of metaphor in the Hebrew Bible and its cognate languages. To name just a few: Sarah J. Dille, Mixing Metaphors: God as Mother and Father in Deutero-Isaiah (JSOT.SS 398; London/New York: T&T Clark Interna tional, 2004); Andrea L. Weiss, Figurative Language in Biblical Prose Narrative: Metaphor in the Book of Samuel (Leiden/ Boston: Brill, 2006); Ellen van Wolde, Reframing Biblical Studies: When Language and Text Meet Culture, Cognition, and Context (Winona Lake, IN: Eisenbrauns, 2009); Job Y. Jindo, Biblical Metaphor Reconsidered: A Cognitive Approach to Poetic Prophecy in Jeremiah 1–24 (Winona Lake, IN: Eisenbrauns, 2010); Pierre Van Hecke, From Linguistics to Hermeneutics: A Functional and Cognitive Approach to Job 12–14 (SSN 55; Leiden/Boston: Brill, 2011). selecting and analyzing metaphors in the hebrew bible 43 ing to study than others. To establish a sub-corpus of metaphors, Steen’s recent proposal on the communicative dimension and metaphorical pro cessing is of assistance.9 Steen introduces the difference between deliberately and non- deliberately used metaphors.10 As he puts it, [d]eliberate metaphor is an overt invitation on the part of the sender for the addressee to step outside the dominant target domain of the discourse and look at it from an alien source domain.11 Some of the conceptual metaphors in a text explicitly ‘ask for our atten tion.’ Now, for exegetes who systematically want to study metaphors in a text, Steen’s pragmatic approach is methodologically interesting: it allows examining not all the metaphors in a text, but primarily the deliberately used ones. Whereas CMT convinced a great many scholars of the ubiq uity of metaphor, few of these metaphors explicitly force the reader to “step outside of the target domain” in order to investigate “an alien source domain.” Indeed, in sharp contrast to the ubiquity of metaphor in gen eral, deliberately used metaphors are characterized by their paucity. Steen explains the latter by referring to the effort they ask of the reader: We cannot keep changing our perspective all the time as we need to stay focused on one conceptual target domain at a time, the topic. Nor can we keep making comparisons or cross-domain mappings all the time, for they cost additional energy.12 Now, how can an author draw the reader’s attention to an alien source domain? There are essentially two ways of doing so: one can use novel metaphor or one can signal the source domain linguistically.13 Steen

9 The present authors do not envision to distinguish between dead or conventional metaphors (metaphors of which the metaphorical meaning is described in the dictionary), on the one hand, and creative metaphors, on the other hand; instead, we suggest with Steen that the linguistic expression of a metaphor in a specific text determines whether the metaphor should be included in the sub-corpus. 10 The notion of deliberately used metaphor touches the theoretical question of meta phorical processing. Some scholars argue that not all metaphorical expressions are neces sarily processed as metaphors. In this line of thought, metaphors that are conventional and not linguistically marked, that is metaphors that are not deliberately used, are understood without the brain having to compare two conceptual domains, whereas novel metaphors and linguistically marked metaphors are metaphorically processed, cf. Brian F. Bowdle & Dedre Gentner, “The Career of Metaphor,” Psychol Rev 112 (2005): 193–216. 11 Steen, “The Contemporary Theory of Metaphor,” 37. 12 Ibid., 38. 13 Ibid., 39. 44 johan de joode and hanneke van loon mentions the following means to linguistically signal a deliberately used metaphor: the appearance of lexical markers (for instance, like), the exten sion of a metaphor beyond one phrase or clause, and the use of extended comparisons and analogies.14 A more complete inventory of possible lin guistic signals can be found in Goatly’s The Language of Metaphors.15 All of the above illustrates the importance of the communicative dimension of metaphor: an author can direct the reader’s attention to certain, spe cific metaphors. Steen’s suggestions enable the literary scholar to select precisely these metaphors.

The Range and Strength of Mappings As argued above, deliberate use is a criterion to select metaphors that shift the reader’s attention to the source domain. While Steen’s findings help us to select metaphors (heuristics), we shall now turn to the interpretation of metaphors (analytics). Exegetes need to know how one can systemati cally interpret the different meanings that deliberately used metaphors evoke. To explain the effect of metaphors, we combine Steen’s findings with those of Adrian Pilkington. To explain the poetic effect of metaphors, Pilkington suggests we inves tigate metaphors in terms of ‘range’ and ‘strength.’ He wonders what dis tinguishes “relatively conventional” from “relatively creative” metaphors and proposes that the two types “vary in terms of the range and strength” of their selective mappings.16 His central idea is useful for CMT: meta phors can vary a) in the number of different correspondences the reader can find (the range of the mapping), and b) in how self-evident the selec

14 “Some linguistic forms typically indicate deliberate metaphor use, including a lexical signal such as like, or the extension of a metaphor beyond one phrase or clause. Extended comparisons and analogies between parts of texts are also deliberate, requiring discourse- strategic as opposed to lexico-grammatical decisions on the part of their producer. Instances of word-play that rest on a contrast between a metaphorical and a non-metaphorical word sense, or other combinations between metaphor and different tropes, such as hyperbole, irony, and so on, all seem to require relatively conscious rhetorical planning. Even entire genres (allegory, parable, poetry) may count as including expected instructions to take any or most metaphors in their texts as deliberate, constituting well-known exceptions that allow for high density and processing of deliberate metaphor.” Ibid., 41–42. 15 Andrew Goatly, The Language of Metaphors (London/New York: Routledge, 1997). For linguistic signals in Hebrew, cf. Daniel Bourguet, Des métaphores de Jérémie (Paris: Gabalda, 1987), 23–44. This is nicely summarized in English by Brian Doyle, The Apocalypse of Isaiah Metaphorically Speaking: A Study of Use, Function and Significance of Metaphors in Isaiah 24–27 (BEThL 151; Leuven: Leuven University Press, 2000), 77–91, 123–128, 139–144. 16 Pilkington, Poetic Effects, 100. In Pilkington’s words, conventional and creative meta phors “vary in terms of the range and strength of the assumptions that they make more salient.” We have adapted his terminology in order for it to be in line with CMT. selecting and analyzing metaphors in the hebrew bible 45 tion of the correspondences is (the strength of the mapping). Whereas most metaphors have a small range of very dominant mappings, some metaphors have a broad range of equivalent mappings.17 An example, though admittedly trivial, might illustrate that point. If we say that a text contains a lot of information, the reader has few choices to make and the exact choice is quite evident: the meaning of the verb to contain refers to physical containment and one can also use it in relation to more abstract entities as, for instance, texts.18 If, however, we say in her text meaning hides behind many corners, the reader has more choices to make and less obvious ones. What are the corners referring to? Why would there be many? Is meaning hidden by the author? Or does meaning hide itself? If so, why would it hide itself? Should the reader reveal what is hidden? In other words, the range of the mapping is broad, viz. the reader can explore many correspondences, and the precise correspondences that the reader ought to map are less evident than in the containment example. Following Pilkington, the connection between the source and target domain determines the range and strength of the mapping; if this con nection is a) not well-established or b) not easy to achieve, there will be numerous and more intricate mappings. If the connection is long-familiar or easy to achieve, however, there will be fewer and easier mappings. In that case, the reader’s work is harder, but his/her reward is higher. We propose to integrate Pilkington’s theoretical insights in the study of delib erately used metaphors by systematically analyzing the range and strength of mappings.

17 Pilkington illustrates his theory with a comment on an example given by Sperber and Wilson: “The example of poetic metaphor used by Sperber and Wilson is the following remark made by Flaubert of the poet Leconte de Lisle (1995: 237): Son encre est pale. ‘His ink is pale.’ Here there are no strong assumptions to the truth of which Flaubert can be said to have committed himself. [. . .] One might derive implicatures such as: “Leconte de Lisle’s writing is weak, lacks contrast, may fade, is sickly, will not last; Leconte de Lisle does not put his whole heart into his work. There are an indefinite number of further implica tures one could add to the list [. . .], and there is no cut-off point that allows one to say that so many implicatures are communicated and no more. It is the range and the indetermi nacy of the implicatures which gives the metaphor its poetic force. These factors explain why it is that metaphors, especially poetic metaphors, can never be adequately translated or paraphrased. They also explain why their interpretation may differ across individuals and be subject to debate amongst literary critics” (ibid., 101). As such, Pilkington explains why some metaphors in literary texts are considered more literary and elevated than oth ers. See also the quotation of Semino and Steen above. 18 This example is part of the conduit metaphor, cf. Michael J. Reddy, “The Conduit Metaphor: A Case of Frame Conflict in Our Language about Language,” in Metaphor and Thought: Second Edition, ed. Andrew Ortony (Cambridge: Cambridge University Press, 1993), 164–201. 46 johan de joode and hanneke van loon

An Example

We propound a step-by-step procedure that has explicitness as its central goal: we want to be as explicit as possible about the selection and the analysis of metaphors. Our first step is to find markers that signal deliber ately used metaphors; the second step is to be explicit about the precise points of comparison between the source and target domain. We apply this method to the metaphors that can be found in Job 6:14–21: 14 לַּמָ֣ סמֵרֵעֵ ֣הּוחָ ֑סֶ ד ֖ וְיִרְאַ ת ֣יׁשַּדַ יַעֲ זֽ ֹוב׃ 15 אַ֭חַ י ּבָ גְד֣ ּוכְמֹו־נָ ֑חַ ל ֖ ּכַאֲפִ יק נְחָלִ ֣ים יַעֲ בֹֽ רּו׃ 16 הַ ּקֹדְרִ֥ ים מִ ּנִי־קָ ֑רַ ח עָ֜ לֵ֗ ימֹו יִתְ עַּלֶם־ׁשָ ֽלֶ ג׃ 17 ּבְ֭עֵת יְזֹרְ ב֣ ּונִצְמָ ֑ תּוּבְ֜חֻּמ֗ ֹו נִדְ עֲכ֥ ּומִּמְקֹומָ ֽ ם׃ 18 יִּ֭לָ֣פְ תּואָרְ ח֣ ֹות ּדַרְ ּכָ ֑םיַעֲל֖ ּובַ ּתֹ֣ הּו וְ ֽ יֹאבֵ דּו׃ 19 הִּ֭בִ יטּואָרְ ח֣ ֹותּתֵמָ א֑ הֲ לִ יכֹ֥ ת ׁשְ֜ בָ֗ א ֽ קִּוּו־לָ מֹו׃ 20 ּבֹׁ֥שּו ּכִֽי־בָטָ ֑ ח ּבָ ֥ אּו עָ֜דֶ֗יהָ ֽ וַּיֶחְּפָ רּו׃ 21 ּכִֽי־עַּ֭תָ ה הֱ יִ ֣יתֶ ם ֹ֑ ל א ּתִֽרְ א֥ ּו תחֲ֜תַ֗ ֽ וַּתִירָ אּו׃ 14 a friend owes loyalty to one who fails, or he forsakes the fear of the Almighty; 15 My comrades are fickle, like a wadi, like a bed on which streams once ran. 16 they are dark with ice; Snow obscures them; 17 but when they thaw, they vanish; in the heat, they disappear where they are. 18 their course twists and turns; they run into the desert and perish. 19 caravans from Tema look to them; processions from Sheba count on them. 20 they are disappointed in their hopes; when they reach the place, they stand aghast. 21 So you are as nothing: at the sight of misfortune, you take fright.19 In chapter 6, Job reacts on Eliphaz’s first intervention in chapters 4 and 5. Job 6:14–21 is structured as follows: verse 14 introduces the theme of the passage; verses 15 to 17 and 18 to 20 are a comparison, and verse 21 explains what the comparison intends to assert.20 The general picture kindness” from his friends, but they are“ חסד is quite clear: Job expects

19 This translation is taken from JPS (1985), except for the words in italics. 20 Cf. David Clines, Job 1–20 (WBC 17; Dallas, TX: Word Books, 1989), 176: “The structure of this section of the speech is in some respects plain: vv 14–21 are one unit, revolving about the image of the seasonal wadi, not directly addressed to the friends, beginning and selecting and analyzing metaphors in the hebrew bible 47

The rivers that Job refers to are temporary rivers .נחל as unreliable as a that are only filled with water in winter or after heavy rain fall.21 With an extended, elaborated metaphor, Job describes in what sense the behavior of the river can be juxtaposed to the conduct of his friends. The point of comparison is given in verse 21: the friends get frightened at the sight at the sight of (ירא) of disaster. This verse hints at verse 14: by taking fright disaster (6:21), they neglect to show kindness and thus forsake the fear of the Almighty (6:14).22 (יראה) In verse 15, the text’s metaphor is signaled by the lexical markers like.” In the succeeding verses, two more linguistic“ ,כ like,” and“ ,כמו means classify the metaphor as deliberately used. Firstly, the metaphor ”.rivers“ ,נחלים is extended up to verse 18 with repeated references to the the ones that“ ,הקדרים This noun functions as the subject of the participle ”;disappear“ ,נצמתו ”;dry up“ ,יזרבו are dark” (verse 16) and the finite verbs be extinguished” (verse 17). It is referred to by means of the“ ,נדעכו and from“ ,ממקומם ;(in them” (verse 16“ ,עלימו suffix third masculine plural on their ways” (verse 18). Secondly, the“ ,דרכם their place” (verse 17); and ,ארחות תמא metaphor is amplified by the introduction of the new topics .travelers from Sheba” in verse 19“ ,הליכת ׁשבא caravans from Tema,” and“ As the new topic is also the subject of the verbs in verse 20, this verse too is part of the description of the source domain. Before we can make the correspondences between the target and the source domain as explicit as possible, we must know where the text refers to the target domain and where it describes the source domain. The tar get domain is referred to in verses 14, 15, and 21: the focus is on Job and his ‘brothers;’ the topic of discussion is the kindness of Job’s brothers (verse 14) and how they take fright when they see calamity (verse 21). ending with distinctive couplets (vv 14, 21) and containing as a center two clearly defined triplets (vv 15–17, 18–20).” in New International Dictionary of Old Testament Theol ”, ָ נ הָ ר“ ,See Allen P. Ross 21 ogy and Exegesis, ed. Willem A. VanGemeren (Grand Rapids, MN: Zondervan Publishing applies most readily to a temporary river that נחל House, 1997), 46–51 (47): “The word flows with great force in the winter or rainy season but leaves only dry channels or deep ravines in the summer. Thus, the word can refer to either a fast flowing stream or torrent, or to the dry river bed.” 22 Cf. John E. Hartley, The Book of Job (Grand Rapids, MN: Eerdmans, 1988); Clines, Job. 1–20. For interpretations that understand Job as the one who forsakes the fear of the Almighty, see Samuel Rolles Driver & George Buchanan Gray, A Critical and Exegetical Commentary on the Book of Job Together with a New Translation (ICC; Edinburgh: T & T Clark, 1921); Marvin H. Pope, Job: Introduction, Translation and Notes (AB 15; Garden City, NY: Doubleday [1965] 19863); Norman C. Habel, The Book of Job. A Commentary (OTL; Phila delphia, PA: Westminster Press, 1985). 48 johan de joode and hanneke van loon

The source domain is described in verses 15–20: the participants are riv ers and caravans, and the topic of discussion is the behavior of the river. Now, what correspondences can be found between, on the one hand, the story of the rivers and the caravans, and on the other hand, the unreliable brothers that do not show kindness, but take fright? If the information given about the target domain is compared with the information that is given about the source domain, it can be established that a) the rivers correspond to the friends, b) the lack of water corresponds to the lack of kindness, and c) the hot circumstances correspond to the sight of calam ity. Just as temporary rivers lack water when it gets hot, the friends lack kindness in the face of trouble.23 In 6:19–20, the caravans correspond to Job. They are disappointed in the rivers they counted on. Although the target domain does not mention how Job feels about the behavior of his brothers, the comparison establishes a mapping between the feelings of the caravans and the feelings of Job. Since the source domain is more elaborate than the target domain, the comparison adds new information to the target domain. In fact, the deliberately used metaphor invites the reader to evaluate for every statement in the source domain whether it has a counterpart in the target domain.24 For example, if the rivers correspond to the brothers and the paths of the rivers are said to twist and end up in waste, what does this say about the brothers? The friends do not literally twist their ways. The reader can construct a mapping in order to explain the metaphor. Here, it seems the friends have steered away from the right path and now continue in a direction that leads nowhere.25 Does Job here accuse the

23 Commentaries generally agree that the rivers’ lack of water corresponds to the friends’ lack of kindness; what the heat refers to is often explained more implicitly. Some commentaries interpret the time of heat as the time of need, cf. Driver & Gray, A Critical and Exegetical Commentary on the Book of Job, 65: “Job’s friends now fail (disappoint) him in the hour of need, like such dried-up wādys;” Pope, Job, 54: “Job’s friends are like the wadis which overflow in the spring thaw, turbid with ice, snow, and silt sediment, but when the need is greatest, in the heat of summer, they fail”; Clines, Job 1–20, 178: “The wadis overflow when their water is not needed; when it is needed they have nothing to offer. So it is with Job’s friends and their loyalty.” As the heat causes the water to disap pear, we argue that it better fits the context that the heat refers to the sight of trouble. So also Hartley, The Book of Job, 138: “This analogy claims that the friends overflow with loyal kindness during the good times, but when the heat of trials comes, they dry up; they turn out to be undependable.” 24 It could be argued that not every description in the source domain needs to have an exact correspondence in the target domain. Deliberately used metaphors, however, do trigger the reader to find as many correspondences as possible. 25 Cf. Habel, The Book of Job, 149: “Job cannot discern his own course (derek, 3:23) but [his friends’] twisting trek into the wasteland is obvious (verse 18). They dwindle away selecting and analyzing metaphors in the hebrew bible 49 brothers of a bad attitude, of a logic so incorrect that it will lead to immo rality? One could also suggest that turning away corresponds to the move ment of forsaking, or literally, leaving of the fear of the Almighty (verse 14). This correspondence would extend the comparison as follows: like the temporary rivers lack water when it gets hot, the friends lack kindness at the sight of disaster; like the lack of water results in wadis that turn to go up into waste, the lack of kindness results in the friends leaving the fear of the Almighty. Whereas the caravans are disappointed about the disappearance of the river, Job is dismayed about his friends not having succeeded in keeping the fear of the Lord. Another example of can be found in verse 16 where darkness, ice, and snow can be taken literally. When snow, however, is said to hide itself, that is yet again a metaphorical expression. The reader has to interpret this metaphor twice: in relation to the source domain (what is the mean ing of snow that hides itself?) and in relation to the target domain (how does hidden snow fit into Job’s refutation of the friends?). Likewise, in the source domain the image of a river filled with ice and snow has positive associations (it promises an abundance of water). If the characterization of the water, however, corresponds to the characterization of the friends’ kindness, the terms ‘dark’ and ‘concealing’ evoke negative connotations. By searching for correspondences between the source and target domain, the main argument of the comparison in verses 15–20 can be established: Job accuses his friends to be unreliable because they withhold their kindness at times of trouble, and he warns them that this will result in that they forsake the fear of the Almighty. While the main line is clear, many elements in the metaphor remain hard to grasp. The above example is not exhaustive, but it does illustrate our point.26 When exegetes deal with deliberately used metaphors in literary texts,

and ‘perish’ (ʾbd) like the wicked whom Eliphaz had introduced (4:9).” Commentators that paths,” contend that in 6:18 the caravans“ אָרְ חֹות caravans,” and not“ אֹרְ חֹות choose to read twist from their ways in search for water, but perish because of thirst, cf. Driver and Gray, A Critical and Exegetical Commentary on the Book of Job, 64; Hartley, The Book of Job, 138; Clines, Job. 1–20, 179. In this reading, the lack of kindness results in Job leaving his way to end up without orientation. We think that this verse refers to the behavior of the paths of the rivers, that is, the behavior of the friends (and not that of Job): there is nothing to sug gest that Job’s disorientation is caused by a lack of kindness on the part of his friends. 26 In the above, we have used the theorems of CMT to describe and analyze possible mappings. We have not dealt with the subsidiary conceptual metaphors that underlie the mapping as such. Here are a few examples (though it would lead us too far to describe the weak/melting one”), affection“ ,מס .them in detail): emotional state is solidity (cf ,(”in the heat“ ,בחמו .ice” and “snow”), trouble is heat (cf“ ,ׁשלג and קרח .is warmth (cf .(”to twist“ ,לפת .being righteous is being straight (cf 50 johan de joode and hanneke van loon their research should make explicit what similarities between the source and target domains can be found. It is tempting to map elements from the source domain implicitly (for that matter, in everyday life we do so continuously), yet the more explicit a mapping is made, the more criti cal fellow scholars can evaluate it. As exegetes, we should be aware of the decisions we (have to) take when selecting and analyzing deliberately used metaphors.

Deliberately Used Metaphor and Hermeneutics

Paying attention to the deliberate use of metaphors has several method ological benefits, a few of which we want to highlight. First, the notion of ‘deliberately used metaphor’ joins at the hips the issues of authorial intent, on the one hand, and reader response, on the other. By using meta phors deliberately, the author can force the reader to reflect on them, try and process them. Thus, he triggers reflection on the part of the reader (cf. note 10). Which elements should he/she map? What is the author aim ing at? The reader thus constructs a mapping based on a wide range of equivalent correspondences and this process is intentionally incited by the author. Furthermore, deliberately used metaphors constitute a natural sub- corpus. Ideally, exegetes study all the conceptual metaphors in a text, yet in view of time restrictions this is not always possible. When one is forced to make a selection of the metaphors to be studied, deliberateness allows the exegete to narrow down his/her sub-corpus based on criteria that are metaphor-related (and not criteria related to historical, thematic or literary features). Though these metaphors are not the only metaphors we can and should study, they might be the most obvious ones to start with. Finally, the author’s invitation to the reader to investigate the source domain is also likely to enhance his/her attention for the literary features of his text. As such, the notion of deliberateness easily connects literary and conceptual studies. Nonetheless, before we conclude, a few disadvantages of our pro posal should be mentioned. The main objection one could raise against the construction of a sub-corpus based on deliberateness, is that non- deliberately used metaphors might reveal as much—if not more—about the message the author wants to get across, and the mental world he attempts to sketch. For example, Job regularly refers to the bitterness of cf. 3:20; 7:11; 10:1). This metaphor is never as such used ;מר נפׁש) his soul selecting and analyzing metaphors in the hebrew bible 51 deliberately in the book (the reader’s attention is not shifted to the source domain), but it is an important characteristic of the author’s depiction of Job. Though one single linguistic expression does not force the reader to step out of the target domain, a set of linguistic expressions spread over a text, might incite him/her to do so. Indeed, what he deliberately communicates is what he controls, but it might also interest us what he communicates with less control. One could likewise object that the author has—before writing—constructed a mental image of the pro tagonist, its narrative world and its relationship with other characters. Exegetes can thus also choose to study all conceptual metaphors (pref erably the non-deliberately used ones) in an attempt to reconstruct the author’s implicit conceptualizations. Furthermore, one could raise a theoretical objection: the degree of deliberateness is evaluated by the effect the text has on the reader (when does his/her attention shift to the source domain?). There is hardly any other way of evaluating the deliberateness of the author’s use of meta phors. Depending on the present-day reader, however, might be shaky ground. The author can use strategies other than interpretative conflict in order to purposefully communicate his message using metaphor and the reader might not spot these metaphors (the author can, for instance, combine metaphors across a text or use them at strategic places in his text). In what sense is that set of images not deliberately used and are we thus prone to find it in many details and characterizations throughout the story? These objections illustrate that deliberately used metaphor is not the only phenomenon worthy of study. In light of the present volume’s insis tence on the literary, however, we do think that the study of deliberately used metaphor is accessible to literary scholars. Though all conceptual metaphors are worthy of our attention, the deliberately used ones might be the first ones to deserve it. If one thus has to make a choice about the metaphors one should study, the deliberately used ones provide a natural sub-group that can be explicitly analyzed with attention for the range and strength of mappings.

Conclusion

Literary studies have often considered metaphor to be a matter of style, yet Cognitive Linguistics has proven the ubiquity of conceptual meta phor. Though conceptual metaphor can explain metaphorical linguistic 52 johan de joode and hanneke van loon expressions in isolation, it is not by definition equipped to deal with con ceptual metaphor in literary texts. In this contribution, we have therefore evaluated whether all metaphors in a literary text are equally signaled. We have done so first from a theoretical point of view. Are there concep tual metaphors in poetry that ask for more attention in literary analysis than others? With Steen, we have argued for the efficacy of the notion of deliberate use. The deliberate use of a metaphor shifts the attention from the target to the source domain. It is signaled and it requires the reader to make an active choice about how to understand the metaphor and its literary function. Pilkington’s notion of the range and the strength of the mapping explain what the reader is confronted with: few, and dominant elements are part of the mapping of a non-deliberately used metaphor and many, equivalent mappings are inherent to deliberately used metaphor. We have also put these theories to practice by formulating some prac tical guidelines for metaphor analysis and by testing them. Quintessen tially, we advocate for explicitness about the source and target domain and in particular about the chosen mapping. We have applied this motto to Job 6:14–21 where clearly (a) a metaphor was used deliberately and (b) the range and strength of its mapping proved to be respectively great and equivalent. A systematic analysis of deliberate metaphor thus provides more insight into the choices a text forces upon the exegete. In terms of hermeneutics, these two findings are significant as they join at the hips the ways in which meaning comes to the fore: the author uses specific conceptual metaphors deliberately; the reader processes these intensively. Hence, we hope that the present proposal fills a gap in cognitive linguistic and literary studies of poetic metaphor in the Hebrew Bible. Literary Theory and Composition History of the Torah: The Sea Crossing (Exod 14:1–31) as a Test Case1

Angela Roskop Erisman

Abstract

This essay argues that literary theory is not just a tool for synchronic studies of biblical literature but that it is essential, if we are to understand its diachronic development and its rootedness in ancient culture. A new reading of the sea crossing narrative in Exodus 14 informed by German reception theory and linguistic pragmatics (relevance theory) suggests a very different understanding of this text’s composition history than that offered by tradition-historical and source- critical methods, and helps us better understand its connections to ancient Near Eastern culture as well as literature elsewhere in Tanakh.

Literary theory became popular in biblical studies during the 1980s, as scholars became increasingly less convinced by the yield of classic historical-critical method, which seemed to focus more on deconstructing a narrative based on stylistic differences than explaining how it works as a whole.2 Scholars employed the formalist theories in vogue at the time, demonstrating that historical critics can be too quick to jump in with a diachronic solution to what may be a synchronic problem. But the reac- tionary character of this early push for literary theory in biblical studies also led scholars to downplay the presence of fractures in the text which may be markers of composition history. The distance between modern literary criticism and classic historical criticism (or Literarkritik) is actually not as vast as it seems, because concern for the literary integrity of a text is at the center of both enterprises.3 As Rolf Knierim pointed out, “the

1 I would like to thank the Department of Theology and the College of Arts and Sci ences at Xavier University (Cincinnati, OH) for their support of my participation in this conference. 2 Brian Peckham, “Writing and Editing,” in Fortunate the Eyes that See: Essays in Honor of David Noel Freedman in Celebration of His Seventieth Birthday, ed. Astrid Beck, et al. (Grand Rapids, MI: Eerdmans, 1995), 364–383 (364). 3 John Barton, “Historical Criticism and Literary Interpretation: Is There Any Common Ground?” in Crossing the Boundaries: Essays in Biblical Interpretation in Honour of Michael D. Goulder, ed. Stanley E. Porter, Paul M. Joyce & David E. Orton (BIS 8; Leiden: Brill, 1994), 3–16 (7). This is easily forgotten when it comes to historical-critical methods, but see 54 angela roskop erisman determination of sources is a possible result of literary-critical work [Lit erarkritik] but not its methodological principle.”4 Setting historical and literary criticism at odds and practicing them in isolation thus obscures the fact that they have essentially the same goal. The question is not whether to use literary theory, but which tools best help us understand the literary integrity of a text as well as identify and explain places where it truly lacks integrity. Formalist theory did biblical studies an enormous service by teaching us to read with an eye to literary concerns we had not considered before, but its ahistorical sensibility does not enable us to deal well with issues like the composite nature of biblical literature or its rootedness in ancient culture. Other literary theories do, and I would like to explore the potential yield of a few literary-critical tools in the following study of the sea crossing narrative. Exodus 14 is a transitional narrative in which the Israelites move from enslavement in Egypt to the freedom (and danger) of the wilderness by crossing the sea on dry ground. A crux of mid-twentieth century scholarship on this text is the question of whether it belongs to an exodus tradition or a wilderness tradition, each of which was understood to have developed orally in a different cultic context before it took shape as literature.5 Martin Noth stated confidently that the sea crossing is the heart of the exodus tradition, but George W. Coats noted various elements that link Exodus 14 with the wilderness narrative and argued that it belongs instead with the wilderness tradition. Brevard Childs resolved this conundrum using the Documentary Hypothesis: The J version of the narrative is associated with the wilderness (Pesah was its exodus event), while the P version identifies the sea crossing as the paradigmatic exodus event, and

Norman Habel, Literary Criticism of the Old Testament (Philadelphia, PA: Fortress Press, 1971), 2–7. 4 Rolf Knierim, “Criticism of Literary Features, Form, Tradition, and Redaction,” in The Hebrew Bible and its Modern Interpreters, ed. Douglas A. Knight and Gene M. Tucker (Min neapolis, MN: Fortress Press, 1985), 123–165 (131). Until the Graf-Wellhausen documentary hypothesis became the paradigm, fragmentary and supplementary hypotheses circulated alongside it and, in fact, still do; e.g., Christoph Levin, “Source Criticism: The Miracle at the Sea,” in Method Matters: Essays on the Interpretation of the Hebrew Bible in Honor of David L. Petersen, ed. Joel M. LeMon & Kent Harold Richards (Atlanta, GA: Society of Biblical Literature, 2009), 39–61. 5 The classic articulation of tradition-historical method is Martin Noth, Überlieferungs geschichte des Pentateuch (Stuttgart: Kohlhammer, 1948); E.T. A History of Pentateuchal Tra ditions (Englewood Cliffs, NJ: Prentice Hall, 1972). literary theory and composition history of the torah 55 the lack of clarity about which tradition Exodus 14 should be associated with arose as the two versions were combined.6 The tradition-historical method yielded some important insights about this text, not least of which is that there appear to be two different end- ings to the exodus narrative, one associated with Pesah and the other associated with the sea crossing. But this approach to the issue of why it contains both exodus and wilderness elements is problematic. Dale Patrick questioned whether the sea crossing narrative could ever have been associated with the wilderness tradition alone, because the story is by its very nature the culmination of Israel’s struggle to get out of Egypt. Patrick himself worked squarely within the tradition-historical method, but he offered a key insight that transcends it, namely that the complaint motif was a “freefloating compositional theme” understandably used in a narrative set at the edge of the wilderness.7 Using Wolfgang Iser’s framing of how literature is read and the implications for how it is written, which he articulated in The Act of Reading: A Theory of Aesthetic Response, we might think of the complaint motif as an element of cultural repertoire. Cultural repertoire is anything in a culture that can be referred to in a text: other texts, geography, social phenomena, history, ideas, motifs. When an author constructs a text, he selects elements of repertoire from various background contexts ideally known to both him and the reader, and blends them together in order to develop the narrative.8 Tradition-historical method thus presented a false choice in asking us to decide whether the sea crossing belongs to the exodus or the wilderness, and source criticism played into the hands of this way of framing the issue. Instead, we might consider the transitional character of this narrative and ask: What elements of cultural repertoire

6 Martin Noth, Exodus: A Commentary (Philadelphia, PA: Westminster Press, 1962), 104–105; George W. Coats, “The Traditio-Historical Character of the Reed Sea Motif,” VT 17 (1967): 253–265; Brevard S. Childs, “A Traditio-Historical Study of the Reed Sea Tradition,” VT 20 (1970): 406–418; George W. Coats, “The Sea Tradition in the Wilderness Theme: A Review,” JSOT 12 (1979): 2–8. Childs’ view continues to be accepted in recent commentaries; e.g., Thomas B. Dozeman, Exodus (Grand Rapids, MI: Eerdmans, 2009), 182–184, 199–200, 300. 7 Dale Patrick, “Traditio-History of the Reed Sea Account,” VT 26 (1976): 248–249. See David H. Aaron, Etched in Stone: The Emergence of the Decalogue (New York: T & T Clark, 2006), 48–52, for a theoretically grounded discussion of floating motifs as elements of cul tural repertoire. 8 Wolfgang Iser, The Act of Reading: A Theory of Aesthetic Response (Baltimore, MD: The Johns Hopkins University Press, 1978). 56 angela roskop erisman are used here, and from what backgrounds were they drawn? What literary goals were they blended together to serve? And how (and how well) do they work together to achieve these goals? When we consider Exodus 14 from this literary-critical angle, we will see that it is not a combination of two independent sources, each of which construed the sea crossing in the context of a different tradition complex, but that much of it is actually a tightly-constructed, purposeful whole. The wilderness complaint motif involves the Israelites complaining in a situation of distress—typically one at home in the wilderness such as lack of food or water—followed by a response from Moses, divine intercession, and a miracle.9 The complaint in Exod 14:10–12 comes when the Israelites, on their way out of Egypt, notice the Egyptian army in hot pursuit. This is not a typical wilderness situation, but the motif is nonetheless used to develop one of the main themes of the narrative. The Israelites’ complaint is prompted by their fear of the pursuing Egyptians (Exod 14:10), yet by the end of the narrative, they have learned to fear YHWH instead (Exod 14:31). Moses’ response in Exod 14:13–14 facilitates this shift: He does not merely calm the Israelites’ fear of being overtaken by the Egyptians but also exhorts them to be in awe of YHWH’s power to save them. The verse 13) and the idea that YHWH will battle on) אל תיראו expression behalf of the Israelites against their enemies (verse 14) are drawn from the background of Deuteronomistic war ideology, where such exhorta- tions appear frequently.10 Blended with the complaint motif, this element of repertoire serves as the pivot on which the theme of fearing YHWH is developed. It also ties the complaint (verses 10–12) and response (verses 13–14) elements of the motif with the dénouement of the narrative (verse 31) in a tightly knit whole, one goal of which is to get the Israelites to feel the right kind of fear (awe) of the right individual (YHWH). Historical critics sometimes lose sight of the core goal of Literarkritik in their effort to apply a particular model of composition history. The com-

9 For definition of the elements of this motif, also referred to as the murmuring motif, see George W. Coats, Rebellion in the Wilderness: The Murmuring Motif in the Wilderness Traditions of the Old Testament (Nashville, TN: Abingdon Press, 1968) and more recently Aaron, Etched, 190–193. 10 E.g., Deut 20:3; Josh 10:25; 1 Sam 7:7–11. See Jan A. Wagenaar, “Crossing the Sea of Reeds (Exod. 13–14) and the Jordan (Josh. 3–4): A Priestly Framework of the Wilderness Wandering,” in Studies in the Book of Exodus: Redaction–Reception–Interpretation, ed. Mark Vervenne (Leuven: Leuven University Press, 1996), 461–470 (465), and William H.C. Propp, Exodus 1–18: A New Translation with Introduction and Commentary (New York: Doubleday, 1999), 495. literary theory and composition history of the torah 57 plaint and response in Exod 14:10–14 are typically assigned by historical critics like Marc Vervenne to J and E, with the holy war language attributed to a (proto-)Deuteronomistic redaction (RJE).11 But it is notoriously difficult to isolate proposed Deuteronomistic redactional layers, and even Vervenne does “not believe that the process of growth of this narrative thread can be entirely reconstructed.”12 If we cannot discern the contours of a redactional layer, however, the only warrant we are left with for positing one is our presumed model of composition history, and we then run the risk of reconstructing a process that simply instantiates our presumed model rather than one that accounts for the shape of the text, which is what Literarkritik is supposed to do. John Van Seters approaches the problem of Deuteronomistic language differently, arguing that J made use of Deuteronomistic materials; indeed, holy war language is not limited to Moses’ response but carries through the entire sea crossing narrative, including YHWH causing the enemy to panic (Exod 14:24; cf. Exod 23:37; Deut 2:15; Josh 10:10; Judg 4:15; 1 Sam 7:10; 2 Sam 22:15) and leaving not a single enemy (Exod 14:28; cf. Judg 4:16).13 Van Seters’ analysis illustrates that an author can draw repertoire from different background contexts to compose a narrative, and that variations in language and style are not always significant for distinguishing one compositional layer from another.14 The scribe has also blended the basic form of a salvation oracle with the complaint motif, and this blend facilitates the development of a second theme, that of salvation.15 Not only does Moses calm the Israelites’ fear ישועת) of the Egyptians, he also anticipates their deliverance from them .(verse 30 ,ויושע יהוה ביום ההוא את ישראל מיד מצרים .verse 13; cf ,יהוה

11 Marc Vervenne, “The ‘P’ Tradition in the Pentateuch: Document and/or Redaction?: The ‘Sea Narrative’ (Ex 13,17–14,31) as a Test Case,” in Pentateuchal and Deuteronomistic Studies: Papers Read at the XIIIth IOSOT Congress, Leuven 1989, ed. Christian Brekelmans & Johan Lust (Leuven: Leuven University Press, 1990), 67–90, (78); Marc Vervenne, “The Sea Narrative Revisited,” Bibl 75 (1994): 80–98 (84–85); and Marc Vervenne, “Exodus Expulsion and Exodus Flight: The Interpretation of a Crux Critically Re-Assessed,” JNSL 22 (1996): 45–58 (50). 12 Vervenne, “Sea Narrative Revisited,” 85. 13 John Van Seters, The Life of Moses: The Yahwist as Historian in Exodus-Numbers (Lou isville, KY: Westminster John Knox, 1994), 134–135. 14 The classic case of this in Pentateuchal studies is arguably the blend of priestly and deuteronomistic style in texts composed relatively late in the formation of the Torah; see Lothar Perlitt, “Priesterschrift im Deuteronomium?” in Deuteronomium-Studien (FAT 8; Tübingen: Mohr-Siebeck, 1994), 123–143. followed by a promise of deliverance, and resulting in trust of the prophet ,אל תיראו 15 is a structure like that of Isa 7:4–9; see Van Seters, Life of Moses, 134–135. 58 angela roskop erisman

”,to cry out“ ,צעק The theme of salvation is set up in verse 10, as the verb is used to articulate the Israelites’ complaint (Exod 14:10).16 Choice of this verb suggests an effort to blend the complaint motif with the broader exodus narrative: The Israelites initially cry out to YHWH in Exod 2:23, and YHWH later indicates that he has heard their cry for salvation and will answer it through Moses (Exod 3:7–10).17 Moses indeed plays a role in YHWH’s act of drying up the sea, and the dénouement of the sea crossing narrative (Exod 14:30–31) indicates that this act of salvation has the desired result: It has inspired in the Israelites the fear of YHWH which they lacked at the beginning of the narrative, and prompted the Israelites to trust Moses as well. The three major themes of the narrative— fear of YHWH, salvation, and trust in Moses—are thus inextricably bound together. Thomas Dozeman notes that the complaint motif “is a rejection both of YHWH’s salvation and of Moses’ authority.”18 Indeed, the complaint in Exod 14:10–12 is key to the development of the narrative, because it establishes the tensions that will ultimately be resolved when the Israelites are saved and learn to both fear YHWH and trust Moses. The Israelites’ wish to return to Egypt is a standard element of the complaint motif (see Exod 16:3; Num 11:46; Num 14:2–3; Num 21:5), but Exod 14:12 is unique in its statement that “it is better for us to serve the Egyptians than to die in the wilderness.” This particular complaint puts the Israelites’ pending salvation in jeopardy because they express a desire to move away from salvation rather than toward it; as William H.C. Propp notes, they show an “ambivalence toward freedom.”19 This ambivalence is the result of being caught between their fear of the Egyptians and their fear of the potential dangers in the wilderness. What they perceive to be in their best interest in this liminal situation prevents them from seeing the possibility of real freedom and from fearing YHWH. It also prompts them to lash out at Moses, and their distrust of his leadership is expressed in two reversals: “Let us be so that we may serve the Egyptians” (Exod 14:12) is a reversal of Moses’ request to Pharaoh to let the Israelites go so they may serve

Exod) לון Exod 17:2; Num 20:3) and) ריב The typical verbs in complaint episodes are 16 Num 21:2 does use .דבר ב and Num 21:5 uses קהל ב Num 14:2). Num 20:2 and 16:3 use ;16:2 when the Israelites call on Moses for help, but their initial complaint to YHWH is צעק .אנן framed with 17 Diana Vikander Edelman, “The Creation of Exodus 14–15,” in Jerusalem Studies in Egyptology, ed. Irene Shirun-Grumach (Wiesbaden: Harrassowitz, 1998), 137–158 (146). 18 Dozeman, Exodus, 301. 19 Propp, Exodus 1–18, 494. literary theory and composition history of the torah 59

YHWH (see Exod 4:23; 7:16, 26; 8:16; 9:1, 13; 10:3, 7).20 We also find a reversal מה זאת עשית לנו להוציאנו את in (להוציא ממצרים) of the exodus formula ,(What have you done to us, bringing us out of Egypt?” (Exod 14:11“ ,מצרים as the Israelites use it to speak of pending doom rather than salvation. In both cases, the Israelites accuse Moses of acting against their best interest when the problem is not Moses but the Israelites’ interpretation of their predicament. By the end of the narrative, their trust in Moses is established along with their grasp of the true nature of the situation. Source critics sometimes separate verses 11–12 from verses 10, 13–14 because they find the theme of fear in verses 10, 13–14 but not in verses 11–12. Moreover, they note that verses 11–12 are concerned with death in the wilderness, while verse 10 is concerned with the pursuing Egyptians.21 One of the contentions formalist literary critics have with historical criticism, as framed by David Gunn and Danna Nolan Fewell, is that “the analysis of sources, fundamental to the method, was basically dependent on aesthetic premises which were often arbitrary and rarely acknowledged.”22 The composite character of biblical literature has long been acknowledged, but what may seem to be fractures in a narrative are not always real fractures. So how does one tell what constitutes a fracture in the text? As Meir Sternberg put it, “the task of decomposition calls for the most sensitive response to the arts of composition,” and Iser’s aesthetic theory offers one way to get control of the discussion: Where elements of cultural repertoire work well together to serve a coherent set of literary goals, we should avoid positing unevenness in the text and drawing conclusions about composition history on that basis.23 When we consider the role of the complaint (verses 10–12) in the development of the sea crossing narrative, we see that there is no unevenness

20 Levin, “Source Criticism,” 45. 21 Noth, Exodus, 106, attributes Exod 14:11–12 to E, while Levin, “Source Criticism,” 44, sees them as part of a late redaction. Coats, Rebellion, 136, adds that verses 11–12 take a negative, accusatory tone in an otherwise positive narrative about Israel’s deliverance. 22 David M. Gunn & Danna Nolan Fewell, Narrative in the Hebrew Bible (Oxford: Oxford University Press, 1993), 8. 23 Meir Sternberg, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading (Bloomington, IN: University of Indiana Press, 1985), 16. Iser’s approach to aes thetics is certainly not the only literary theory one could use; I prefer it to formalist theo ries because his notion of repertoire drawn from a shared cultural background has the potential to help us understand how ancient texts are rooted in the cultures in which they were produced as well as how we as readers (separated now by significant temporal and cultural distance) receive these texts. The point is simply to have a solid and transparent means of control so that we are not building arguments about composition history on a foundation of ultimately irrelevant observations. 60 angela roskop erisman in the text. The Israelites fear both the Egyptians and the possibility of death in the wilderness; the latter is simply expressed as anger directed at Moses in order to develop the theme of trust in him alongside the themes of fear of YHWH and salvation by placing all three in question. Indeed, the Israelites complain to both YHWH and Moses in order to express lack of confidence in both of them.24 Moses’ response (verses 13–14) then moves all three problems toward their resolution: He exhorts the Israelites not to be afraid, announces their pending salvation which will inspire their fear of YHWH, and begins to win their trust himself. In other wilderness complaint episodes, Moses responds to the Israelites by complaining to God (Num 14:5; 20:6) or falling on his face before God (Exod 17:4; Num 11:1–15), but here he speaks to the people, and what he says is going to happen does happen in the end, establishing him as the trustworthy character he is said to be in Exod 14:31. To separate verses 11–12 from verses 10, 13–14 is therefore to break up elements of the complaint motif and destroy the fabric of the narrative. The complaint motif is not limited to the Israelites’ complaint and Moses’ response but includes the elements of divine intervention and miracle as well, and the latter are used in the sea crossing narrative to help establish trust in Moses. Divine intervention in the rock/water episodes involves YHWH giving Moses instructions on how to carry out his role in the miracle, which he follows in one case (Exod 17:6) and fails to follow in the other (Num 20:7–11). The sea crossing narrative also involves instructions, which Moses follows to the letter. The first command is issued in Hold out your arm over the sea,” and the“ ,ונטה את ידך על הים :Exod 14:16 expression is repeated nearly verbatim when the sea is dried up in verse 21 Moses held out his arm over the sea”) and again“ ,ויט משה את ידו על הים)

24 Joel S. Baden, The Composition of the Pentateuch: Renewing the Documentary Hypoth esis (New Haven, CT: Yale University Press, 2012), 200, argues that the Israelites’ complaint (should be assigned to a different source (P (ויצעקו בני ישראל אל יהוה) to YHWH in verse 10 than the rest of verses 10–14 (J), because he sees it as a doublet of the Israelites’ complaint to Moses in verse 11, but this analysis fails to reckon with fact that fear of YHWH and trust in Moses are established through their cooperation in bringing about a single event. The development of these themes naturally involves the Israelites addressing complaints to both characters in order to establish the lack of confidence in them which is remedied by as discussed above, introduces the theme of ,ויצעקו the end of the narrative. Note also that salvation in this episode and is thus connected with Moses’ response in verses 13–14, which assures the Israelites that salvation will indeed come. Baden is right to note the presence of a double complaint in verses 10–11, but this doublet is significant for the development of character and theme in the narrative and thus should not be taken as a source-critical indicator. literary theory and composition history of the torah 61 when it is brought back over the heads of the Egyptians in verses 26–27. This command-execution structure is typically assigned by source critics to P and understood to accompany Moses’ splitting of the sea, while the complaint and response elements are assigned to J, where YHWH dries up the sea himself. This source division not only breaks up elements of the complaint motif used to develop the narrative, but also destroys the idea that YHWH and Moses work in tandem to get the Israelites out of Egypt and earn their awe and trust.25 This is evident in Exod 14:16–17, where Moses’ and YHWH’s respective roles are contrasted in clauses with fronted pronouns: As for you, raise your staff, and hold ,ואתה הרם את מטך ונטה את ידך על הים out your arm over the sea . . . As for me, I will harden the hearts of the ,ואני הנני מחזק את לב מצרים Egyptians . . .26 We as readers see that the staff in Moses’ hand is not actually what dries up the sea; YHWH does this himself using a strong east wind (Exod 14:21). Moses’ act of holding his staff over the sea appears to the Israelites as a sign of the salvation YHWH is working, inspiring their trust in Moses’ leadership just as the precise command-execution structure inspires it in the reader. The miracle itself comes in the form of YHWH drying up the sea so the Israelites can escape Egypt by crossing on dry land. Because this act evokes a common myth in the ancient Near East in which a deity creates the world by conquering chaos in the form of a personified sea, the sea crossing

25 Marc Vervenne, “The Protest Motif in the Sea Narrative (Ex 14,11–12): Form and Structure of a Pentateuchal Pattern,” EThL 63 (1987): 257–271, limits the motif to the pro test against leaving Egypt (manifest in the sea crossing narrative in Exod 14:11–12), but it involves a more extensive plot structure which must be accounted for in an effort to explain its function in the narrative; cf. Aaron, Etched, 190–193. Moreover, because the themes (fear of YHWH, salvation, trust in Moses) developed by means of the command- execution structure (typically P) are resolved in Exod 14:30–31 (typically J, like the complaint and response), they should not be broken into two different sources; for typical source divisions, see Brevard S. Childs, The Book of Exodus: A Critical, Theological Commen tary (OTL; Philadelphia, PA: Westminster, 1974), 220. 26 Peter Weimar, Die Meerwundererzählung: Eine redaktionkritische Analyse von Ex as unnecessary because ואתה Wiesbaden: Harrassowitz, 1985), 177–178, sees) 14,31–13,17 YHWH is already speaking to Moses in verse 15 and relegates it to a redactional effort to bridge verses 15 and 17. But Weimar misses the purpose of fronting pronouns in Hebrew syntax, and Vervenne, “‘P’ Tradition,” 84, is correct that they play an important role in the structure of the narrative. 62 angela roskop erisman narrative is often viewed as a historicized myth.27 Michael Fishbane offers an important critique of this approach in Biblical Myth and Rab binic Mythmaking, where he points out that biblical scholars have often brought to their exegesis a misguided understanding of myth (as a primi- tive intellectual form) and how it was employed by Israelite scribes (who retained vestiges of myth in order to neutralize them in favor of history).28 By studying uses of the divine combat motif in a variety of poetic texts, Fishbane demonstrates that scribes did not merely tolerate myths but actively made them, arguing that we should consider how a mythic motif is used in any particular instance in order to understand the “complex symbiosis” between the motif and the concerns of the literary context.29 We might think of the divine combat motif as an element of cultural repertoire, typically used in order to invoke these mighty primordial acts of creation as an expression of confidence in or appeal to YHWH’s power to save in a situation of communal or individual crisis (e.g., Ps 74:15; 2 Sam 22:8–21//Ps 18:8–20; Isa 43:1b-2, 16–19; 44:27; 50:2; 51:9–11). The miracles in other complaint episodes are specific to their respective situations, and the same is true here: The divine combat motif is blended with the complaint motif to create the dramatic act of salvation which gets the Israelites out of Egypt and inspires awe in YHWH as well as trust in Moses. מה זאת עשית לנו להוציאנו When the Israelites complain, they ask Moses What have you done to us, taking us out of Egypt?” (Exod“ ,ממצרים 14:11). This question is anticipated by Pharaoh in verse 5, when he says, What have we done, releasing“ ,מה זאת עשינו כי שלחנו את ישראל מעבדנו Israel from our service?” Propp calls this a case of “unconscious irony,” but it is not irony, and it is not unconscious on the part of the scribe.30 Pharaoh’s realization of his error is a deliberate echo of the complaint motif which establishes YHWH’s act of drying up the sea as a response to the Egyptians as much as to the Israelites, and the miracle is also used to develop a secondary theme: YHWH’s judgment against Egypt.31 Fishbane notes that the divine combat motif is sometimes used to express YHWH’s

27 E.g., Carola Kloos, Yhwh’s Combat with the Sea: A Canaanite Tradition in the Religion of Ancient Israel (Amsterdam: Van Oorschot, 1986); Bernard F. Batto, Slaying the Dragon: Myth making in the Biblical Tradition (Louisville, KY: Westminster/John Knox, 1992), 116–117. 28 Michael Fishbane, Biblical Myth and Rabbinic Mythmaking (Oxford: Oxford Univer sity Press, 2003), 1–7. 29 Fishbane, Biblical Myth, 21. 30 Propp, Exodus 1–18, 494. 31 I refer to judgment of Egypt as a secondary theme because it is not one of the three major themes explicitly resolved in the dénouement of the narrative (Exod 14:30–31); rather, it finds its last expression in verse 18 before the sea crossing is narrated. literary theory and composition history of the torah 63 power to judge as well as to save: Ezekiel 29, for example, is an oracle of judgment against Pharaoh which depicts him as the sea monster defeated by YHWH, and the oracle announcing YHWH’s pending destruction and exile of Egypt by means of Nebuchadnezzar in Ezekiel 30 also employs the divine combat motif (verse 12).32 The sea event in Exodus 14 is an act of judgment against the Egyptians as well, evident in Exod 14:21, where YHWH dries up the sea with a strong east wind. The east wind, or is a common tool of judgment in prophetic passages, in many ,רוח קדים cases used for drying up or scorching something, be it Israel as a vine or Jonah after his shade plant had been destroyed.33 The sea event was designed as an act of judgment as well as salvation from the plan’s very inception. YHWH notes that he will bring the Egyptians out after the so the Egyptians will know that I am“ ,וידעו מצרים כי אני יהוה Israelites YHWH” (Exod 14:4), an expression which appears repeatedly in the judgment oracle in Ezekiel 30 (verses 8, 19, 26). Again, source-critical studies have tended to break up what is inextricably connected: Verses containing this expression are typically attributed to P, while YHWH drying up the sea in Exod 14:21 is attributed to J.34 Since the two work together to develop the theme of YHWH’s judgment against Egypt, to separate them is to destroy the fabric of the narrative. Exodus 14 certainly resonates with the ancient Near Eastern mythic background, as all uses of the divine combat motif do, but it is not a historicized myth. Rather, the scribe who composed the sea crossing narrative blended the divine combat motif with other elements of cultural repertoire, including the complaint motif and Deuteronomistic war ideology as well as expressions from salvation and judgment oracles, in order to create a new narrative about Israel’s salvation in a particular situation. As Fishbane notes, “[m]ythmaking is also a learned and literary act that, far from being a feature of degeneration or decreased spontaneity, is often a key factor in the revitalization of earlier sources and is a sign of ongoing cultural creativity.”35 The sea crossing narrative appears to have become a paradigm of salvation for the Israelites—a myth in its own right—evident

32 Fishbane, Biblical Myth, 61; other examples include Nah 1:3–4; Jer 51:34–37. 33 General examples include Jer 18:17; Job 27:21; examples of the wind causing a ship wreck include Ezek 27:26; Ps 48:8; and examples of a drying, scorching wind include Gen 41:6, 23; Ezek 17:10; 19:12; Hos 13:15; Jonah 4:8; see Jean-Louis Ska, “Séparation des eaux et de la terre ferme dans le récit sacerdotal,” NRT 103–104 (1981): 512–532 (527, 530). 34 E.g., Childs, Book of Exodus, 220. 35 Fishbane, Biblical Myth, 20. 64 angela roskop erisman in poetic texts such as Psalm 78, which use the divine combat motif, but appear also to have this specific narrative in mind.36 Exodus 14 is thus a quintessential example of the kind of mythmaking Fishbane discusses. The itinerary genre is yet another element of cultural repertoire used to construct the sea crossing narrative. A series of three itinerary notices brings the Israelites out of Eygpt in the present form of the text: The Israelites set out from ויסעו בני ישראל מרעמסס סכתה Exod 12:37 Ramesses to Sukkot. They set out from Sukkot and ויסעו מסכת ויחנו באתם בקצה Exod 13:20 camped at Etham, at the edge המדבר of the wilderness. Speak to the Israelites so that דבר אל בני ישראל וישבו ויחנו Exod 14:2 they turn and camp before לפני פי החירת בין מגדל ובין הים Pi-hahiroth, between Migdol לפני בעל צפן נכחו תחנו על הים and the sea, before Baal- zephon; you should camp next to the sea. Source-critical studies have typically viewed Exod 13:20 as a J notice. Etham is taken to be the setting of the J sea crossing narrative which follows, and its location at the edge of the wilderness makes the sea crossing the first event in the wilderness narrative. The P notice in Exod 14:2 is understood to provide a different setting for the sea crossing narrative, one located in Egypt rather than the wilderness, making the sea crossing the last event of the exodus.37 A major problem with this view is that the sequence of stops at Rameses, Sukkot, and Etham established in Exod 12:37 and 13:20 takes the Israelites out of Egypt through Wadi Tumilat, which is nowhere near a sea. This problem illustrates how important it is to consider the background from which repertoire is drawn and its function in a narrative. The geographical names in these itinerary notices are not empty placeholders but are used to create setting and, because they refer to places whose loca- tions we are more or less clear about, we can judge whether or not they

36 Psalm 78:12–13 refers to splitting the sea specifically in Egypt and taking the Israelites through it. See also Ps 114, which refers specifically to a sea event in Egypt; Isa 11:15–16, which compares the departure from exile in Assyria with YHWH’s drying up of the Egyp tian sea to create a path out; and Isa 43:16–19, which refers to the destruction of horses and chariots, specific to Exodus 14. Other uses of the divine combat motif are more generic. 37 E.g., Childs, “Traditio-Historical Study”; Dozeman, Exodus, 182–183, 199–200; Thomas B. Dozeman, “The Priestly Wilderness Itineraries and the Composition of the Pentateuch,” in The Pentateuch: International Perspectives on Current Research, ed. Thomas B. Dozeman, Konrad Schmid & Baruch J. Schwartz (Tübingen: Mohr Siebeck, 2011), 257–288 (262–265). literary theory and composition history of the torah 65 constitute a plausible setting.38 Exodus 14:2 includes places near a sea, but the edge of the wilderness at Etham (Exod 13:20) is not a plausible setting for the sea crossing narrative at all. Just as working with unclear aesthetic premises can lead to identification of fractures in a narrative which are not really there, it can also result in failure to see inconsistencies that are present and significant. Reinhard Kratz understands the whole series of itinerary notices in Exod 12:37, 13:20, and 14:2 to have been formed around a previously existing sea crossing narrative in order to make it a stage on the itinerary out of Egypt.39 Joel Baden holds a similar view, assigning all three notices to his P source and understanding Exod 14:2 as a “direct continuation” of Exod 13:20.40 But the notion that all three itinerary notices are of a piece (whether a single redaction or a single source) is problematic, because the itinerary notice in Exod 14:2 is not like Exod 12:37 and 13:20 in form. Again, background context places an important constraint on interpretation: No extant itinerary document in the ancient Near East changes form in the middle, so it is highly unlikely that we have a coherent string of notices here.41 Exodus 12:37 and 13:20, along with most of the itinerary notices throughout the for arrival in חנה for departure and נסע wilderness narrative, use the verbs their narrative preterite forms, and I have argued elsewhere that this series of notices is part of an effort by a priestly scribe to emplot the wilderness narrative like an Assyrian annal.42 But Exod 14:2 is framed as direct speech to divert the Israelites’ route away from their departure into שוב and uses the wilderness through Wadi Tumilat in the Pesah narrative, where there is no sea. Baden argues that the “route toward the sea is entirely unnecessary for the Israelites’ escape,” because in Exod 14:2 they are simply “retracing their steps for the sole purpose of luring the Egyptians toward the water, where Yahweh will destroy them.”43 This view neglects the geography as well as the form: In Exod 12:37 and 13:20, the Israelites are nowhere near a sea, so they are not merely retracing their steps in Exod 14:2, and the route articulated there is absolutely necessary to establish a plausible setting

38 On exodus geography, see Angela R. Roskop, The Wilderness Itineraries: Genre, Geog raphy, and the Growth of Torah (Winona Lake, IN: Eisenbrauns, 2011), 237–252. 39 Reinhard G. Kratz, The Composition of the Narrative Books of the Old Testament, trans. John Bowden (London: T & T Clark, 2005), 284. 40 Baden, Composition, 199, 206. 41 Roskop, Wilderness Itineraries, 186 (the point is illustrated by texts discussed on 50–82). 42 Roskop, Wilderness Itineraries, 136–184, especially the list of itinerary notices on 178. 43 Baden, Composition, 205. 66 angela roskop erisman for the sea crossing narrative. The itinerary notice in Exod 14:2 may best be viewed as a revision to the priestly annalistic version of the wilderness narrative which takes them toward a seaside setting on the coastal route out of Egypt appropriate for the sea crossing.44 The revision signaled by Exod 14:2 cannot be limited to the redactional addition of an itinerary notice, since elements of the itinerary genre are used throughout the sea crossing narrative. Verse 9 again uses the camp- ,and repeats elements of the setting established in verse 2 חנה ing verb does not occur until verse 15 when YHWH נסע while the departure verb commands Moses to tell the Israelites to move forward toward the sea, only after—and in response to—their complaint that involves a desire to go back to Egypt. These verses are all typically attributed to P, and use of language characteristic of the itinerary genre in these cases could, as Frank Moore Cross suggested, mean that the itinerary notice is part of a P -used in verse 19 to indi נסע reworking of a JE narrative.45 But we also find cate the movement of the pillar of cloud and fire, and this verse is typically attributed to JE.46 The pillar serves to guide the Israelites on their wilderness journey, so it typically sets out in front of them (cf. Exod 40:36–38; Num 9:15–23; 14:14; Deut 1:32–33; Ps 78:14 and 105:39; Neh 9:12, 19). In Exod 14:19, however, it sets out behind them to act as their rearguard, a creative adaptation of the itinerary genre to serve the need, unique to this narrative, to protect the Israelites from the Egyptians and give them a head start into the seabed.47 Elements of the itinerary genre are thus blended with the complaint motif, the divine combat motif, Deuteronomistic war ideology, and prophetic imagery to aid the development of the entire sea crossing narrative.

44 Roskop, Wilderness Itineraries, 215–218; this is true of not just Exod 14:2 but all atypical formulations of the itinerary genre in the wilderness narrative, as I discuss on pp. 185–232. 45 Frank Moore Cross, Canaanite Myth and Hebrew Epic: Essays in the History of the Religion of Israel (Cambridge, MA: Harvard University Press, 1973), 310. 46 E.g., Noth, Exodus, 106; Coats, Rebellion, 131; Childs, Book of Exodus, 220; Propp, Exo dus 1–18, 480. 47 Baden, Composition, 198, argues that the pillar of cloud and fire (which he assigns to J) is at odds with the notion that the Egyptians pursue the the Israelites into the sea (which he assigns to P; the Israelites do not cross the sea in his reconstruction of J), because the Egyptians would need to see the Israelites in order to pursue them, which the pillar pre vents them from doing. It is not difficult, however, to understand that the pillar of cloud and fire keeps the Egyptians at bay long enough for the Israelites to get a head start, ulti mately preventing the Egyptians from catching up with them in the middle of the seabed. In fact, one must envision such a scenario in order for the Israelites to get to the other side before the sea returns to its normal state, so that only the Egyptians (and not the Israelites as well) are wiped out when it does. literary theory and composition history of the torah 67

Childs was fundamentally right that the itinerary notice in Exod 14:2 brings the Israelites at least nominally back into Egypt after they had already arrived at the edge of the wilderness in Exod 13:20, so the sea crossing, rather than Pesah, could be depicted as the final departure from Egypt.48 But the itinerary notice is not a feature of a P alternative to a JE sea narrative set in the wilderness. Rather, it serves to accommodate the sea crossing episode to the exodus narrative as a new paradigmatic exodus event. Accommodation refers to “the presentation of new information in a backgrounded way” so that, although new, its “presence in the text world is presented as an unremarkable fact,” and we read the new information as though it naturally belongs there.49 The priestly narrative, structured by the itinerary notices in Exod 12:37 and 13:20, took the Israelites out of Egypt through Wadi Tumilat in an escape associated with the plague of the firstborn and the observance of Pesah; it had no sea crossing. Exodus 14 as a whole is a post-priestly addition to the exodus narrative, and the scribe(s) who wrote it used the itinerary genre to link this new episode to the previous chain of itinerary notices and take the Israelites farther north to an appropriate setting at the sea. A number of other efforts to accommodate the sea crossing narrative to the exodus and wilderness narratives also suggest that it was written specifically for this context and explain its transitional character.50 Exodus 14 is tightly linked to the plague narrative by a number of shared expres- I [YHWH] will harden the heart” of Pharaoh and the“ ,וחזקתי את לב :sions Egyptians (Exod 14:4, 8, 17; cf. Exod 4:21; 7:13, 22; 8:15; 9:12, 35; 10:20, 27; ”the Egyptians shall know that I am YHWH“ ,וידעו מצרים כי אני יהוה ;(11:10 (Exod 14:4, 18; cf. Exod 5:2; 7:5, 17; 8:6, 18; 9:14–16, 30; 10:2); Pharaoh’s regret ;in Exod 14:5; cf. Exod 4:21, 23; 5:1–2 שלח) at sending the Israelites away 6:1, 11; 7:2, 14, 16, 26–27; 8:4, 16–17, 24–25, 28; 9:1–2, 7, 13, 17, 28, 35; 10:3–4, 7, 10, 20, 27; 11:1, 10; 12:33; 13:15, 17); and Moses’ use of his staff (Exod 14:16; cf. Exod 4:1–5; 7:8–13; 7:19–25; 8:1–3; 9:22–26; 10:12–13 as well as Exod 17:5;

48 Coats, “Sea Tradition,” 4, argues that Exod 14:2 does not bring the Israelites back into ”.is “simply one of several [verbs] that can appear in itinerary formulas שוב Egypt because the ,שוב In fact, Exod 14:2 is the only itinerary notice in the wilderness narrative that uses The fact that Exod 15:22 has the Israelites departing into .חנה and נסע typical verbs being a different wilderness than Exod 13:20 supports Childs’ interpretation against Coats’. 49 Paul Werth, Text Worlds: Representing Conceptual Space in Discourse (New York: Long man, 1999), 280, 56. 50 I do not necessarily mean to imply that no tradition of a sea crossing existed prior to the composition of Exodus 14, just that it had not yet become part of the standard exodus narrative and that the version we have had not been written. If there was such a tradition (and there seems to have been, if Exodus 15 is an early Hebrew poem), it was yet another element of cultural repertoire used to construct the narrative. 68 angela roskop erisman

Num 20:8–11).51 The first two expressions are found in verses typically attributed to P and are used to support the source-critical conclusion that P construes the sea crossing as the last event in Egypt and the culmination of the plague narrative.52 While these expressions may occur in other priestly texts, however, not every instance of them must be assigned to the same source or redactional layer; as Suzanne Boorer pointed out, “the relative levels of texts cannot be determined on the basis of formulaic comparisons alone.”53 Given the character of the sea crossing narrative as a revision to P, these expressions are better understood as efforts to blend it in, so it reads as a continuation of what precedes. Yet another sign of accommodation to the exodus narrative is the effort to set the sea crossing at the very end of the night (Exod 14:20–21, 24) in order to frame it as a continuation of the departure associated with the plague of the firstborn, which takes place in the middle of the night (Exod 11:4; 12:29–31, 42). Efforts to accommodate the sea crossing narrative to the previously existing text also look forward to the wilderness. Diana Edelman argued that Exodus 14 “does not presume the existence of the wilderness wandering traditions,” but use of the complaint motif to provide its basic framework is an important accommodation to those very traditions.54 The pillar of cloud and fire is another key element of the wilderness narrative, where its primary role, “repeated at crucial junctures in the P History,” is to guide the Israelites through the wilderness.55 But it also functions as a divine melammu, whose terror-inspiring brilliance overwhelms the enemy and protects the Israelite army.56 The pillar is thus a fitting element of the priestly annalistic version of the wilderness narrative, and we should not be surprised to see discussion of its role accompany the first P itinerary

51 These connections to the plague narrative are well-rehearsed in the secondary lit erature; see Dennis J. McCarthy, “Plagues and Sea of Reeds: Exodus 5–14,” JBL 85 (1966): 137–158; Edelman, “Creation,” 146–147; Dozeman, Exodus, 312. The hardening of Pharaoh’s heart is, in fact, intimately connected to the use of Dtr holy war ideology; see Robert R. Wilson, “The Hardening of Pharaoh’s Heart,” CBQ 41 (1979): 18–36 (34). 52 Childs, Book of Exodus, 223. 53 Suzanne Boorer, The Promise of the Land as Oath: A Key to the Formation of the Pen tateuch (Berlin: Walter de Gruyter, 1992), 444. 54 Edelman, “Creation,” 143. 55 Dozeman, Exodus, 310. 56 For the functions of the pillar of cloud and fire, see Thomas W. Mann, “The Pillar of Cloud in the Reed Sea Narrative,” JBL 90 (1971): 15–30, and especially Moshe Weinfeld, “Divine Intervention in War in Ancient Israel and in the Ancient Near East,” in History, Historiography, and Interpretation: Studies in Biblical and Cuneiform Literatures, ed. Hayim Tadmor & Moshe Weinfeld (Jerusalem: Magnes Press, 1986), 121–147 (132–136). literary theory and composition history of the torah 69 notice to bring the Israelites into the wilderness in Exod 13:20–22.57 Just as Exod 14:2 ties into the P itinerary notice in Exod 13:20, so reference to the pillar of fire and cloud in the sea crossing narrative picks up the priestly repertoire in Exod 13:21–22, but emphasizes its melammu-like function moreso than its guiding function in keeping with the character of the sea narrative as a battle scene: In Exod 14:19–20, the pillar functions as a rearguard to protect the Israelites from the Egyptians, while in verse 24 it throws the Egyptians into a panic so the Israelites can escape.58 Such efforts to accommodate Exodus 14 both backward to the exodus and forward to the wilderness account for its transitional character better than a process of tradition history and combination of sources. We have now seen a number of cases in Exodus 14 alone which illustrate that not every perceived tension in a narrative is a sign of composition history. Indeed, creating and resolving tensions is how an author develops a narrative. Iser argued that our brains are constantly trying to grasp the meaning of a text as we read. We build a coherent understanding of the discourse by negotiating textual cues created by the author, fitting each new piece of information into the picture we have already built at any given point in the process. A horizon, or point of tension, occurs when we encounter a new feature that prompts us to renegotiate that understanding.59 Some horizons are productive tensions created by the author in a well-constructed narrative and are eventually either resolved or left

57 Exodus 13:17–22 are often taken as part of the sea crossing narrative; e.g, Childs, Book of Exodus, 220–221, 223–224. But it should now be clear that verses 20–22 are part of the P narrative to which the sea crossing was later added. Verses 17–19 are a still later addition aimed at repairing the confusion generated when the sea the Israelites crossed was identi fied as Yam Suf (Exod 15:4, 22), which conflicted with the typical use of Yam Suf elsewhere in Tanakh to refer to the two northern arms of the Red Sea (e.g., Exod 23:31; Num 14:25; 21:4; Deut 1:40; 2:1; 1 Kgs 9:26; Jer 49:21). Exodus 13:17–19 restores Yam Suf to its typical referent; see Roskop, Wilderness Itineraries, 247–252. Consequently, the sea crossing narrative begins only in Exod 14:1. 58 The messenger of God and the pillar of cloud and fire in Exod 14:19 are often assigned to two different sources, E and J, respectively. But, as Van Seters notes, both embody the “guiding presence of God [. . .] who provides victory over Israel’s enemies” (Life of Moses, 130). Important instances in other literature of the divine melammu (or the god himself— the two are interchangeable) throwing the enemy into panic are cited by Weinfeld, “Divine Intervention,” 133–136. Yet another effort to accommodate the sea-crossing narrative to the ביד רמה military character of the priestly narrative to which it is added is the expression in Exod 14:8, which C.J. Labuschagne identifies as a “military, or semi-military, expression, signifying readiness to fight and the will to prevail” (“The Meaning of beyād rāmā in the Old Testament,” in Von Kanaan bis Kerala: Festschrift für J.P.M. van der Ploeg zur Vollendung des 70. Lebensjahres am 4.9.1979, ed. Wilhelm C. Delsman, et al. [Neukirchen: Neukirchener Verlag, 1982], 143–148 [146]). 59 Iser, Act of Reading, 96–99. 70 angela roskop erisman in such a way as to contribute to its meaning. When we arrive at Exod 14:21, for example, we find Moses holding his arm over the sea and splitting it, but YHWH also driving it back with an east wind, and we might be prompted to wonder who really dried up the sea and how it was done. We may not be able to reconcile splitting the sea with drying it up. But, as I argued above, we can see how a partnership between YHWH and Moses in drying up the sea is critical for the thematic development of the narrative culminating in Exod 14:30–31, helping us resolve part of this horizon: YHWH’s act demonstrates his power over Israel’s enemies, while Moses provides a sign of it that inspires the Israelites to trust him as well as YHWH. Other horizons are unproductive but simply constitute poor composition. In Exod 14:15, YHWH says to Moses, “Why do you cry out to me?” despite the fact that Moses does not cry out to God in verses 13–14. This horizon serves no meaningful purpose in the narrative. Dozeman noted, “It is as though YHWH has forgotten the setting and the circumstances of the story.”60 But it is not YHWH who has forgotten; it is the scribe. David H. Aaron emphasizes the importance of attention to aesthetics in the study of biblical literature: Aesthetics are as essential to a document as form, content, and ideological purpose. Too often scholars ignore this aspect of the text, worrying that aesthetic judgments are too prone to subjectivity to be of value. Aesthetic criticism can be done on the basis of aesthetic expectations that emerge from the literature itself rather than on the basis of some externally established rubric, especially when it pertains to aspects of narrative plausibility.61 Elements of cultural repertoire are sometimes employed poorly, revealing a scribe’s lack of skill or—more likely in this case, given the otherwise tight construction of the narrative—momentary carelessness. The horizon created by YHWH’s question in Exod 14:15 is easily explained by the fact that the complaint motif often involves Moses transferring the Israelites’ complaint to YHWH (e.g., Exod 15:25; 17:4; Num 11:2; 21:7). Vervenne suggested that “the usual scheme [. . .] is probably used and modified in a special way,” but he did not explain how.62 As I argued above, the scribe

60 Dozeman, Exodus, 314. Even the ancient translators saw the aesthetic problem here. Coats, Rebellion, 130, points out that the Peshitta “has felt this problem and inserted an introduction before vs. 15 to report that Moses also cried to Yahweh. But this reading can hardly be considered more than a correction of the inconsistency.” 61 Aaron, Etched, 191. 62 Vervenne, “Protest Motif,” 261. literary theory and composition history of the torah 71 manipulated the response element of the complaint motif in order to develop the theme of trust in Moses by having him reassure the people instead of transferring the Israelites’ complaint to YHWH. When the scribe moved on to craft YHWH’s reply to Moses, however, he appears to have reverted to what is typical in the motif without noticing that it no longer quite worked with the changes he made.63 Iser did not deal with composite literature, but his idea of horizons in a narrative provides biblical scholars with a useful way to frame our discussion of composition history. While some horizons are productive and others result from poor composition, still others are signs of diachronic development. In Exod 14:21–22, we see that the sea is driven back with a strong east wind and split in two. It is impossible to form a coherent picture of the narrative with both actions, especially since YHWH com- -leaving a different char ,(ובקעהו) mands Moses to split the sea in verse 16 acter responsible for each action—YHWH for driving it back, and Moses for splitting it. This horizon is unproductive: It does not develop plot, characterization, or theme, and the conflict it creates is never resolved.64 The whole episode, as we have seen, coheres quite tightly except for a few minor spots that are closely focused on splitting the sea and crossing in verse 21 ויבקעו המים ;in verse 16 ובקעהו :between two walls of water

63 Another example of poor composition can be found in Exod 14:20. Here the pil lar of cloud, together with the darkness of the nighttime setting, prevents the Israelites indicates that the pillar lit ויאר את הלילה and the Egyptians from engaging in conflict, yet up the night, which would compromise this purpose. Marc Vervenne, “Exodus 14,20 MT- LXX: Textual or Literary Variation?,” in Lectures et Relectures de la Bible: Festschrift P.-M. Bogaert, ed. Jean-Marie Auwers & André Wénin (Leuven: Leuven University Press, 1999), 3–25, noted a number of efforts in the versions to make sense of the awkward syntax in may ויאר את הלילה this verse, but argued that the Masoretic Text is not corrupt and that be a pre-Masoretic gloss, harmonizing Exod 14:19–20 with Exod 13:21–22, which emphasizes the daytime as well as nighttime forms of the pillar. Whether a gloss or an original part of the narrative, I am more inclined to see it as a clumsy anticipation of the pillar’s role as a melammu in verse 24, because the awkwardness stems from having to juggle the two roles the pillar plays in this narrative: obscuring the view of the enemy, which requires cloud and darkness, and acting as a melammu to throw the enemy into panic, which requires This problem would have been avoided if the .ויאר את הלילה the brightness indicated by scribe had simply saved the mention of the pillar’s brightness for the point in the narrative where it was immediately relevant. 64 Paul R. Noble, “Synchronic and Diachronic Approaches to Biblical Interpretation,” JLT 7 (1993): 130–148 (136), gets at this concept when he notes, “a diachronic reconstruction is predicated not simply upon repetitions, tensions, etc. in the text, but upon it displaying redundant repetitions and unfruitful tensions. In other words, it is built upon the observa tion of features in the text that lack sufficient intrinsic motivations—e.g., repetitions that go beyond what is needed for the text’s information-conveying purposes, and make no other positive contribution.” 72 angela roskop erisman

should also be included here, since it is a doublet of וישם את הים לחרבה) והמים להם המה מימינם in verse 22); and ויבאו בני ישראל בתוך הים ביבשה in verses 22 and 29. Source critics are right to consider this ומשמאלם horizon a sign of composition history.65 Especially where we can link elements of a horizon to a distinct set of literary goals, we are justified in suggesting the presence of a separate source or revision. Given the limited character of the passages in which splitting the sea surfaces, we are better off seeing a minor revision to the text in this case.66 While an independent source might have an entirely different set of themes and literary goals, we can expect that a revision will seek to somehow extend the base narrative. To explain this revision, we might ask the same questions we asked in our effort to understand the construction of the base narrative itself: From what background context is this element of cultural repertoire (splitting a body of water and crossing it with water on either side) drawn, and how does it alter or expand this narrative? We are sometimes able to identify more than one possible background context from which an element of cultural repertoire might have been drawn. The sea crossing narrative, as I noted above, is typically understood in the context of ancient Near Eastern mythology. Dozeman, for example, sees the version in which the sea is dried up as having been influenced by the Canaanite Yamm myth, while the divided sea reflects Enuma Elish (where Marduk cleaves Tiamat in two) and introduces the theme of new creation.67 But Moses splitting the sea can also be understood against the background of prophetic activity elsewhere in Tanakh, particularly 2 Kings 2 where Elijah and Elisha split the Jordan by slapping it with Elijah’s ,(verses 8, 14) הנה והנה mantle and cross it on dry land with the water in Exod 14:22, 29. The function of the miracle מימינם ומשמאלם similar to they perform is to demonstrate the transfer of prophetic authority from

65 E.g., Childs, Book of Exodus, 218–219; Levin, “Source Criticism,” 50. ,He turned aside the wheels of their chariots“ ,ויסר את אפן מרכבתיו וינהגהו בכבדת 66 and they drove with difficulty” (Exod 14:25) may also be part of this revision. The notion that the chariots are stuck conflicts with the fact that the sea returns to its normal state and covers them as they are fleeing (verses 25, 27–28). This creates unproductive tension in the narrative, which reads clearly without these two clauses: The Egyptians realize what is happening once they are thrown into a panic, decide to flee, and fail to get out before the sea covers them. It is difficult, however, to see what is gained in the revision by having the Egyptians be stuck since, either way, they drown in the sea; perhaps the motivation was heightened drama. I tentatively attribute these clauses to the revision simply because I can think of no good warrant for arguing that they constitute a separate revision or a gloss. 67 Dozeman, Exodus, 300, 304. literary theory and composition history of the torah 73

Elijah to Elisha for the benefit of the group of prophets who came down from Jericho to watch. Which context best helps us understand the role of Moses splitting the sea in Exodus 14? Dan Sperber and Deirdre Wilson offer a way to address this question in Relevance: Communication and Cognition. When we process a new piece of information in a discourse, we choose the background context that will maximize the relevance of the information to the discourse as we have interpreted it thus far.68 New information is judged relevant to the extent that it strengthens the understanding already constructed, weakens that understanding by calling it into question, or develops that understanding further.69 Relevance is a matter of degree; new information may be more or less relevant, depending on how significantly it strengthens, weakens, or develops the discourse.70 When we have a choice of multiple background contexts in which to understand that information, one context may yield much more in terms of relevance than others and will therefore produce the strongest interpretation. This is exactly what plays out as we consider the relative merits of relating the divided sea to the context of either ancient Near Eastern mythology or prophetic activity elsewhere in Tanakh. If Moses splitting the sea is seen as an echo of Elijah and Elisha splitting the Jordan, this act characterizes Moses as a legitimate prophet. Indeed, Moses’ prophetic character is already an issue in the base narrative, where his trustworthiness is bolstered by an effort to characterize him like Samuel, who is recognized as a trustworthy prophet by all Israel in 1 Sam 3:20. The complaint motif is blended with repertoire drawn from Samuel in Moses’ response: Exod 14:13 אל תיראו התיצבו וראו את ישועת יהוה אשר יעשה לכם היום כי אשר ראיתם את מצרים היום לא תסיפו לראתם עוד עד עולם Have no fear! Stand by and witness YHWH’s act of salvation which he will perform for you today. For the Egyptians whom you see today you will never see again.

68 Dan Sperber & Deirdre Wilson, Relevance: Communication and Cognition (Cambridge, MA: Harvard University Press, 1986), 132–142. 69 Sperber & Wilson, Relevance, 108–117. This is arguably just as true for new informa tion added in the course of a revision as it is for new information within a single compo sitional layer. 70 Ibidem, 122–125. 74 angela roskop erisman

1 Sam 12:16–17 גם עתה התיצבו וראו את הדבר הגדול הזה אשר יהוה עשה לעיניכם: הלוא קציר חטים היום אקרא אל יהוה ויתן קלות ומטר ודעו וראו כי רעתכם רבה אשר עשיתם בעיני יהוה לשאול לכם מלך Now stand by and see the marvelous thing that YHWH will do before your eyes. It is the season of the wheat harvest. I will pray to YHWH and He will send thunder and rain; then you will take thought and realize what a wicked thing you did in the sight of YHWH when you asked for a king. Exodus 14:13 and 1 Sam 12:16 are the only two passages in Tanakh other stand by and“ ,התיצבו וראו than 2 Chr 20:17 which have the expression in 1 Sam 12:16 with את הדבר הגדל הזה witness.” Exodus 14:13 replaces ,which brings in one of the main themes of the passage ,את ישועת יהוה salvation. This theme is recapped in the dénouement of the sea-crossing narrative: Exod 14:30–31 ויושע יהוה ביום ההוא את ישראל מיד מצרים וירא ישראל את מצרים מת על שפת הים: וירא ישראל את היד הגדלה אשר עשה יהוה במצרים וייראו העם את יהוה ויאמינו ביהוה ובמשה עבדו: Thus YHWH saved Israel that day from the Egyptians. Israel saw the Egyp- tians dead on the shore of the sea. And when Israel saw the wondrous power which YHWH had wielded against the Egyptians, the people stood in awe of YHWH; they trusted YHWH and His servant Moses.

1 Sam 14:22–23 וכל איש ישראל המתחבאים בהר אפרים שמעו כי נסו פלשתים וידבקו גם המה אחריהם במלחמה: ויושע יהוה ביום ההוא את ישראל והמלחמה עברה את בית און: When all the men of Israel who were hiding in the hill country of Ephraim heard that the Philistines were fleeing, they too pursued them in battle. Thus YHWH saved Israel that day, and the fighting passed beyond Beth-aven. 1 Sam 12:18 ויקרא שמואל אל יהוה ויתן יהוה קלת ומטר ביום ההוא ויירא כל העם מאד את יהוה ואת שמואל: Samuel prayed to YHWH, and YHWH sent thunder and rain that day, and the people stood in awe of YHWH and of Samuel. Thus YHWH saved Israel“ ,ויושע יהוה ביום ההוא את ישראל The phrase that day” (Exod 14:30) is a quote of 1 Sam 14:23, where salvation from the Philistines rather than the Egyptians is at issue. Moreover, the conclusion of the sea-crossing narrative mimics the conclusion of 1 Sam 12:18, where literary theory and composition history of the torah 75 the people fear YHWH and Samuel; only Exod 14:31 places fear of YHWH and trust of Moses in separate clauses suitable for the two themes running through the narrative. When we read the character of Moses in the sea crossing narrative, then, we read with the trustworthy prophet Samuel in the background and come to view Moses in a similar light. Reading Moses’ splitting of the sea as a similar allusion to 2 Kings 2 thus strengthens our understanding of his character as a trustworthy prophet. It also develops our understanding of his prophetic character further, as it now looks like Moses performs a miracle. As noted above, Exod 14:16 is revised so that YHWH commands Moses to split the sea himself rather than just hold his staff over it as a sign while YHWH does the work. This revision blurs the distinction between Moses and YHWH that is sharply maintained in the base narrative, not unlike Deut 34:10–12, which elevates Moses to an incomparable status among prophets and highlights the signs and wonders he performed in Egypt as indicators of his uniqueness.71 Viewed against the background of prophetic activity elsewhere in Tanakh, then, the revision which has Moses split the sea is highly relevant to the development of character and theme in the sea crossing narrative. Relating the divided sea to the context of ancient Near Eastern mythology, as Dozeman does, also produces a degree of relevance because it resonates with use of the divine combat motif to construct the narrative. But it is a much weaker degree of relevance, especially when we consider that drying up the sea with an east wind is a use of the divine combat motif already filtered through prophetic imagery. Moreover, the theme of new creation that Dozeman identifies here does not strengthen or develop other prominent themes in the narrative quite the way amp- ing up Moses’ prophetic character does. The ancient Near Eastern mythic context is certainly a viable background for the sea crossing narrative, but understanding Moses’ splitting of the sea as an allusion to Elijah and Elisha maximizes its relevance to the narrative and best explains the goal of this revision to the sea narrative. What I have done here is straightforward literary criticism, but it demonstrates that attention to literary concerns does not have to— indeed should not—come at the expense of serious engagement with historical issues such as the diachronic development of a biblical text or its

71 For discussion of Deut 34:10–12, see Stephen B. Chapman, The Law and the Prophets: A Study in Old Testament Canon Formation (Tübingen: Mohr Siebeck, 2000), 113–131. 76 angela roskop erisman rootedness in ancient culture. Indeed, Iser has helped us see that attention to how horizons function (or fail to function) in a narrative and the background contexts from which cultural repertoire is drawn is essential to the aesthetic experience of reading. Baruch Schwartz has argued that a particular model of composition history should be evaluated not on the basis of how theoretically plausible it is, but on how well it explains the literary problems in a text.72 It should now be evident that the Documen- tary Hypothesis is not an optimal solution to the literary problems in the sea crossing narrative, and that a supplementary model is preferable in this case. But the weaknesses of particular historical-critical studies do not, as early advocates of literary theory in biblical studies often suggested, render the historical-critical project (Literarkritik) irrelevant for the experience of reading the text as a whole. I have intended to show that literary-critical tools are essential to the success of this project and hope I have illuminated the richness we would miss in this narrative, if we failed to understand its historical development as well as its narrative artistry.

72 Baruch J. Schwartz, “Does Recent Scholarship’s Critique of the Documentary Hypoth esis Constitute Grounds for Its Rejection?,” in The Pentateuch: International Perspectives on Current Research, ed. Thomas B. Dozeman, Konrad Schmid & Baruch J. Schwartz (Tübingin: Mohr Siebeck, 2011), 3–16 (9–10). Eyewitness Accounts in the Books of Samuel? A Reappraisal

Klaas Smelik1

Abstract

The reconstruction of the history of Ancient Israel has been dominated by the concept of the eyewitness account. The books of Samuel, for instance, have been considered as largely based on such accounts. In the authoritative study by Leon hard Rost Die Überlieferung von der Thronnachfolge Davids published in 1926, it is argued that two main sources have been used in composing these Bible books, both written by a contemporary of king David and king Solomon, who had access to the testimonies of eyewitnesses. For this reason, there were no objections to relying heavily on the books of Samuel when reconstructing the history of the Early Monarchy. Literary analysis of the narratives found in the books of Samuel, however, suggests that the concept of an eyewitness account is misleading. Due to the literary skill of the authors, it is suggested to the reader that the narrator was present during the events described in the narratives but in reality this was not the case. This different approach considering the biblical account as a piece of literary art has a great impact on our understanding of the Hebrew Bible and also for the reconstruction of the history of Ancient Israel.

Modern scholarly discussion of the traditions in the books of Samuel has been shaped primarily by Leonhard Rost’s book Die Überlieferung von der Thronnachfolge Davids, which appeared in 1926.2 In Old Testament Stud ies, seldom has a hypothesis been so successful and long living.3 Even now, his analysis dominates the debate, although serious objections have been made against the validity of his theory. Before entering into the dis cussion, let us first have a closer look at Rost’s proposals.

1 I like to thank Patrick Wing for his revision of the English text. 2 Leonhard Rost, Die Überlieferung von der Thronnachfolge Davids (BWANT III, 6; Stutt gart: Kohlhammer, 1926), reprinted in Leonhard Rost, Das kleine Credo und andere Studien zum Alten Testament (Heidelberg: Quelle & Meyer, 1965). E.T. The Succession to the Throne of David, transl. Michael D. Rutter & David M. Gunn (Sheffield: Almond Press, 1982). Cita tions in the present paper are taken from the 1926 German edition. 3 “This hypothesis has been among the modern durable of modern scholarship,” P. Kyle McCarter, Jr., I Samuel: A New Translation with Introduction, Notes and Commentary (AB; Garden City, NY: Doubleday, 1980), 23. 78 klaas smelik

Leonhard Rost and the Succession Narrative

In his study, Leonhard Rost isolated the materials of 2 Sam 6:16 and 20ff.; 7:11b and 16; 9:1–10:5 (10:6–11:1); 11:2–12:7a; 12:13–25 (26–31); 13:1–14:24; 14:28–18:17; 18/19–20:22; 1 Kings 1:1–21; 2:5–10; 2:12–27a and 2:28–46 as a sin gle literary unit dominated by the thematic problem of the succession to David’s throne. He called this source in German die Thronfolgegeschichte, translated into English as the Succession Narrative, although the designa tion Court History is also in use. Rost dated the Succession Narrative to the early period of Solomon’s kingship. Next to the Succession Narrative, Rost isolated another literary unit, which he called die Ladeerzählung (in English: the Ark Narrative). Accord ing to Rost, the Ark Narrative comprised 1 Sam 4:1b-18a; 19–21; 5:1–11bα, 12; 6:1–3bα, 4, 10–14, 16, 19–7:1; 2 Sam 6:1–15; 17–20a.4 In his view, the Ark Nar rative was even older than the Succession Narrative;5 it was a hieros logos, a cult legend which “depicts the fate of the Ark from its removal from Shiloh until its installation in Jerusalem.”6 It was written by a priest for the benefit of the visitors to the Ark Sanctuary in Jerusalem, who wanted to know how the Ark had arrived at this spot. Rost even supposed that the five golden mice featured in the story could be seen there for a small remuneration!7

General Agreement

Rost’s study has strongly influenced subsequent investigation of these traditions and it has been the basis for treating them as a discrete liter ary unit within the Deuteronomistic History. In his influential monograph Überlieferungsgeschichtliche Studien (Part One) from 1943, Martin Noth concurred with Rost’s analysis and assumed that the Deuteronomistic Historian included these sources into his work while making only minor changes in the text.8

4 Rost, Überlieferung, 46. 5 Ibidem, 38: “zur Zeit Davids bzw. im Anfang der Regierung Salomos entstanden.” 6 Ibidem, 36: “die die Geschicke der Lade von der Wegführung aus Silo an bis zur Aufstellung in Jerusalem erzählt.” 7 Ibidem, 37. 8 Martin Noth, Überlieferungsgeschichtliche Studien. Teil 1: Die sammelnden und bear beitenden Geschichtswerke im Alten Testament (Schriften der Königsberger Gelehrten Gesellschaft, Geisteswissenschaftliche Klasse 18,2; Halle: Niemeyer, 1943). E.T. The Deuter onomistic History (Sheffield: Sheffield Academic Press, 1981). eyewitness accounts in the book of samuel? 79

Since then, there was a general agreement among scholars9 that in the books of Samuel several early sources are still discernible such as the Ark Narrative and the Succession Narrative but also, for instance, the History of David’s Rise in 1 Samuel 16-2 Samuel 5, another narrative also recognized as a self-contained literary unit by Rost, who, however, did not elaborate on it in his study from 1926.10 All these sources were—according to gen eral opinion—hardly reworked when included into the Deuteronomistic History. Also the date of the Succession Narrative seemed not to constitute a problem: it was written in the early period of Solomon’s kingship, as Rost already proposed. In regard to the aim of the narrative, there was general agreement as well. It was to justify and uphold the claims of Solomon himself, not primarily the Davidic dynasty. The leitmotif of the work is expressed in the words ‘who shall sit on the throne?’ (1 Kgs 1.20, 27). The purpose of the narrator is to relate how it came to pass that David was succeeded by Solomon and to defend Solomon with regard to the most incriminating of the political executions during his first years.11 There was discussion, however, about the identification of the author of the Succession Narrative. Various proposals have been made.12 But all scholars agreed that it must have been a contemporary of the events, a person in some way connected to the royal court of Jerusalem. The main point of discussion in this period was the precise delimi tation of the text. In contrast to Rost’s detailed proposal quoted above, general opinion favored a less complicated delimitation: the Succession Story comprised 2 Sam 9–20. Other propositions were made but without

9 Cf. John Barton, “Dating the ‘Succession Narrative’,” in In Search of Pre-Exilic Israel: Proceedings of the Oxford Old Testament Seminar, ed. John Day (London: T&T Clark, 2004), 95–106: “Thirty years ago it was all so simple. There was definitely a Succession Narrative [. . .] and it was evidently very early, the product of an eyewitness, or at least based on eyewitness testimony, and written at the court of Solomon [. . .].” (95). 10 The History of David’s Rise was the subject of several studies in the fifties and sixties of the last century; see for instance, Arthur Weiser, Samuel: Seine geschichtliche Aufgabe und religiöse Bedeutung (FRLANT 81; Göttingen: Vandenhoeck & Ruprecht, 1962) and Jakob H. Grønbæk, Die Geschichte vom Aufstieg Davids (1 Sam. 15–2 Sam. 5): Tradition und Komposi tion (AThD 10; Copenhagen: Prostant & Munksgaard, 1971). 11 Tryggve N.D. Mettinger, King and Messiah: The Civil and Sacral Legitimation of the Israelite Kings (CB.OT 8; Lund: CWK Gleerup, 1976), 31. 12 Theodoor Vriezen, for instance, was convinced that Nathan’s son Zabud had written the Succession Narrative; cf. Theodoor C. Vriezen, “De compositie van de Samuëlboeken,” Orientalia Neerlandica (1948): 167–189, esp. 170. 80 klaas smelik criticizing the thesis as a whole. Leonhard Rost seemed to have given the definitive answer in this matter.

Criticism on Rost’s Hypothesis

Notwithstanding the general agreement among scholars sketched above, serious criticism of Rost’s proposals arose in the late sixties, early seven ties of the last century.13 The Amsterdam scholar Martinus Beek formu lated his doubts in a lecture given in 1972.14 We are not dealing with an eyewitness report but with literature, was his opinion.15 It is clear that we are dealing with literature here. The narrator was not an eyewitness to David’s flight and the emotions that it elicited among his faithful ones, any more than Aeschylus was witness to the uproar at the court of Xerxes when he returned from Greece defeated; he did not hear the speech of Hushai anymore than Shakespeare heard the speech of Mark Anthony at the forum of Rome after the murder of Julius Caesar. Historical motifs are perceptible in the background, but above all else what we hear is essentially what narrators and tragic poets are telling us.16 Lienhard Delekat criticized the view that the Succession Narrative was written in majorem gloriam Salomonis—to quote Rost.17 In his article Ten denz und Theologie in der David-Salomo-Erzählung from 1967, this scholar argued that it was just the other way round.18 The author did not wish to extoll David and Solomon—he wanted to show the dark side of their king ship. In this respect, Delekat pointed to the anti-monarchic tendency in the parts on Bathsheba, Absalom and Solomon’s accession to the throne. When Ahithophel suggests that David is to be killed in order to avoid civil

13 A survey in: Randall C. Baily, David in Love and War: The Pursuit of Power in 2 Samuel 10–12 (JSOT.SS 75; Sheffield: Sheffield Academic Press, 1990), chapter 1, and in: Michael Zach, Die Ambivalenz des David-Bildes in II Sam 9–20; I Kön 1+2 (Oldenburg: BIS-Verlag, 2006). 14 First published separately: Martinus A. Beek, David en Absalom (Amsterdam: n.p., 1972). Reprinted in Hier blijven half alle ogenblikken: Keuze uit het werk van M.A. Beek, ed. Lenie van Reijendam-Beek (Baarn: Ten Have, 1988), 69–84. E.T. in Voices from Amsterdam: A Modern Tradition of Reading Biblical Narrative, ed. Martin Kessler (Atlanta, GA: Scholar Press, 1994), 155–168. 15 Ibidem, 1 (69) [E.T. 156]. 16 Ibidem, 11 (76–77) [E.T. 163]. 17 Rost, Überlieferung, 128. [In my opinion, better Latin would have been: ad majorem Salomonis gloriam]. 18 Lienhard Delekat: “Tendenz und Theologie in der David-Salomo-Erzählung,” in Das ferne und nahe Wort: Festschrift Leonard Rost zur Vollendung seines 70. Lebensjahres am 30. November 1966 gewidmet, ed. Fritz Maas (BZAW 105; Berlin: Töpelmann, 1967), 26–36. eyewitness accounts in the book of samuel? 81 war, the author agrees with him—at least according to Delekat. He infers this from the narrator’s qualification of Ahithophel’s counsel as a good advice in the following passage: And Absalom and all the men of Israel said, The counsel of Hushai the Archite is better than the counsel of Ahithophel. For the LORD had appointed to defeat the good counsel of Ahithophel, to the intent that the LORD might bring evil upon Absalom. (2 Sam 17:14) In his study The Story of King David from 1978, David Gunn also opposes the idea that the Succession Narrative was written to justify the accession of David’s throne by Solomon.19 He suggests that the story is, in fact, an examination of power, its use and abuse, and its effects. It is a story, not a piece of propaganda, as suggested before. The author had no political inten tions. The Succession Narrative is “a work of art and entertainment.”20 Rost’s view that the Succession Narrative was written in majorem glo riam Salomonis is indeed improbable. The narrative contains too much that is unfavorable to both David and Solomon, especially in 1 Kings 2:5–46, where David like a biblical Godfather21 orders his son to kill sev eral of his opponents, although he had promised to spare their lives, and where Solomon does not show any mercy to these unfortunate men. He does not even hesitate to let his father’s general Joab be butchered inside the Jerusalem sanctuary. Other scholars, like François Langlamet, Timo Veijola and Ernst Würth wein, have also expressed their doubts about Rost’s appraisal of the author’s intention, although they acknowledge a subsequent pro-Solomonic redac tion of the text.22 In their view, it is far from obvious to characterize the Succession Narrative as a justification of David’s and Solomon’s actions, let alone as a glorification of both rulers.23 Rost suggested that the author

19 David M. Gunn, The Story of King David: Genre and Interpretation (JSOT.SS 6; Shef field: JSOT Press, 1978). 20 Ibidem, 38. 21 Baruch Halpern calls David even a ‘serial killer’ and lists his victims as the ‘Ten Little Indians’; cf. Baruch Halpern, David’s Secret Demons (Grand Rapids, MI: Eerdmans, 2001), 73ff. 22 François Langlamet, “Pour ou contre Salomon? La rédaction prosalomonienne de I Rois, I–II,” RB 83 (1976): 321–379 + 481–528; Timo Veijola, David: Gesammelte Studien zu den Davidüberlieferungen des Alten Testaments (SFEG 52; Göttingen, Vandenhoeck & Ruprecht, 1990); Ernst Würthwein, Die Erzählung von der Thronfolge Davids: Theologische oder politi sche Geschichtsschreibung? (ThS 115; Zürich, Theologischer Verlag, 1974). 23 Pace P. Kyle McCarter, Jr., “‘Plots, True or False’: The Succession Narrative as Court Apologetic,” Interp. 35 (1981): 355–367, esp. 360 n. 12: “Apologetic writing presents unfavorable circumstances forthrightly in order to cast a favorable light on them by a 82 klaas smelik was a court official who wrote in majorem gloriam Salomonis but tyrants usually prefer a less detached minister of information . . . A next point of criticism is the continued search for the author of the Succession Narrative. Eric Seibert states: “Attempts to identify a specific author amount to pure conjecture.”24 And indeed no scholar ever pro posed a convincing identification.25 Moreover, objections have been formulated against the common dat ing of the Succession Narrative in the early years of Solomon’s kingship. Gunn, for instance, gives five arguments against: 1. 2 Samuel 12:20 mentions David going into the “House of the Lord”—an anachronism since the Temple was only built after his death. 2. 2 Samuel 13:8 uses the phrase “in olden days” in referring to princess Tamar’s special tunic;26 this suggests a distance in time. 3. 2 Samuel 18:18 reads: “it is called Absalom’s monument to this day,” another indication that the author was not a contemporary. 4. The narrator’s use of the designations ‘Judah’ and ‘all Israel’ seems to be inconsistent—probably due to the fact that he was writing long after the United Kingdom had ceased to exist. 5. moreover, the style of story-telling implies that at least a few generations have elapsed, perhaps many more.27 Note, however, that Gunn refrains from dating the text because in his opinion there is no “prospect of inspiring any more conviction than does the current uncritical acceptance of near-contemporary authorship.”28 Other scholars are less reluctant and advocate a rather late dating of the Succession Narrative, as we will see later on.

variety of literary means. By its very nature, then, it holds conflicting ideas in literary tension. The elimination of the literary blandishments of the author by appeal to higher critical or other considerations, therefore, will inevitably produce a recital of unfavorable circumstances, but it will also distort the writer’s intended product beyond recovery. It is a mistake to rely heavily on the criterion of narrative tension for identifying redactional material in these stories, when such tension is the very essence of the writer’s technique.” 24 Eric A. Seibert, Subversive Scribes and the Solomonic Narrative: A Rereading of 1 Kings 1–11 (London: T&T Clark, 2006), 103 n. 23. 25 Note that already Rost had his doubts in this regard: “Aber vielleicht ist es überhaupt müßig zu fragen, ob Ebjathar oder Ahimaas die Thronfolgegeschichte verfaßt haben.” 26 Emendated text. 27 See Gunn, Story of King David, 32–33. 28 Ibidem, 33. eyewitness accounts in the book of samuel? 83

Delimitation of the Narrative

The way Rost decided which parts of 2 Samuel originally belonged to the Succession Narrative and which not was determined by his conviction that the succession to David’s throne was the central theme of the account. For this reason, he included even some verses of 2 Samuel 6, although he assigned the main part of this chapter to the Ark Narrative. The actual beginning of the Succession Narrative remained, however, a problem to all scholars who followed Rost. At present, chapter 9 is commonly con sidered to be the opening of the narrative and instead of Rost’s compli cated delimitation listed above, the Succession Narrative is now limited to 2 Samuel 9–20 and 1 Kings 1–2 (as remarked before). But some scholars consider chapters 2–4, dealing with the short reign of Saul’s son Ishboshet, to have belonged to the Succession Narrative as well. Randall Bailey gives even five reasons to do so.29 According to general opinion, the first two chapters of the book of Kings constitute the conclusion of the Succession Narrative. Rost even took these chapters as the starting point for his reconstruction of the whole text. This choice was, however, criticized by scholars who pointed to differences in style and also in the characterization of David.30 In their opinion, it is unlikely that 2 Samuel 9–20 and 1 Kings 1–2 stem from the same hand. But when 1 Kings 1–2 are not a part of the Succession Narrative, the sub ject of the tale cannot be the succession to David’s throne. New themes were therefore suggested.31 The—in my view—most interesting of these propositions is that made by R.A. Carlson in his thought-provoking study David, the Chosen King from 1964.32

29 Cf. Baily, David in Love and War, 14–15. 30 Cf. James W. Flanagan, “Court History or Succession Document? A Study of 2 Samuel 9–20 and 1 Kings 1–2,” JBL 91 (1972): 172–181; Gillian Keyes, The Wages of Sin: A Reappraisal of the ‘Succession Narrative’ (JSOT.SS 221; Sheffield: Sheffield Academic Press, 1996), 54–63; and Henry Wansbrough, “The Finale of the Davidic Succession Narrative,” in Biblical and Near Eastern Essays: Studies in Honour of Kevin J. Cathcart, ed. Carmel McCarthy & John F. Healey (London: T&T Clark, 2004), 37–56. 31 See, for instance, Roger N. Whybray, The Succession Narrative: A Study of II Sam 9–20; I Kings 1 and 2 (SBT 2,9; London: SCM Press, 1968), who sees a connection with wisdom literature. This suggestion is, however, severely criticized by Robert P. Gordon, 1 & 2 Samuel: A Commentary (Exeter: Paternoster Press, 1986), 42–43. 32 R.A. Carlson, David, the Chosen King: A Traditio-Historical Approach to the Second Book of Samuel (Stockholm etc.: Almqvist & Wiksell, 1964). 84 klaas smelik

Compared to Rost and most scholars of his time, Carlson had a com pletely different approach. He applied the traditio-historical approach of biblical criticism, typical of the Scandinavian school. He did not recognize any independent documents within 2 Samuel, thus denying the very exis tence of the Succession Narrative.33 In his view, 2 Samuel is dominated by the contrast between blessing and curse. He describes the first part of the book (2 Sam 2–7) as “David under the blessing” and the second part (2 Sam 9–24) as “David under the curse.” David’s adultery with Bathsheba and the subsequent murder of Uriah turned God’s blessing into a curse. The theme of 2 Samuel is, therefore, the contrast between obedience and disobedience to the Lord, and the consequences of this choice, exempli fied in the figure of king David.

A Literary Approach to 2 Samuel 6

A very serious objection to Rost’ isolation of the Ark Narrative and the Succession Narrative as independent texts is the already mentioned inabil ity to achieve agreement about the limits of these sources. Here, a literary approach to the text can be very helpful. Instead of dividing the given text into various sources following the methods of the historical-critical school, we can try and understand the text as a literary unity. Only when and where this proves to be impossible, source criticism and a division of the text into distinct layers must be applied. We will elaborate on this by focusing on 2 Samuel 6. According to Rost, the chapter is based on two different sources: 2 Samuel 6:1–15; 17–20a belong to the Ark Narrative 2 Samuel 6:16 and 20–23 belong to the Succession Narrative But is this plausible? Are there compelling arguments to divide the chapter in this manner, arguments derived from the text itself? Or is all dependent on the general theory that there was a tale about the Ark and another one about the succession to David’s throne? Let us pay closer attention to this chapter wherein we are told how David brought the Ark from the house of Abinadab to the newly conquered capital Jerusalem. In the first part of the chapter, the narrator is focused on the special nature of the Ark; in

33 See also Serge Frolov, “Succession Narrative: A ‘Document’ or a Phantom?” JBL 121 (2002): 81–104. eyewitness accounts in the book of samuel? 85 the second part, on the conflict between David and his wife Michal, the daughter of king Saul. For this reason, Rost believed that 2 Samuel 6 could be attributed to the two different sources mentioned above but when we pay attention to the occurrence of some special words in the chapter, it becomes clear that both parts of the text are closely interconnected. We give here the text of 2 Samuel 6 in English translation, divided into two parts according to Rost’s delimitation. Interconnected words and expres sions are marked by the same color:

2 Samuel 6 Verses assumed to have been derived from the Ark Narrative

1 Again, David gathered together all the chosen men of Israel, thirty thou sand. 2 And David arose, and went with all the people that were with him from Baale of Judah, to bring up from thence the ark of God, whose name is called by the name of the LORD of hosts that dwelleth between the cherubims. 3 And they set the ark of God upon a new cart, and brought it out of the house of Abinadab that was in Gibeah: and Uzzah and Ahio, the sons of Abinadab, drave the new cart. 4 And they brought it out of the house of Abinadab which was at Gibeah, accompanying the ark of God: and Ahio went before the ark. 5 And David and all the house of Israel played before the LORD on all manner of instruments made of fir wood, even on harps, and on psalteries, and on timbrels, and on cornets, and on cymbals. 6 And when they came to Nachon’s threshingfloor, Uzzah put forth his hand to the ark of God, and took hold of it; for the oxen shook it. 7 And the anger of the LORD was kindled against Uzzah; and God smote him there for his error; and there he died by the ark of God. 8 And David was displeased, because the LORD had made a breach upon Uzzah: and he called the name of the place Perezuzzah to this day. 9 And David was afraid of the LORD that day, and said, How shall the ark of the LORD come to me? 10 So David would not remove the ark of the LORD unto him into the city of David: but David carried it aside into the house of Obededom the Gittite. 11 And the ark of the LORD continued in the house of Obededom the Gittite three months: and the LORD blessed Obededom, and all his house. 12 And it was told king David, saying, The LORD hath blessed the house of Obededom, and all that pertaineth unto him, because of the ark of God. So David went and brought up the ark of God from the house of 86 klaas smelik

Obededom into the city of David with gladness. 13 And it was so, that when they that bare the ark of the LORD had gone six paces, he sacri ficed oxen and fatlings. 14 And David danced before the LORD with all his might; and David was girded with a linen ephod. 15 So David and all the house of Israel brought up the ark of the LORD with shouting, and with the sound of the trumpet. 17 And they brought in the ark of the LORD, and set it in his place, in the midst of the tabernacle that David had pitched for it: and David offered burnt offerings and peace offerings before the LORD. 18 And as soon as David had made an end of offering burnt offerings and peace offerings, he blessed the people in the name of the LORD of hosts. 19 And he dealt among all the people, even among the whole multitude of Israel, as well to the women as men, to everyone a loaf of bread, and a portion of dates, and a cake of raisins. So all the people departed everyone to his house. 20 Then David returned to bless his house.

Verses assumed to have been derived from the Succession Narrative

16 And as the ark of the LORD came into the city of David, Michal the daughter of Saul looked through a window, and saw king David dancing and leaping before the LORD; and she despised him in her heart. 20 [. . .] And Michal the daughter of Saul came out to meet David, and said, How did the king of Israel get him honour today, who uncovered him self today in the eyes of the handmaids of his servants, as one of the vain fellows shamelessly uncovereth himself! 21 And David said unto Michal, It was before the LORD, which chose me before thy father, and before all his house, to appoint me ruler over the people of the LORD, over Israel: therefore will I play before the LORD. 22 And I will yet be more vile than thus, and will be base in mine own sight: and of the maidservants which thou hast spoken of, with them will I get me honour. 23 Therefore Michal the daughter of Saul had no child unto the day of her death.

According to Rost, verse 16 should be isolated from the context. In his view, it stemmed from the Succession Narrative instead of the Ark Narra tive; but when we pay attention to the words used in this verse, we see that they are clearly connected to the preceding and also the following verse 17–19. This contradicts Rost’s division. Moreover, Michal’s reaction in verse 20 presupposes not only verse 16 but also verse 14, that according to Rost belongs to the other source. The eyewitness accounts in the book of samuel? 87 reason for Michal to despise her royal spouse was David’s ritual dance before the Ark, with only a linen ephod as garment, described in verse 14. Without saying this explicitly, the author suggests that the royal genitals were visible when David was leaping around. In Michal’s eyes, this was a very unseemly behavior for a king—and even in our modern times most people would agree with her.34 David’s dance before the Ark was a kind of happening for which tabloids are prepared to kill . . . However, when we separate verse 14 from verse 20, supposing that they stem from two different sources, as suggested by Rost, there is no reason for Michal’s bitter words nor for her reference to the handmaids of David’s servants, who have seen their master uncovered. The reference to the linen ephod as the king’s garment in verse 14 is essential to understand Michal’s furious reaction in verse 20. Else, there was nothing for her to complain about. Therefore, the division of the chapter into two separate sources is impossible without making the end of the story incomprehensible. When we read 2 Samuel 6 as a literary unity, however, the text is not only about the bringing of the Ark inside the new capital but also about the possibility of David begetting an heir by the daughter of his prede cessor Saul.35 In this chapter, there are clear reminiscences to a fertility ritual:36 not only in the dancing and leaping of a semi-nude king but also in David’s blessing of the people and the gifts he presents to them—both men and women!—which are connected with fertility as well.37 More over, the detail that Michal is looking down through a window can be an allusion to the erotic type scene of a woman at a window, not only known from ancient Near Eastern texts but also from the visual arts of the period. These allusions should not be considered to be a description of what really happened in the past, but as literary devices in order to mislead the reader. The suggestion is made that after the conquest of Jeru salem and the arrival of the Ark in the city, it is time for David to unite

34 While it seems obvious that Michal is very angry because her husband has shown his genitals publicly to other women, most scholars do not consider this to be the real reason for her indignation; cf. Telling Queen Michal’s Story: An Experiment in Comparative Interpretation, ed. David J. Clines & Tamara C. Eskenazi (JSOT.SS 119; Sheffield: Sheffield Academic Press, 1991). In this way, these colleagues show that they have no idea of Near Eastern female jealousy. Michal is not an English lady keeping up appearances . . . 35 See my more elaborate analysis of the chapter in Klaas A.D. Smelik, “De intocht van de Ark in Jeruzalem,” ACEBT 4 (1983): 26–36. 36 Cf. J.R. Porter, “The Interpretation of 2 Samuel VI and Psalm CXXXII,” JThS 5 (1954): 161–173. 37 Cf. Hos 3:1; Song 2:5. See also Klaas A.D. Smelik, Converting the Past: Studies in Ancient Israelite and Moabite Historiography (OTS 28; Leiden: Brill, 1992), 52 n. 92. 88 klaas smelik the house of Saul and his own house by begetting a son, who will be at the same time David’s son and Saul’s grandson. But the surprising outcome of the story is that Saul’s daughter Michal will not be the queen mother of Israel—she remains childless to her dying day. In rebuking David, she resembled her father Saul and therefore she shares his fate—the house of Saul will never become an eternal dynasty like the house of David. This unexpected ending raises the question who will be queen mother instead. A question that will be dealt with in the sequel of 2 Samuel and in the first two chapters of the book of Kings. This literary approach to 2 Samuel 6 is completely different from the analysis of Bible texts as practiced by contemporary German scholars as Walter Dietrich and Thilo Rudnig.38 They too question Rost’s reconstruc tion of the Succession Narrative but in an opposite way. According to them, there was no Succession Narrative as a literary unit, in the way Rost proposed. Having isolated some passages dating from David’s time, they concentrate on the many editorial layers in the text. The thesis of a liter ary unit is completely abandoned as is evident from the title Dietrich gave to his contribution: “Das Ende der Thronfolgegeschichte.”39 Dietrich points to the many interconnections between the Succes sion Narrative and another source isolated by Rost: the History of David’s Rise. For this reason, he questions Rost’s reconstruction and I concur with Dietrich in this respect. But are these interconnections the result of editorial reworking of existing sources, as Dietrich suggests, or are they indications that we are dealing here with a more extensive literary unit stretching from 1 Samuel 15 to 1 Kings 2? The problem with redaction criticism is that it resembles the well- known story of the Sorcerer’s Apprentice, who was in the end confronted with an ever increasing number of brooms: the number of editorial layers in a Bible text similarly increases with each redaction-critical study that is published. In this respect, John Van Seters makes the following thought- provoking remark:

38 See Walter Dietrich, “Das Ende der Thronfolgegeschichte,” in Die Sogenannte Thron folgegeschichte Davids: Neue Einsichten und Anfragen, ed. Albert de Pury & Thomas Römer (OBO 176; Freiburg: Universitätsverlag, 2000), 38–69, and Thilo Alexander Rudnig, Davids Thron: Redaktionskritische Studien zur Geschichte von der Thronnachfolge Davids (BZAW 358; Berlin: Walter de Gruyter, 2006). 39 Mark that Jeremy Hutton, The Transjordanian Palimpsest: The Overwritten Texts of Personal Exile and Transformation in the Deuteronomistic History (BZAW 396; Berlin & New York: De Gruyter, 2009) also questions the literary unity of the Succession Narrative and similar texts, but he pleads for a very early date of the various sources he discerns. eyewitness accounts in the book of samuel? 89

The consequences of the analytic method of redaction criticism, as reflected in Dietrich and others, can be seen in the recent work of Thilo Rudnig, in which, in my view, the number of editors and revisers has proliferated to the point of absurdity.40 There is also another possibility. That is to return to Noth’s original view of the Deuteronomist as author instead of redactor or various redactors, but extending his authorship to passages considered by Noth as derived from older sources like the Succession Narrative. In my dissertation Saul from 1977,41 I already defended this approach, and Frolov does the same in his already quoted article: It looks increasingly probable that its subdivision into fragmented narra tive substrate and Deuteronomistic redactional layer(s) is not sufficiently warranted. This calls for a serious consideration of the possibility that the Former Prophets originated as a large, continuous, and integral Deuterono mistic composition.42

Historicity of the Account

We turn now to the historicity of the account. Our first observation is that a literary approach to the books of Samuel does not necessarily infer that the historical reliability of the Bible text is questioned,43 as witnessed by Jan Fokkelman’s point of view in his elaborate study of the Succession Nar rative (called by him: David’s sin and its consequences). He states: Furthermore, I see it as being highly probable that the portrait of David as it is given to us here by a very spiritual and detached person is an accu rate portrayal of the historical David. [. . .] I presume that the David of the narrative, however fictionally portrayed, is not fictitious. It does not seem

40 John Van Seters, The Biblical Saga of King David (Winona Lake, IN: Eisenbrauns, 2009), 32. 41 Klaas A.D. Smelik, Saul: De voorstelling van Israëls eerste koning in de Masoretische tekst van het Oude Testament (Amsterdam: P.E.T., 1977). 42 Frolov, “Succession Narrative,” 104. It is remarkable that Frolov does not understand the importance of 2 Samuel 10–12 for the composition as a whole. He even suggests that these chapters could be a later addition: “Despite being fully integrated into the main story line, chs. 10–12 are not essential for the coherence of the plot” (103). This is not true. Carlson rightly underlines that David’s affair with Bathsheba described in 2 Samuel 11 is essential, because it constitutes the turning point in the whole story about David, as men tioned above. 43 See also the remark by Barton, “Dating the ‘Succession Narrative’,” 97: “[. . .] treating the text primarily as literature does not in itself entail a late dating [. . .].” 90 klaas smelik

conceivable that a writer of ancient Israel would have entirely fabricated this king in the way a modern writer can.44 Actual proof for this assertion is, however, lacking. This is rather unsatisfy ing. Is there no way to combine a literary and a historical approach in a more objective way? I believe there is.45 In my already mentioned dissertation, I have followed four ways in order to ascertain the historical value of various parts of the books of Samuel: – The epigraphic evidence from the Land of Israel – The genre of Ancient Near Eastern historiography – Archaeological evidence – Literary analysis of the Bible text. Using this combined method, we can avoid circular reasoning. Only when these four approaches point in the same direction, can we be certain that our historical conclusions will be correct. When applied to 2 Samuel, we see that: – the epigraphic evidence from the Land of Israel suggests that literacy in Israel and Judah in the tenth century bce was very limited. It is unlikely that in this period literary masterpieces such as the Succession Narrative could have been produced. – Ancient Near Eastern historiography is very different in style and interest compared to the biblical narratives about the past. Only the Mesopota mian genre of historical fiction resembles the nature of biblical stories to a certain degree. Passages comparable to Ancient Near Eastern histo riography can be found in the Bible, showing that the genre was known and practiced in Ancient Israel. However, these passages are completely different in style compared to biblical narrative. The list of David’s heroes in 2 Sam 23:8–39, for instance, resembles Ancient Near Eastern histori ography far better than the story about David’s affair with Bathsheba or that of Absalom’s revolt. – the archaeological evidence from the tenth century bce shows a great discrepancy with the representation of this period in the Hebrew Bible. Jerusalem, for instance, was in reality a very small town and not the beau tiful city praised in the book of Psalms. John Van Seters writes: “We now know with a high degree of confidence that the sociohistorical context in

44 Jan P. Fokkelman, Narrative Art and Poetry in the Books of Samuel: A Full Interpreta tion Based on Stylistic and Structural Analyses, Volume 1: King David (II Sam. 9–20 & I Kings 1–2) (SSN 20; Assen: Van Gorcum, 1981), 424. 45 See also Klaas A.D. Smelik, “The Use and Misuse of the Hebrew Bible as a Historical Source”, in The Rediscovery of the Hebrew Bible, ed. J.W. Dyk (ACEBT.SS 1; Maastricht: Shaker Publishing, 1999), 121–140. eyewitness accounts in the book of samuel? 91

the Court History of David simply cannot be supported by the archaeo logical for the 10th century and must belong to a much later age.”46 – literary analysis of the books of Samuel shows that the common identifi cation of various sources or redactional layers in the text is less probable than usually assumed.47 Moreover, a literary approach gives us the oppor tunity to study the narrative art of the authors, their literary techniques and the implications of their way of writing for the historical value of their texts. They used their literary skill to influence the reader. Their intent was not to write history—as supposed by Gerhard von Rad48—nor to divert—as supposed by Gunn—but to convince, as we will see in the concluding part of this contribution.

Eyewitness Account or Narrator’s Skill

In regard to the historical value of this part of the Bible, an important point of consideration is the general assumption that the genre of the Succession Narrative is that of an eyewitness account. Whybray describes this remarkable agreement among scholars as follows: There is almost universal agreement that the author was a contemporary, or near-contemporary, of David and a member of the court, who was therefore in an excellent position to write an authentic history of the reign.49 But what is meant by the word ‘eyewitness’? An eyewitness is a person present at an event who gives a report on what he or she has seen. But in 2 Sam 15:9 we read: And Amnon said, Have out all men from me. And they went out all men from him. Therefore, the dramatic description of Tamar’s rape by Amnon cannot stem from an eyewitness—unless we assume that either Amnon or Tamar

46 Van Seters, The Biblical Saga of King David, xii. See also Israel Finkelstein & Neil Asher Silberman, David and Solomon: In Search of the Bible’s Sacred Kings and the Roots of the Western Tradition (New York: Free Press, 2006). 47 For a critical discussion of the literary approach to the books of Samuel see, Greger Andersson, Untamable Texts: Literary Studies and Narrative Theory in the Books of Samuel (LHB/OTS 514; New York & London: T&T Clark, 2009). 48 Gerhard von Rad, “Der Anfang des Geschichtsschreibung im alten Israel,” Archiv für Kulturgeschichte 32 (1944): 1–42; reprinted in Gerhard von Rad, Gesammelte Studien zum Alten Testament (ThB 8; Munich: Kaiser Verlag, 19714), 148–188. E.T. “The Beginnings of His torical Writing in Ancient Israel,” in Gerhard von Rad, The Problem of the Hexateuch and Other Essays (Edinburgh: Oliver & Boyd, 1966), 166–204. 49 Whybray, Succession Narrative, 11. 92 klaas smelik wrote the Succession Narrative. This is impossible: Amnon was killed by Absalom two years later and Tamar lived in seclusion—moreover, women in Israel and Judah were illiterate in this period. When we have ruled out the possibility that an eyewitness wrote the Succession Narrative, is it possible that we are dealing here with a hearsay witness, a person who testifies what someone else said or wrote? This is actually what scholars mean, when they are referring to an eyewitness account. When they suggest that for instance Abiathar was the author of the Succession Narrative, they do not suppose that the priest had been lying under Amnon’s bed with an ostracon and a pen in his hand to make notes of the conversation between sister and brother. They assume that Abiathar collected his information from people connected to David’s court. They only use the designation ‘eyewitness account’ in order to stress the historical reliability of the Succession Narrative. But even a hearsay wit ness could not have rendered the words spoken between Amnon and his sister, since there was no one else present, as remarked by the narrator himself. This also applies to other parts of the Succession Narrative where we find private conversations—for instance when David tries to persuade Uriah to sleep with his wife Bathsheba. And who had access to the secret correspondence between David and general Joab, when the king ordered him by letter to have Uriah killed?50 In an article published in 1981, Peter Ackroyd warned us: [. . .] hence, because of the intimacy of the details—some of the crucial incidents involve a bedroom scene of one kind or another—the authorship has been sought among the persons directly involved or sufficiently close to the main actors in the drama to be able to give a precise account of what Nathan really said to David, or of what Amnon said to Tamar. But such an approach is naïve: The detail of conversations in narratives like these is to be attributed to the narrator’s skill and not to precise information.51 In this respect, two questions arise: – if it is obvious that the Succession Narrative is not an eyewitness account, why have scholars assumed that it was? What made them believe that the story was written by an eyewitness?

50 2 Samuel 12:14–15: “And it came to pass in the morning, that David wrote a letter to Joab, and sent it by the hand of Uriah. And he wrote in the letter, saying, Set ye Uriah in the forefront of the hottest battle, and retire ye from him, that he may be smitten, and die.” 51 Peter R. Ackroyd, “The Succession Narrative (so-called),” Interp. 35 (1981): 383–396, esp. 385. eyewitness accounts in the book of samuel? 93

– if not eyewitness accounts, to what genre texts belong tales like the Ark Narrative and the Succession Narrative—assuming that they have ever existed as separate entities? In regard to the first question, we have to take into account that most of these scholars were theologians who were inclined to stress the histori cal trustworthiness of the Bible. Their interest was focused on retrieving reliable historical information even from late Bible texts like the books of Chronicles. When it was not too obvious that we are dealing with fiction, as is the case with the book of Judith, they were inclined to believe that what is described in the text is historically reliable. This explains partly for what Ackroyd called their naïve approach. On the other hand, the author of the Succession Narrative has used all his literary skill to give the reader the impression that we are witnesses of the events described. The combination of these two factors resulted in the assumption that the Succession Narrative was written by a contemporary who had access to court secrets. Note, however, that Rost himself reck oned with the possibility that the Succession Narrative was only a fanciful creation of the author (in German: “bloßes Spiel der Phantasie”), but he considered this less probable than the assumption that the author is pro viding us real historical facts (“wirkliche Geschichtstatsachen”).52 These scholars did not take into account the historical impossibility that this kind of developed literature was written already in the tenth cen tury bce; they rather praised the uniqueness of the narrative considering the time of its origin. For many biblical scholars, the author of the Succes sion Narrative replaced Herodotus as the “Father of History.”53 However, the reason for this earlier generation of scholars to consider the Succes sion Narrative to be the beginning of modern historiography, turned out for a later generation to be the definitive proof that the account was of a much later date. John Van Seters54 even suggests that the Succession Nar rative was composed in the late Persian period as a satirical response to the Deuteronomistic Historian’s positive account of David’s reign. He sub mits that the Sitz im Leben of the narrative is the mercenary dominated

52 Cf. Rost, Überlieferung, 126. See also Otto Eissfeldt, Einleitung in das Alte Testament (Tübingen: Mohr, 19643), 187: “Ein einfacher Augenzeugenbericht liegt ja in II Sam 13–20 + 1 Kön 1–2 keinesfalls vor, sondern eine von hoher Erzählungskunst gestaltete und ausge schmückte Darstellung, die immerhin einiges von einem—guten !—historischen Roman an sich hat.” 53 Cf. Gordon, 1 & 2 Samuel, 41. 54 Van Seters, The Biblical Saga of King David. 94 klaas smelik martial culture of the Near East in the late fifth and early fourth century bce. According to Van Seters, the Deuteronomistic Historian did not insert the Succession Narrative into his work, as assumed by most scholars. It was the other way round: the author of the Succession Narrative added his text to the Deuteronomistic History.55 Also in this respect, a literary analysis can be helpful in order to decide the matter. In the example I will give now, narrative analysis is combined with another fruitful method in the field of modern biblical studies, to wit intertextuality. Intertextuality is the shaping of the meaning of a text by other texts. It can include an author’s borrowing and transformation of a prior text but also a reader’s referencing of one text in reading another. In this case, I mean with ‘intertextuality’ that one Bible text can be con nected with another in such a way that the meaning of the latter text shapes our understanding of the former. Actually, we see this method already used in rabbinic literature. For instance, when in Isa 36:6 the Assyrian envoy quotes an expression that also occurs in the book of Ezekiel (19:6), the latter passage offers a clue to the hidden meaning of this part of the book of Isaiah, so often misunderstood. By connecting both passages, the one in Isaiah and the one in Ezekiel, the reader understands what the author of Isaiah intended to convey. He wanted to warn the reader of what I have named “distortion of true prophecy.”56 The Assyrian envoy is quoting an authentic prophetic expression from the book of Ezekiel but in another sense than the prophet had meant. His aim is to delude the Judahite king Hezekiah by misusing Ezekiel’s words. Of course, this is historically impossible: the prophet lived a hundred years later than the Assyrian envoy, but in biblical narrative this is no problem. In the same way, the question of dating the narratives in the books of Samuel can be solved by the method of intertextuality. This time, we do not focus upon the Succession Narrative but on the related Ark Narrative. We remember that this tale was, according to Rost and followers, even older than the Succession Narrative. We turn now to 1 Samuel 4, a chapter in which the capture of the Ark by the Philistine army in the time of Eli, the priest of Shiloh, is the main theme.57

55 His view is criticized by Barton, “Dating the ‘Succession Narrative’,” 103. 56 See Smelik, Converting the Past, 93–128. 57 For a more elaborate analysis of the biblical text ibidem, 35–58. eyewitness accounts in the book of samuel? 95

At the end of the chapter, the reader is told twice that Eli’s daughter-in- law laments: “The Glory has gone into exile from Israel!”58 Her words refer to the preceding part of the narrative. They are an allusion to the loss of the Ark, which was captured by the Philistines after their glorious victory. But why does this dying woman designate the Ark as “the Glory” and why to go into exile,” instead of—for instance—a“ ,גלה does she use the verb to take,” a verb used four times in this chapter?59 Her“ ,לקח niph’al of exclamation seems out of context—for us readers a clue to search for a parallel elsewhere. We can find the original context of this expression in the book of Eze kiel as well. Chapters 8–11 describe how the Glory of the God of Israel first leaves the Temple and afterwards Jerusalem on the eve of the capture of is כבוד the city by the Babylonians in 586 bce.60 The same Hebrew word used. When we read twice in 1 Samuel 4: “The Glory has gone into exile from Israel!” we, the readers, are supposed to connect these words with the central theme in the book of Ezekiel: Exile and Restoration. There fore, we have a definite clue for dating the Ark Narrative: it reflects a 6th- century bce context.61 In this way, the approach of intertextuality can be helpful for dating passages by means of other Bible texts.

Eyewitness Account or Engaged Literature

However, intertextuality can do more, as will become clear, when we now turn to our second question: – if not eyewitness accounts, to what genre texts belong tales like the Ark Narrative and the Succession Narrative—assuming that they have ever existed as separate entities? In Rost’s view, the Ark Narrative was a cult legend, written by a priest in order to explain to the visitors to the Ark Sanctuary in Jerusalem how the Ark had arrived at this spot. But this does not account for the special usage of the designation ‘the Glory’ in this text nor for the use of the verb

58 See 1 Sam 4:21–22. 59 In verses 11, 17, 19, and 21. 60 See also Peter R. Ackroyd, The First Book of Samuel (CBC; Cambridge: Cambridge U.P., 1971), 52. Cf. however Philippe De Robert, “Le gloire en exil: Réflexions sur 1 Samuel 4, 19–22,” RHPR 59 (1979): 351–356, especially 351–352, who argues that Ezekiel 8–11 is an echo of 1 Samuel 4:19–22 (that is just the other way round). 61 See Smelik, Converting the Past, 55. 96 klaas smelik

The peculiar choice for these two expressions must have a special .גלה reason and historical context, and again the book of Ezekiel provides a key for the solution of our problem. When we combine both texts, the capture of the Ark by the Philistines turns out to be an allusion to the pil lage and destruction of the Temple of Jerusalem by the Babylonian army, the lost Ark being comparable with the holy vessels that are taken away out of the Temple, and the Philistines with the Babylonians. In the same way as the holy vessels were brought to Babylon, the Ark went into exile on foreign soil. There is more. In Jeremiah 7, the inhabitants of Jerusalem are warned no to trust in the doctrine of the inviolability of the Temple propagated by the Jerusalem priesthood. The Lord is prepared to have his own sanctuary destroyed by the enemy, when his people do not mend their evil ways. In order to prove this, the reader is reminded of the destruction of the Lord’s sanctuary at Shiloh. Inspired by this text, the author of 1 Samuel 4 wrote a narrative in which the Israelites put their faith in the presence of the Ark when solemnly carried from the Shiloh sanctuary onto the battle field, with the surprising outcome that God thwarts their expectations of a secure victory by having the Ark captured by the enemy. It is up to the readers to understand that this narrative is not about the past; it is an allusion to their present situation, in which God acted in the same way with the Temple of Jerusalem and its holy vessels as He did with the Ark in 1 Samuel 4. Trust ye not in lying words, saying, The temple of the LORD, the temple of the LORD, the temple of the LORD, is this. [. . .] Behold, ye trust in lying words, that cannot profit. [. . .] But go ye now unto my place which was in Shiloh, where I set my name at the first, and see what I did to it for the wickedness of my people Israel. [. . .] Therefore will I do unto this house, which is called by my name, wherein ye trust, and unto the place which I gave to you and to your fathers, as I have done to Shiloh. And I will cast you out of my sight, as I have cast out all your brethren, even the whole seed of Ephraim. (Jer 7:4, 8, 12 & 14–15)62

62 It is clear from this verse that the destruction of the Shiloh sanctuary did not take place in Samuel and Eli’s days but later, during the Assyrian campaign against Israel eyewitness accounts in the book of samuel? 97

In the same way, the Succession Narrative is not an eyewitness account, written to justify Solomon’s accession to the throne, but rather a tale in which the reader is confronted with the tragic fate of a pious king who had committed a deadly sin. Carlson was right: what Rost called die Thronfolgegeschichte, is actually a narrative about the contrast between David blessed and David cursed. Therefore, we are not dealing here with historiography, we are not dealing here with literature intended to divert the reader, we are dealing here with engaged literature:63 literature con veying a message. These kinds of texts can be used as historical sources for the period in which they were written, but they have much less importance for the reconstruction of the historical events in the period described. The historical account of David’s reign cannot be based on 2 Samuel alone without confronting serious problems.

(Ephraim) at the end of the eight century bce, as is also suggested by the results of the excavations at this site. Cf. Smelik, Converting the Past, 56. Contra John Day, “The Destruc tion of the Shiloh Sanctuary and Jeremiah vii 12, 14,” in Studies in the Historical Books of the Old Testament (VT.S 30), ed. John Adney Emerton (Leiden: Brill, 1979), 87–94. 63 In German: Tendenzliteratur, in French: Littérature engagée.

A Battle of Wits and Words: Hushai, Ahithophel and the Absalom Rebellion (2 Samuel 16–17)1

Robert P. Gordon

Abstract

The narrative of the verbal contest between Ahithophel and Hushai in 2 Sam 16–17 is unusually rich in figurative language and more generally noteworthy for the rhetorical strategies by which each of these attempts to influence Absalom at a crucial point in his rebellion against his father. This verbal duel is described in such a way as to suggest that the contest between David and Absalom is settled even before the respective armies engage in battle. The setting in Absalom’s war council has unsurprising features of a royal council. Both Ahithophel and Hushai are in the business of ‘suasion’, and figures of speech would be appropriate to them as members of ‘the wise’ and as royal counsellors. At the same time, Ahithophel’s terser contributions are especially reminiscent of the oracular response, as would be expected from a strict reading of 16:23. Hushai’s generous use of figures of speech makes for narrative retardation, and serves his plan to dissuade Absalom from making an immediate strike against David. This may even be reflected secondarily in his gentilic ‘the Arkite’—‘the lengthener’. There is no indication that Hushai’s plan, though accepted, was fully implemented: this actually becomes less important once he has bought time for David. Redactional layering of the narrative into strands that make Hushai respectively spy and rival counsellor to Ahithophel is judged unnecessary, partly on the ground that David’s prayer in 15:31, that God would ‘make foolish’ the counsel of Ahithophel, is answered when Hushai prevails over his rival. Other aspects of the narrative that are discussed include the relevance of the (mainly) divine council type-scene featuring the question ‘What shall we do?’ (16:20), and the possibility that, in 16:23, the text has been modified in a couple of respects, whether at the authorial or scribal stage, to the disadvantage of Ahithophel.

In the 1980s, I worked on the books of Samuel, producing a short guide to those books in 1984 and a middlebrow commentary in 1986. At the then time of writing, few passages in the fine literary mosaic that is 1–2 Samuel impressed or amused me as did the story of Ahithophel and (especially) Hushai. I still find it amusing when I imagine Hushai spinning out his

1 I am most grateful to my colleagues in Antwerp and Ghent for their kind invitation to participate in the conference whose proceedings are presented in this volume. 100 robert p. gordon similes and metaphors, glancing anxiously at a nearby sundial, and drawl- ing out his syllables in order to save precious minutes for David, to let him get to the Jordan and away from the threat of attack by Absalom and his followers. I hereby fulfil a longstanding promise to myself to return to this story, in order to savour some of its features away from the larger duties of commentary and introduction writing. The story of Absalom’s rebellion, as various writers have noted, begins with a gate and ends in a gate. Absalom presents himself to disgrun- tled Israelites as the one who will listen to their grievances and repair his father’s neglect of this “duty of the gate”. He stations himself by the city gate in order to intercept the early morning arrivals from outlying places, and he tells them that the king has not deputed anyone to listen to their just causes. “And so Absalom stole the hearts of the people of Israel” (2 Sam 15:6; cf. verse 13). The end of the rebellion is marked by David’s taking his seat in another gate—possibly the gate of Mahanaim—where he reviews his victorious troops (19:9[8]). Thereafter, he “won over the hearts of all the men of Judah as though they were one man” (19:15[14]).

Verbal Combat

In this paper, I focus on the roles of Hushai and Ahithophel during Absa- lom’s rebellion, and especially on their contributions in the setting of Absalom’s war council (2 Samuel 16–17). From the moment Hushai meets Absalom, he begins to weave a wordy web. Having presumably left off his mourning attire and washed his hair (see 15:32), he greets Absalom with “Long live the king! Long live the king!” (16:16). As has often been noted, there is ambiguity here. The king could be Absalom or David. However, in a comparable situation of disputed sovereignty, at the time of David’s death, people said “Long live King Adonijah!” (1 Kgs 1:25) and “Long live King Solomon!” (1 Kgs 1:34, 39). Instead of “Long live King Absalom!”, however, Hushai voices a double acclamation—the only one in the Hebrew Bible. He has smothered the absence of Absalom’s name in a (doubtless) noisy repetition of the coronation formula. The Septuagint omits the second occurrence of the formula, but the MT, rather than accidentally repeating itself, may be making a deliberate point by its repetition.2

2 P. Kyle McCarter, II Samuel: A New Translation with Introduction, Notes and Com mentary (AB 9; Garden City, NY: Doubleday, 1984), 380, follows the Greek, but comments: “One could argue for haplography in LXX or, as assumed here, dittography in MT.” Baruch a battle of wits and words 101

With ambiguities of this sort, Hushai insinuates himself into Absalom’s circle of advisers and, as the narrator would have it, the contest between the supporters of David and Absalom is going to be settled before ever they meet on the battlefield. The counterposed advice of Ahithophel and Hushai marks the spot where the struggle between Absalom and David is decided. This is a case of single combat. The story is told with exquisite skill, and many bouquets have come its way. Jan Fokkelman, for example, speaks of the narrator’s “dialectical mastery” in handling the competing contributions of Ahithophel and Hushai,3 while John Van Seters says that 17:1–14 is “a brilliant piece of wisdom narrative and illustrates the difference between wise counsel and foolish flattery”.4 Much has been made of the fact that this is a contest between two of Israel’s “wise”, though, strictly, the words “wise” and “wisdom” do not occur in the narrative. On the other hand, the related terms “counsel” and “counsellor” do occur (15:12; 16:23; 17:7, 11, 15[x2], 21).5 More is made of the counsellor role of Ahithophel, but Hushai as David’s “Friend” (16:16) also becomes a counsellor to Absalom, regardless of what exactly a “King’s Friend” might normally be expected to be or to do. The conceptual closeness of “wisdom” and “counsel” is not in question, and is evident in a statement such as “wisdom is with those who take counsel” (Prov 13:10).

Absalom’s Council

Absalom is a rebel and not a king, and even his rebel status appears not to have lasted for very long, yet 16:20–17:14—irrespective of what exactly is implied by “all the elders of Israel” (17:4) and “all the people of Israel” (17:14)—deserves to be included in the frugal material in the Hebrew Bible relating to the “royal council”, and it is here that I shall begin. The most

Halpern, David’s Secret Demons: Messiah, Murderer, Traitor, King (Grand Rapids, MI: Eerd mans, 2001), 45, suggests that the double acclamation adverts to the fact that there are two “kings” in Israel at this point. 3 Jan P. Fokkelman, Narrative Art and Poetry in the Books of Samuel. I. King David (II Sam. 9–20 & I Kings 1–2) (SSN 20; Assen: Van Gorcum, 1981), 220–221. 4 John B. Van Seters, The Biblical Saga of King David (Winona Lake, IN: Eisenbrauns, 2009), 312. 5 Meir Sternberg, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading (Bloomington, IN: Indiana University Press, 1985), 344–345, notes that in the Hebrew Bible agents who turn out to be sagacious are rarely so described by the narrator. See also R. Norman Whybray, The Succession Narrative: A Study of II Samuel 9–20; I Kings 1 and 2 (SBT 2/9; London: SCM, 1968), 57–58. 102 robert p. gordon obvious texts for the council are 1 Kings 12 and 22,6 and both feature the presentation of rival advice. One of the sayings repeated in Proverbs is “In the multitude of counsellors, there is safety” (11:14; 24:6; cf. 15:22). Stated as a general principle, it has obvious relevance to proceedings of the royal council. In each of the two Kings narratives, a monarch is not satisfied with the advice that he is given, and he therefore turns to alternative sources of guidance. In 1 Kings 12, Rehoboam consults with his father’s advisers on how to govern the complaining northerners led by Jeroboam, but he rejects their counsel and puts faith in his own circle of friends (verses 13–14); in 1 Kings 22, Jehoshaphat of Judah is not persuaded by four hundred prophets predicting the recovery of Ramoth-gilead and asks for a prophet of YHWH to be brought (verse 7). In both instances, the seeking of a second opinion seems natural and unexceptionable. We should note, moreover, that there are sufficient points of correspondence between the story of Rehoboam and his counsellors and Absalom’s war council to suggest to Van Seters that there is a “direct literary relationship” between the two narratives.7 As Absalom listens to Ahithophel’s wise counsel, he is convinced by its good sense, but he is out of his depth and he summons Hushai for an independent opinion. Apparently, Hushai has been excluded from the original meeting of the war council, and is summoned only after Ahithophel has spoken. The plural in “Give (pl.) [us] your advice” (16:20) has been interpreted as addressing Ahithophel and Hushai, and more specifically the former, rather than the members of Absalom’s council.8 In the light of 17:5, however, it is difficult to imagine that Hushai was present during Ahithophel’s speech, which indeed Absalom summarizes for him (17:6) before he makes his own proposal. The narrator credits the decision to call Hushai and its consequences to the overruling of YHWH (17:14), but the mere act of consulting Hushai for confirmation or disconfirmation of Ahithophel’s advice should not of itself cause surprise in the light of the other narratives, or of what would be expected to happen in royal councils throughout the Ancient Near East.9 From a narratival standpoint, courtly practice simply assists the divine purpose.

6 Esth 1:13–21, describing Ahasuerus’s consultation of his wise men on the problem of Vashti, also qualifies. 7 The Biblical Saga, 314. 8 Cf. Jonathan Grossman, “The Design of the ‘Dual Causality’ Principle in the Narrative of Absalom’s Rebellion,” Bibl 88 (2007): 558–566 (562). 9 Pace Keith Bodner, David Observed: A King in the Eyes of his Court (HBM 5; Sheffield: Sheffield Phoenix Press, 2005), 133, who suggests that Absalom “inexplicably” summons Hushai after there has been general agreement about the wisdom of Ahithophel’s counsel. a battle of wits and words 103

The Art of Suasion

Ahithophel and Hushai are in the business of “suasion”. As Christopher Begg notes, Josephus has David in trepidation, lest Ahithophel prove too persuasive (πείσειεν) in what he recommends to Absalom.10 Hushai’s task is to frustrate the incisive and, for David, potentially disastrous recom- mendations of Ahithophel. It is a commonplace, as we have noted, to view both protagonists as presenting their cases in the manner of the wise, using figures of speech as natural expressions of their wisdom and erudition. In a nearby text, the wise woman of Tekoa, in the course of persuad- ing David to reinstate Absalom after his murder of Amnon, talks about the quenching of her coal (2 Sam 14:7; NRSV “my one remaining ember”) and reminds the king that “We are like water spilled on the ground, which cannot be gathered up” (verse 14). Royal counsellors, as being among the “wise”, had a predilection for simile.11 Figurative speech is a feature of the royal council in the couple of narratives that we have already noted. When Rehoboam calls his advisers together, the wise old heads give their answer with brevity and good sense worthy of an Ahithophel: “If you will become a servant to these people today and will serve them and give them a favourable answer, then they will always be your servants” (1 Kgs 12:7). Contrariwise, the younger set advise: “Tell them . . . My little finger is thicker than my father’s waist . . . My father scourged you with whips, but I will scourge you with scorpions” (1 Kgs 12:10–11). Again, in 1 Kings 22, when the king of Israel seeks divine counsel, the response of the four hundred prophets is “Go, for the Lord will give it (sc. Ramoth-gilead) into the king’s hand” (verse 6; cf. verse 12). When, however, Micaiah is called in and sarcastically repeats the salvation oracle of the other prophets (verse 15), he then changes his tone and tack: “I saw all Israel scattered on the hills like sheep without a shepherd, and YHWH said, ‘These people have no master. Let each one go home in peace’” (verse 17). As to whether Ahithophel actually speaks in the manner expected of the wise counsellor, much depends on what exactly he says in 17:3. The MT has something of the order of “I will bring back all the people to you. Like the return of the whole (is) the man you are seeking: all the people will be at peace.” The Septuagint offers an alternative reading that

10 Christopher T. Begg, “David’s Flight from Jerusalem According to Josephus,” Herv. Teol. Stud. 62 (2006): 1–22 (9). See Josephus, Ant. 7.202. 11 Cf. Robert Polzin, David and the Deuteronomist: A Literary Study of the Deuteronomic History. Part Three. 2 Samuel (ISBL; Bloomington, IN: Indiana University Press, 1993), 171. 104 robert p. gordon has frequently been accepted as superior to the MT and as being capable of explaining it: “I will bring back all the people to you as when a bride returns to her husband; you are seeking the life of but one man, and all the people will be at peace.” If the alternative reading is accepted, there is the added lure of a simile that seems appropriate to the context. Israel is viewed as a bride and Absalom is the husband whom she has lately deserted. That the image of bride and husband would not exactly fit the relationship between the people and their upstart king is probably beside the point: Hushai flatters Absalom by disguising the true situation. Adjudication between the two texts remains difficult, and I am not so confident about the superiority of the Greek as I was in the early 1980s.12 Ahithophel is in a hurry, and perhaps our narrator is clever enough to reflect this in his clipped Hebrew. On the assumption of the superiority of the Septuagint Vorlage, Samuel R. Driver attempted to show how the MT came about, by a combination of accidental omission and rearranging of letters, but his explanation is not wholly convincing.13 Caution is certainly advisable when weighing the Greek of this verse. The freedom exercised by the Greek translator of Proverbs in dealing with proverbs is well-known, and we may have an instance of the same in this very chapter when—in addition to the MT’s comparison of David to “a bear robbed of her whelps in the field” (KJV) in verse 8—the Greek has “and like a wild sow in the plain.” Kyle McCarter is one of the few to regard the addition as original,14 but many have joined Julius Well- hausen in being less confident.15 If this is the case in verse 8, we should not be too readily persuaded of the originality of the Greek’s bride simile in verse 3. The point is especially important when comparing the speeches of Ahithophel and Hushai, particularly in view of Song-Mi Suzie Park’s observation that the text appears to pile the figures of speech on Hushai and leave Ahithophel to unadorned prose.16 For if we were to conclude that the Greek reading is superior, then even Ahithophel’s brief speech in 17:1–3 would contain a simile and, since Hushai’s oration is several times

12 Robert P. Gordon, I & II Samuel: A Commentary (Exeter: Paternoster, 1986 [2004]), 280. 13 See Samuel Rolles Driver, Notes on the Hebrew Text and the Topography of the Books of Samuel (Oxford: Clarendon, 19132), 320–321. 14 McCarter, II Samuel, 382. 15 Julius Wellhausen, Der Text der Bücher Samuelis (Göttingen: Vandenhoeck & Rup recht, 1871), 200, noting that in Hebrew thinking pigs were associated with uncleanness, rather than ferocity. 16 Song-Mi Suzie Park, “The Frustration of Wisdom: Wisdom, Counsel, and Divine Will in 2 Samuel 17:1–23,” JBL 128 (2009): 453–468 (457–458). a battle of wits and words 105 longer,17 there would be no significant difference stylistically or statisti- cally between the two. On balance, I am inclined to accept the MT as expressing an intelligible sense appropriate to the occasion; the Septu- agint figure of the disloyal bride remains a powerful alternative whose claim to originality cannot altogether be dismissed. The statement in 16:23 comparing the counsel of Ahithophel with making inquiry of YHWH is sometimes loosely interpreted as according it the status of prophetic utterance. However, the expression used in the MT points to consultation by oracular means more usually associated with priestly mediation (e.g., 1 Sam 14:37; 22:10; 28:6). Most of the occurrences of this expression, moreover, have to do with seeking direction about the timing or manner of engaging in battle, precisely as in 17:1–14. This has relevance for our evaluation of Ahithophel’s language when he gives his advice to Absalom. In prophecy, there is regular use of simile and metaphor—a feature that is also well represented in Mesopotamian prophecy18—whereas oracular responses are brief and unadorned. At their simplest, oracular responses consist in “Yes” or “No” answers, though in most cases the biblical narratives flesh them out in short sentences.19 A typical answer is “Go, attack the Philistines and save Keilah” (1 Sam 23:2).20 Ahithophel’s speeches exemplify something of this oracular terseness. To Absalom’s “What shall we do?” (16:20) Ahithophel gives a one-sentence answer that tells the rebel prince what he should do and what will be the consequences (verse 21). Moreover, in neither of his contributions, in 16:21 and 17:1–3, is there a suggestion of courtly—at least in the sense of deferential—language.21 Insofar as it features in the story of Ahithophel and Hushai, it is Hushai who uses it, in 16:16–19 (and 17:11b?).

Filibuster and Frustration

Hushai’s objective is the buying of time for David, and his use of figurative language is crucial in this regard. This comes on the heels of a slowed down account of the evacuation of Jerusalem. Van Seters has well remarked that

17 See Robert Alter, The Art of Biblical Narrative (New York: Basic Books, 1981), 74. 18 E.g., the recurrent “beneath straw, water runs” in ARM 26 197 13–14; 199 44; 202 10–11. 19 Cf. Judg 1:1–2; 20:23, 28; 1 Sam 23:2; 30:8; 2 Sam 5:19. 20 Judg 1:2 has an example of a more expansive response: “Judah is to go; I have given the land into their hands.” 21 See Charles Conroy, Absalom Absalom!: Narrative and Language in 2 Sam. 13–20 (AnBib 81; Rome: Biblical Institute Press, 1978), 140. 106 robert p. gordon the evacuation of the city proceeds “at a snail’s pace”, because the narrator wants to introduce a number of scenes and characters each of which provides a sidelight on David’s reign.22 And Van Seters understandably wonders why David is climbing the Mount of Olives when he is supposed to be making for the Jordan.23 Then there also seems to be deliberate narrative retardation leading up to the presentation of Hushai’s counsel: Ahithophel advises Absalom to lie with his father’s concubines, which advice Absalom acts upon; there follows the summary statement about Ahithophel’s reputation as a counsellor; next comes his second address to the war council; finally, Hushai is allowed to make his case.24 When given the floor, he confects a wordy text out of similes and metaphors on a scale scarce paralleled in Old Testament narrative. Even the pace of his speech is made to serve his purpose of delaying Absalom’s pursuit of David. As we have noted, at one point in the speech the Septuagint or its Vorlage appears to get carried away, adding “and like a wild sow in the plain” in verse 8, and so contributing to the element of retardation. Other features of Hushai’s speech also make for prolongation: instances of infini- tive absolute with finite verb, repetitions of terms, and similar. Fokkelman notes about 15 of what he calls “duplications” within Hushai’s speech.25 In this connection, we might even wonder whether Hushai’s designation as “the Arkite” acquires significance. A couple of writers have suggested that ,”quickly”) in 17:16 makes play on the name “Hushai“) מהרה the use of :(”hasten, hurry“) חוש since it easily lends itself to association with BH “Now send quickly and tell David, ‘Do not spend the night at the fords are synonyms, and wordplay חוש Pi.) and) מהר in the desert . . .’”26 BH is possible, even if not certain. Similarly, Hushai the “Arkite” lengthens out his speech in a way that would make it hard for an attentive Hebrew be long”; Hi. “lengthen”): Hushai“) ארך hearer or reader not to think of BH is the “lengthener” who uses his speech as a delaying tactic on behalf of the fugitive king whom he actually supports.27 By contrast, when Hushai

22 Van Seters, The Biblical Saga, 308. 23 Van Seters, The Biblical Saga, 311–312. 24 Cf. Conroy, Absalom Absalom!, 47 n. 9. 25 Fokkelman, Narrative Art, 221 n. 28. 26 See Robert Alter, The Art of Biblical Narrative (London: George Allen and Unwin, Hebr.) (Ramat-Gan: Revivim, 1987), 71) מדרשי שמות במקרא ,Moshe Garsiel ;206–205 ,(1981 (E.T. Biblical Names: A Literary Study of Midrashic Derivations and Puns [Ramat Gan: Bar- Ilan University Press, 1991], 105). Garsiel also suggests that in the scene involving Ittai in coming (אתה) you” [sg.]): “Why are you“) אתה with”) and“) את there is play on BH 22–15:19 .([A Literary Study, 219 =] 144 ,מדרשי שמות) (verse 19) ”?(אתנו) with us .(”[to produce Aρχι (ἑταῖρος) (“Chief [Friend הארכי The Septuagint graecized MT 27 a battle of wits and words 107 reports on the war council to the priests Zadok and Abiathar, his message is summarized in two occurrences of “thus and thus”: “thus and thus did Ahithophel counsel Absalom and the elders of Israel and thus and thus have I (Hushai) counselled” (17:15). This is, of course, the narrator insert- ing his summary expressions into Hushai’s speech, even though Hushai manifestly must have gone into some detail about what he had heard when with Absalom.28 The contrast with the wordy Hushai of the earlier verses is strong. The acceptance of Hushai’s proposal by the war council and the victory of David and his men within a short span of narration time might easily lead to the assumption that Hushai’s plan is carried through. The message that Hushai sends to David certainly does not make that assumption (17:15–16). It reaches the king in the form “Set out and cross the water immediately, for Ahithophel has advised such (and such) against you” (17:21). It is needed, not least in view of David’s puzzling procession up the Mount of Olives, for he seems to be clearer about the urgency of fleeing the capital than he is about how he should proceed thereafter. Neither in Hushai’s message “as originally given” nor in its reported form in 17:21 is anything being left to chance. As far as the battle account itself is concerned, a number of writers have concluded that Hushai’s advice was not implemented, and certainly not au pied de la lettre. Even the statement in 17:24 that Absalom “crossed the Jordan with all the men of Israel” does not imply the assembling of all Israel “from Dan to Beersheba”, as advised by Hushai (17:11), for an equally comprehensive expression is already used of those accompanying Absalom in 16:15 (“all the people the men of Israel” [KJV]), before ever he consulted his war council. What surely has happened is that the battle of words having been regarded as decisive, the question of the full implementation of Hushai’s plan has become irrelevant. While the statement that Ahithophel saw that his advice had not been followed (17:23) does not require that Hushai’s was accepted, sufficient uncertainty is created by Hushai’s speech for David to be able to recover his situation, and for Ahithophel to assume the worst and, at some point or other, to remove himself from the possibility of any further part in the proceedings.

28 See Sternberg, The Poetics of Biblical Narrative, 385, on the narrator’s tendency to flaunt his contempt for mechanical repetition by interpolating his summary within the direct speech of his characters. 108 robert p. gordon

This suspicion of a mismatch between Hushai’s successful intervention in Absalom’s council and the account of the battle in chapter 18 has encouraged a few scholars to press the redactional button and claim that the whole of Hushai’s speech in 17:7–13, together with 17:14, which includes a summary statement about God’s frustration of Ahithophel’s counsel, plus the last clause in 17:15 and the report of Ahithophel’s suicide in 17:23, is secondary. Thus pared down, the story tells simply of Ahithophel’s per- tinent counsel and its acceptance by “Absalom and all the elders of Israel” (17:4). On this view, Hushai’s role is simply to insinuate himself into Absa- lom’s circle and act as David’s informant. However, the redactional approach is leaden-footed, and fails to reckon with the fact that Ahithophel’s plan was not acted upon. If the original story told simply of unanimous agreement in Absalom’s camp that Ahithophel should attack “tonight” (17:1), it provides no explanation for its non- implementation, whereas the MT version featuring Hushai does exactly that. The redactional approach simply shows that if you remove the right pieces from a narrative, you can produce a different sort of story. It also requires excisions within the passage that introduces Hushai in 15:30–37, and especially the mention of David’s prayer in 15:31, which has its counterpart in the theological statement in 17:14b—rejected ex hypothesi— about God’s frustrating of Ahithophel’s counsel. As we have noted, in ([.Hi] פרר) the reconstructed original the role of Hushai is to frustrate the counsel of Ahithophel simply by spying and informing. He is to pass on whatever he hears from within Absalom’s war council to the priests Zadok and Abiathar, who in turn will relay the information to their sons and thence to David. However, it is unlikely that Hushai’s role is is indeed used of the nullifying פרר limited to espionage activity. The verb effects of intelligence passed on, as in Neh 4:9 (15): “When our enemies heard that we were aware, and that God had frustrated their plan . . . ,” while in Ezra 4:5 the “people of the land” hire counsellors in order to frustrate the plans of the Judean builders, but there is more to consider. Hushai meets David at or near the summit of the Mount of Olives “where God was worshipped” (15:32) and just after the king had voiced the prayer “O LORD, turn the counsel of Ahithophel into foolishness” (15:31). At this point, the narrative cries out that Hushai is the answer to Dav- id’s prayer. Providence has intervened in this meeting between the king and his trusted friend.29 So what about the nullification of Ahithophel’s

29 See Grossman, “The Design,” 560. a battle of wits and words 109 counsel? On the one hand, nothing in the proposal that David puts to Hushai requires anything more than intelligence-gathering: he outlines how information given to the priests who are remaining in Jerusalem can be relayed to him. This appears to be the means of neutralizing, or “frustrating”, Ahithophel’s advice. But what should we make of David’s This ?(סכל־נא) prayer about the turning of Ahithophel’s counsel into folly is hardly satisfied by espionage. Are we not to see in Hushai’s tour de force the answer to the prayer, as Hushai egregiously spins a rope from sand, serving up a hilarious mix of earnest strategy and self-contradictory non- sense that commends itself to all but his rival? It is a “making foolish” in this sense that the theological statement in 17:14b picks up: “YHWH had commanded to frustrate the good counsel of Ahithophel, so that YHWH could bring harm upon Absalom.” I have already noted that the accounts of royal councils in 1 Kings 12 and 22 involve rival counsels, just as happens in the MT version of Absalom’s war council. There is the further parallel that in all three narratives the outcome is attributed directly to divine intervention: “this turn of events was from YHWH, to fulfil the word YHWH had spoken to Jeroboam son of Nebat through Ahijah the Shilonite” (1 Kgs 12:15; cf. “this is my doing”, verse 24); “YHWH has put a lying spirit in the mouths of all these prophets of yours. YHWH has decreed disaster for you” (1 Kgs 22:23). The suspect theological statement in 2 Sam 17:14b is just as integral to the narrative in which it appears: “YHWH had commanded to frustrate the good counsel of Ahithophel, so that YHWH could bring harm upon Absalom.”

Personal Factors

On the purely human plane, Hushai is faced with an immense task if he is to negate Ahithophel’s counsel. The account of Ahithophel’s proposal in 2 Sam 17:1–3 issues in the statement that Absalom and all the elders of Israel were convinced (17:4). Hushai has, therefore, to try to overthrow what is virtually a done deed. He is an outsider and he is liable to suspicion because of his prior association with David. “Call also Hushai the Arkite and let us hear what he has to say”, says Absalom (17:5), and the use of the gentilic may be drawing attention to the fact that Hushai is not one of Absalom’s intimates—he is “not one of us”. Perhaps more significant still is Absalom’s reference to “what is in his mouth” (lit.), often rendered in English versions by “what he has to say” (NIV, REB, NRSV, ESV). When this expression is used elsewhere, it can refer to speech that has been put in the speaker’s mouth by someone else, sometimes YHWH 110 robert p. gordon

(e.g., Num 23:12, 16) and sometimes another human (e.g., Exod 4:15). More particularly, in the story of the wise woman of Tekoa it is said that Joab put words in her mouth (2 Sam 14:3, 19) so that she could act out her part and trick David into welcoming Absalom back to court. We are entitled to ask, then, whether in Absalom’s use of this expression he indicates his suspicion that Hushai has been embedded (as they say) in order to act on David’s behalf. Are we entitled, indeed, to ask whether the Absalom of our narrative recalls Joab’s arrangement described just two or three chapters earlier? If so, his decision to hear Hushai out already hints at the irratio- nality that 17:14b attributes to the intervention of YHWH. Hushai’s use of metaphor and simile creates an impression of wisdom and sagacity that helps him outmanoeuvre his rival. The individual figures of speech have been examined from time to time and their composite effect has been admired, so that I do not propose further discussion of that sort. There are differences between the approaches of Ahithophel and Hushai, nevertheless, that are deserving of brief comment. Ahithophel’s speech in 17:1–3 is heavy on the first-person singular: he uses seven verbs with himself as subject.30 The implication of this is not entirely clear. It is often assumed that he is putting himself forward as leader of the attack on David—though there is no evidence that he has military experience— and is thus enhancing his own position in the new regime. Fokkelman, however, sees nothing that is particularly sinister: “Ahithophel wants to achieve the final elimination of the old king personally, possibly also in order not to expose the successor to the risk of waging war at night.”31 Oth- ers are inclined to think that Ahithophel means simply, “If I were in your place, here is what I would do.”32 Bodner notes the commissive element in Ahithophel’s statement “committing himself to undertake a program”.33 He thinks that Ahithophel is “executing an individually minded agenda”,34 since he believes that Ahithophel is settling an old score. Bodner takes further than most the idea that, because Ahithophel was the grandfather of Bathsheba, he was motivated by revenge for David’s seduction of his granddaughter.35 The fact that Absalom abused his father’s concubines

30 Conroy, Absalom Absalom!, 64 n. 83. 31 Fokkelman, Narrative Art, 212. 32 So André Caquot and Philippe de Robert, Les Livres de Samuel (CAT 6; Geneva: Labor et Fides, 1994), 537 (“à ta place, voici ce que je ferais”). See also John Mauchline, 1 and 2 Samuel (NCB; London: Oliphants, 1971), 279, citing GKC § 108f. 33 Bodner, David Observed, 132. 34 Ibid., 132. 35 Ibid., 129–130. a battle of wits and words 111 on the palace roof is taken as calculated riposte to David’s seduction of Bathsheba, whom he first noticed while he was walking on that same roof (11:2//16:22).36 And while Ahithophel himself does not explicitly mention the palace roof in 16:21, the next verse describing what happened certainly does. Otherwise, we should interpret the roof as the obvious place that would answer to the terms of Nathan’s judgment pronouncement: “in broad daylight before all Israel” (12:12). Since Ahithophel’s relationship to Bathsheba is never directly mentioned in the text, it is difficult to be certain. That Bathsheba is the daughter of Eliam is stated in 2 Sam 11:3, and that Eliam is the son of Ahithophel in 2 Sam 23:34. There is, therefore, a danger of psychologizing the narrative beyond what it can bear. It is some narrative time since the “Bathsheba episode”, and we should be required to think of Ahithophel acting as David’s counsellor, and similarly revered by Absalom, during a longish period of disgruntlement over the treatment of his granddaughter. On the other hand, one possible interpretation of Ahithophel’s having to be summoned from Giloh would be his prior alienation from David. It just remains to note that, if Bodner is correct, and Ahithophel’s head is still full of a perceived injustice to Bath- sheba, a revisit of the textual crux in 17:3 would be justified. A metaphor that portrays David as interloper in a marriage would come easily to a man so scandalized by what David had done to Bathsheba’s marriage that, vengefully, he instigated the abuse of David’s concubines by the king’s own son. When evaluating Ahithophel’s egocentric contribution, we should also take into account the council type-scene, better represented in descriptions of the divine council, in which volunteers are sought to act on behalf of the deity or deities. The relevance of the divine council to the present discussion depends, of course, upon the assumption that this concept is in some degree modelled on earthly realities experienced in the royal council. There are examples both in the Hebrew Bible and outside it. In these, there is a question, and sometimes a response in the first person singular: “Whom shall I send?” / “Here I am” (Isa 6:8); “Who will entice Ahab into attack- ing Ramoth-gilead?” / “I will entice him” (1 Kgs 22:20–21); “Which of the gods will slay Zu?” (Myth of Zu 9);37 “Who among [the gods can remove] the illness?” / “I myself will act the craftsman” (KTU 1.16 verse 10–27)38—these

36 Ibid., 130. 37 See ANET, 111. 38 See ANET, 148; Nicholas Wyatt, Religious Texts from Ugarit: The Words of Ilimilku and his Colleagues (BSS 53; Sheffield: Sheffield Academic Press, 1998), 235. 112 robert p. gordon corresponding to Absalom’s “What shall we do?” (2 Sam 16:20) and Ahithophel’s response with “I will/would . . .” (2 Sam 17:1–3). Whatever the precise significance of Ahithophel’s use of the first person in 17:1–3, Hushai seems to score a point when he plays to Absalom’s nar- cissistic tendencies, talking about all Israel being gathered to Absalom and 39.(17:11 ;קרב) assuring him of victory, if his “face” or “person” goes into battle Fokkelman40 and André Caquot and Philippe de Robert41 follow the MT in reading “into battle”, but others see the word as an Aramaism and perhaps too late to be admitted here, preferring “in the midst”.42 As the MT stands, an “Aramaism” would presumably be a plank for those who think that Hushai’s speech is a later interpolation into the story. And if Hushai could be shown to be of non-Israelite origin,43 we might even have to think of an appropriate ethnolect being put in his mouth, though that is a remote possibility. Most writers follow the Septuagint B text in reading “in their midst”, which, if the intended sense, raises a point. For why should Hushai suggest that Absalom be only “in the midst of” his vast army and not at their head? This is hardly flattering, as the Targum seems to appreciate, reading “at the head of us all” (most Sperber Mss) or “at our head”.

In Conclusion

Absalom, who wanted to be recognized as ruler and judge administer- ing justice in the kingdom (15:4), makes the wrong choice at the crucial moment because he is fated to do so. He falls for Hushai’s flattery, though not necessarily because Hushai’s figurations sound like divine speech.44 There may even be a case for regarding the oracular concision of Ahithophel as more “divine” than the figurative confection served up by Hushai. Oracular speech tends to be direct and concise, as we have seen. And oracular speech is especially appropriate in this story, since the situation described is about inquiry before battle. The statement in 17:14 that “YHWH had commanded to frustrate the good counsel of Ahithophel, so that YHWH could bring harm upon Absalom” is integral to the account

39 Sternberg, The Poetics of Biblical Narrative, 361 (where Sternberg also comments that by the time of the battle “Absalom is little more than a pretty face”). 40 Fokkelman, Narrative Art, 218. 41 Caquot and de Robert, Les Livres de Samuel, 520. 42 So Driver, Notes on the Hebrew Text, 322. 43 See McCarter, II Samuel, 371–372. 44 Park, “The Frustration of Wisdom,” 463. a battle of wits and words 113 of Absalom’s rebellion and not one of a series of later additions. This verse resonated from the beginning with David’s statement in 15:34 that Hushai could help by frustrating Ahithophel’s counsel. It emphatically attributes everything to do with the downfall of Absalom to YHWH, who is twice mentioned in a slightly overloaded sentence: “YHWH had commanded . . . so that YHWH could bring harm”. And since Ahithophel is the agent for the fulfilment of Nathan’s judgment oracle on David (see 2 Sam 12:11–12) as surely as Hushai is instrumental in rescuing David from the designs of Absalom, it is difficult to view the Absalom narrative as a warning to kings on the dangers of relying on “counsel” in preference to the divine word mediated through prophecy.45 In this narrative, it is not the king proper who seeks the counsel, while, for his part, David is under prophetically announced judgment and therefore unable to act under prophetic auspices. At the same time, it is through the operation of wisdom—Hushai’s “wisdom”—that Absalom is undone.

דבר Addendum on the euphemizing use of

There are hints in MT 2 Sam 16:23 of disapproval, whether authorial or scribal, of Ahithophel. If we follow the qere, his advice was reckoned “as if a man inquired of God”. The kethiv, however, says that it was “as if he (or “one”) inquired of God”, leaving open the possibility that Ahithophel is the inquirer.46 Given the potential difference in emphasis between the two readings, the qere may represent a deliberate refocusing of the MT. Driver correctly notes that the meaning of the but it is still possible that the qere is ,איש qere is available without the insertion of specifically intended to block off the possibility that Ahithophel is the subject of “inquired”: “as if [Ahithophel] inquired of God”.47 This prompts a further observation of a similar sort. The MT talks about inquiring of “the word of God”, which rep- is שאל where ,בדבר ask, inquire”) and“) שאל resents a unique combination of BH דבר used for oracular consultation of God. Since there are other instances where appears to have a euphemizing “buffer” function, so that, for example, “rejecting/

45 Pace Robert Polzin, “Curses and Kings: A Reading of 2 Samuel 15–16,” in The New Literary Criticism and the Hebrew Bible, ed. J. Cheryl Exum and David J.A. Clines (JSOT.SS 143; Sheffield: JSOT Press, 1993), 225; idem, David and the Deuteronomist, 173–178. In David and the Deuteronomist, Polzin summarizes: “Kingship contaminates Israel, royal counsel contaminates divine inquiry . . .” (177). 46 Cf. Grossman, “The Design,” 561–562, who comments on the difference between the qere and the kethîv, with the latter reflecting a situation in which Ahithophel is the one ask ing for divine guidance. Absalom in theory has access to the priestly oracular apparatus, since Zadok and Abiathar do not accompany David to the Jordan. Their support for David (15:24–29), however, probably compromises them in Absalom’s eyes as far as any cultic functioning on his behalf is concerned. 47 Driver, Notes on the Hebrew Text, 320. 114 robert p. gordon despising YHWH” becomes “rejecting/despising the word of YHWH” (see the table below), it is possible that a still earlier (i.e., pre-qere) intervention in the text has already taken place, again to the disadvantage of Ahithophel.48

2 Sam 12:9 mt why have you despised the word of YHWH? 4Q51 – LXX L+ why have you despised the LORD?49

2 Sam 12:1450 (דבר) mt you have despised the enemies of YHWH in this matter (דבר) 4Q5151 you have despised the word of YHWH in this matter LXX B (L) you provoked the enemies of the LORD in this matter

1 Kgs 22:5 mt seek first the word of YHWH LXX inquire52 of the LORD today

// 2 Chr 18:4 MT seek first the word of YHWH LXX seek the LORD today LXX be2 seek the word of the LORD today

2 Kgs 1:16 mt to inquire of his word LXX* –

דבר A fuller discussion would consider whether even “early” euphemizing use of 48 should be regarded as original or secondary, and so also with the kind of material noted in the Addendum. 49 On the originality of the shorter reading, see Carmel McCarthy, The Tiqqune Sopherim and Other Theological Corrections in the Masoretic Text of the Old Testament (OBO 36; Freiburg: Universitätsverlag and Göttingen: Vandenhoeck & Ruprecht, 1981), 205. 50 The textual fluidity in this verse is now widely interpreted as an indication that the original form of the Hebrew was “you have despised YHWH”. The 4Q51 version (“you have .is especially inelegant (”דבר of YHWH in this דבר despised the 51 See Frank Moore Cross et al., eds., Qumran Cave 4: XII: 1–2 Samuel (DJD XVII; Oxford: Clarendon, 2005), 143; Eugene Ulrich, ed., The Biblical Qumran Scrolls: Transcriptions and Textual Variants (VT.S 134; Leiden: Brill, 2010), 303. 52 L “Let us seek.” See Natalio Fernández Marcos and José Ramón Busto Saiz, El Texto Antioqueno de la Biblia Griega: II: 1–2 Reyes (TECC 53; Madrid: Instituto de Filología del CSIC, 1992), 74. The Intentional Use of Polysemy: (Judg 3:19) דבר סתר A Case Study of

Karolien Vermeulen

Abstract

In Judg 3:19, Ehud talks himself into king Eglon’s chambers by saying, “I have a hidden/covered word/thing, matter’) for you.” The double polysemy‘) דבר סתר in this phrase has led many scholars to interpret this in terms of moral judgment of the character. An untruthful man uses devious language, while a hero’s words are a skillful tool. The language (linguistic) is but a symptom of the character’s intentions. The current contribution addresses this symptomatic approach by analyzing the mechanisms of polysemy in the text on the one hand and its effect on the interpreter on the other hand. It will argue that the ambiguity plays a double role. First, Ehud’s words serve the story, as they enable the character to accomplish the impossible: murder the king. The polysemy has a well-defined narratological function, creating and consequently filling a gap in the story world. Secondly and more importantly, the words of Ehud provide the exegete with a tool to interpret the story in whatever framework he or she prefers. Various and opposite readings of the passage demonstrate that the gap also functions outside the narrative in the interpreter’s world. Thus, polysemy works inside the narrative’s borders and beyond, giving the one disambiguating the benefit of the doubt in both instances.

But let’s return for the moment to purely strategic speech acts, where the aim is to subvert, obscure, or bypass understanding. In cases of lying, deception, and other manipulative speech acts, there are two levels of communication, one carefully hidden (the intention of the communicator, the real aim of the speech act) and the surface, or apparent, speech act. Language is used strategically for deceptive purposes when these two levels are kept strictly apart, and the listener is fooled by the speech act’s appearance as a communicative act. Strange as this may seem, the strategic use of language for deceptive purposes is only possible, it only works, because the primary purpose of language use is to generate understanding. Communication for the purpose of understanding is, and is taken to be, the communicative norm. Only because mutual understanding is the essence of communication is it possible for someone to successfully mislead people through the use of language. It would be impossible (or very, very difficult) to lie or deceive by means of speech acts if lying and deceiving were generally understood to be the primary aims of communication: Who would be fooled by deceptive 116 karolien vermeulen

speech acts when everyone knew speech acts were always employed to lie and manipulate?1 In Judg 3:19, an example of double polysemy occurs that has been the subject of debate for a long time among translators and commentators of the Hebrew Bible.2 Ehud, an Israelite judge, addresses Eglon, the Moabite :in itself part of the larger phrase דבר סתר king, with the ambiguous words for you, O King.” The polysemy דבר סתר I have a“ ,דבר סתר לי אלך המלך has been called “the protagonist’s [i.e., Ehud’s] ‘special ability’,” the result of “his ability of improvisation,” an “effective piece of dramatic irony” as well as part of the judge’s “Odysseus-like, Huck Finn-like [. . .] resources to survive.”3 Yet simultaneously, the words have been read as “an obvious pretense, since Ehud has no particular religious qualifications, and in any event Ehud worships an entirely different god” and “a major problem that concerned [. . .] any reader or interpreter of the Bible [. . .] to justify the obvious trickery employed by Ehud in assassinating Eglon, the king of Moab.”4 Whereas most interpreters use the ambiguity in Judg 3:19 to make moral claims about Ehud—he is either bad or good, the words rarely have been discussed as an utterance fooling both the audience inside and outside the story. The symptomatic approach of the language and its role in the story as found in the majority of older studies have prevented scholars from seeing the polysemy’s other effects, effects that surpass the literary, the scatological, and the humorous, considered “somewhat coarse” by the modern reader.5 This essay argues that the ambiguity—initially a character’s tool—became an interpreter’s tool that generates many, even opposite readings. By exploring the linguistic given of polysemy, the story creates possibilities for all audiences inside and outside the textual world.

1 Jerome E. Bickenbach & Jacqueline M. Davies, Good Reasons for Better Arguments: An Introduction to the Skills and Values of Critical Thinking (Peterborough: Broadview Press, 1997), 17. 2 For a historical overview, see David Gunn, Judges through the Centuries (Malden: Blackwell Publishing, 2005), 34–52. 3 Quotes are taken from resp. Eric Christianson, “A Fistful of Shekels: Scrutinizing Ehud’s Entertaining Violence (Judges 3:12–30),” Interp. 11,1 (2003): 53–78 (61); Yairah Amit, The Book of Judges: The Art of Editing (BIS 38; Leiden: Brill, 1999), 172; Robert Alter, The Art of Biblical Narrative (New York: Basic Books, 1981), 40; Susan Niditch, War in the Hebrew Bible: A Study in the Ethics of Violence (New York: Oxford University Press, 1993), 119. 4 Quotes are taken from resp. Geoffrey Miller, “Verbal Feud in the Hebrew Bible: Judges 3:12–30 and 19–21,” JANES 55,2 (1996): 105–117 (115); Louis Feldman, “Ehud,” in Louis Feld man, ed., Studies in Josephus’ Rewritten Bible (Leiden: Brill, 1998), 137–152 (141). 5 J. Alberto Soggin, Judges: A Commentary (OTL; Philadelphia: Westminster, 1981), 53. the intentional use of polysemy 117

Yet, very often these options have not been perceived as such due to the reader’s limiting perspective. Contrary to the common assumption that it is ambiguity that confuses people, it is the mentioned perspective that leads them astray. To clarify my point, I will first touch on the Hebrew text and its challenges to the reader. Both lexical-semantic, that is to say, the polysemy on the word level, and narratological issues, that is to say, the polysemy embedded in the text, will be addressed. Then, I will discuss several views on the verse as reflected in various interpretations. Special attention will be paid to the role attributed to the polysemy based wordplay in the interpretation. I will include some reflections on the various translations of the passage in an appendix. In conclusion, possible functions will be proposed for the polysemy of the phrase under analysis.

Hidden Hebrew Things

The Polysemy of the Hebrew Phrase The story of Ehud and Eglon in Judges 3 takes a new turn, when Ehud invites himself over to Eglon’s private chambers with the following words: I have a dĕbar sēter for you, O King.”6 As scholars“ ,דבר סתר לי אליך המלך ,דבר have pointed out, the Hebrew text provides a double polysemy with ,both “secret” and “covering.”7 As such ,סתר both “thing” and “word,” and ”,being “word,” “message דבר Ehud has a secret message for the king, with the advice of“ ,דבר בלעם :advice,” as attested among others in Num 31:16“ דבר :a vain word;” and in Esth 1:12“ ,דבר שפתים :Balaam;” in 2 Kgs 18:20 stands for “matter,” Ehud is alluding to דבר the Royal order.” If“ ,המלך the event of the murder, still secret at that moment. Similar uses of the the matter of Uriah,” or Exod“ ,דבר אוריהו :noun can be found in 1 Kgs 15:5 every case of sin.” A third possibility consists of Ehud“ ,כל דבר פשע :22:8 :such as in Qoh 1:10 ,דבר referring to the hidden sword by using the word anything“ ,כל דבר אשר בארץ :there is a thing,” or in Judg 18:10“ ,יש דבר

6 Judg 3:19. 7 The double polysemy has been noted by John Hamlin, At Risk in the Promised Land: A Commentary on the Book of Judges (ITC; Grand Rapids, MI: William B. Eerdmans Publishing see among ,דבר Co., 1990), 74–75. For a discussion on the ambiguity created by the word others Emil Kraeling, “Difficulties in the Story of Ehud,” JBL 54,4 (1935): 205–210; Geoffrey Miller, “Verbal Feud in the Hebrew Bible,” 105–117; Eric Christianson, “A Fistful of Shekels, 53–78; J. Clinton McCann, Judges (IBC; Louisville, KY: John Knox Press, 2002), 44; and Lil lian Klein, The Triumph of Irony in the Book of Judges (Sheffield: The Almond Press, 1988), .סתר .and HALOT II, 772, s.v דבר .See also HALOT I, 211–212, s.v .39–37 118 karolien vermeulen

,is of a more palpable and concrete nature דבר ,on earth.” In these cases being a “thing,” as is the sword.8 leads to additional readings, that stress סתר Exploring the polysemy of the covering as in “a thing of covering” or “a word of covering.” The former refers to the sword wrapped by Eglon’s fat, when it will pierce him.9 In case of the latter, it hints at the message as an annunciation of Eglon’s death realized in his flesh enclosing the dagger.10 Support can be found in 1 Sam 25:20 where a covering of mountains occurs, or in Job 24:15 with as “secret” occurs in Prov סתר a concealing of the face.11 The reading of a gift in“ ,מתן סתר :secret conversation,” and in Prov 25:23“ ,לשון סתר :21:14 secret.” Notice that it is used once in combination with speech and once with a material object. For the story of Judges 3, this results in the supplementary readings of “a secret message” and “a secret thing.” Obviously, the interpretation of the “secret message” can still take different concrete forms. Scholars have suggested among others messages of a political, private, sexual, and religious nature.12 In case of the “secret thing” reading, a

girded“ ,ויחגר אותה מתחת למדיו על ירך ימינו The sword is mentioned in Judg 3:16 as 8 under his cloak on the right thigh.” וישלח אהוד את יד שמאלו ויקח את החרב מעל ירך ימינו ויתקעה בבטנו :Judg 3:21–22 9 And Ehud send out his left hand and took the sword from“ ]. . .[ ויסגר החלב בעד הלהב his right thigh and he drove it into his belly (Eglon’s) [. . .] and the fat enclosed the blade.” Notice that ‘fat’ can be taken from a medical point of view as the internal fat, or can describe a chubby, even obese person. While most commentators choose the latter, Law חלב and בריא son Stone does not find this reading supported by the use of the words elsewhere. Although I do not agree with his final conclusion—“Ehud appears not as a duplicitous trickster but as a courageous, risk-taking man armed with a custom weapon undertaking a dangerous mission,” his analysis shows how traditional readings gain credit, even if they conflict with basic things as philology (Lawson Stone, “Eglon’s Belly and Ehud’s Blade: A Reconsideration,” JBL 128,4 [2010]: 649–663 [663]). ”.A thing of God I have for you“ ,דבר אלהים לי אליך :Judg 3:20 10 ”;And she came done under the cover of the mountain“ ,וירדת בסתר ההר :Sam 25:20 1 11 Also the eye of the“ ,ועין נאף שמרה נשף לאמר לא תשורני עין וסתר פנים ישים :Job 24:15 adulterer watches the twilight, saying, ‘No eye will see me,’ and he put on a covering of the face.” 12 For a political interpretation, see among others Robert Boling, who calls Ehud’s “piece of single-handed diplomatic treachery” an effective way of attaining peace for eighty years (Judges: Introduction, Translation and Commentary [AB; Garden City, NY: Doubleday & Company, 1975], 84–88 [88]). According to Marc Zvi Brettler, a reading of the chapter should take into account the political setting in order to understand it historically. It is in light of this that “the purpose of these devices within a context that describes a battle between Moab and Israel” must be considered (The Creation of History in Ancient Israel [London/ New York: Routledge, 1995], 83–87 [83]). In addition, see also Edward Bloom & Lillian Bloom, Satire’s Persuasive Voice (Ithaca, NY: Cornell University Press, 1979), 218–221 and Yehezkel Kaufmann, The Book of Judges: A Commentary (Jerusalem: Kiryat Sefer, 1962), 108 [in Hebrew]. For a reading as private encounter, see among others Boling’s translation: “I have confidential message for you, O King” ( Judges, 84). For a sexual-ironic interpretations, the intentional use of polysemy 119 disparity arises between Ehud’s intention—the thing is the sword—and Eglon’s interpretation—the thing is anything but a sword (unless one would adhere to a euphemistic interpretation of the English ‘sword’ here, which is not my intention as such). In short, the Hebrew expression in Judg 3:19 carries multiple meanings, each of them illuminating one aspect of the broader picture. Whereas such a word study may give insight in the semantic range of a word or words, it is in the context of the story that the double polysemy becomes interesting for interpretation and that it proves its true functionality. It is one thing to throw in a couple of polysemous words; it is a totally different thing to deplete its contribution to the narrative.

The Functionality of Polysemy in the Story their ,מושיע Ehud has been introduced as the leader of the Israelites, their “savior,” who will free them from the Moabites (Judg 3:15). Yet, after this introduction, his role seems to be restricted to bringing tribute to the king of the oppressors, a servant like task. It is said that Ehud carries a weapon, but a little later he walks out of the room without having done anything (Judg 3:16). This raises the assumption that the left-handedness of Ehud might be a bad thing, even though some scholars have argued that left-handed warriors formed some kind of elite troops.13 In the story, however, Ehud is not presented as an exceptional soldier. On the contrary, he is depicted as an obedient servant, who is, for reasons unknown at that point, armed. Even when he returns to the king in Judg 3:19, this time alone, he does not seem to have any intention on using his hidden dagger. Instead, he utters an ambiguous phrase, which prompts the king to evacuate the room and take Ehud up to his private chambers (Judg 3:19–20). After a repetition of the phrase in a slightly altered form, Ehud finally pulls his weapon and kills the king (Judg 3:20–21).

see Alter, The Art of Biblical Narrative, 38–41; Brettler, The Creation of History in Ancient Israel, 82–83; McCann, Judges, 44–45; Soggin, Judges, 50–55. For a religious nature of the message, see among others Trent Butler: “Often it appears as ‘word of God’ or ‘prophetic oracle’. Eglon has to get to the root of the matter and determine what the secret is. He knows such secrets can only be shared on a one-on-one basis” (WBC 8; Judges [Nashville, TN: Nelson, 2009], 71). Also Kraeling, “Difficulties in the Story of Ehud,” 206, and Moshe Garsiel, “The Story of Ehud Son of Gera (Judges 3:12–30),” in Reflections on Scripture: Selec tions from Studies of the Yishai Ron Memorial Bible Circle, II (Tel Aviv, 1977), 57–77 (69) [in Hebrew]. 13 See Baruch Halpern, The First Historians: The Hebrew Bible and History (San Francisco, CA: Harper & Row, 1988), 39–75; Amit, The Book of Judges, 180. 120 karolien vermeulen

The story thus presents itself as problematic; more in particular the problem is Ehud’s. God has sent him as a savior, but the odds are against him. He is left-handed, at least a bad omen, and forced to bring tribute to the oppressor of his people. Moreover, other people accompany him when meeting the king, so that it does not matter how many daggers he has under his cloak, he would never be able to eliminate his enemy without being caught.14 As a consequence, Ehud is stuck in a situation that does not seem to have an outcome. It is at that point, the point where the story seems to stop evolving and the dynamics between the characters is frozen, that the polysemous phrase comes in and creates an opportunity for Ehud to kill the king. When Ehud utters his ambiguous phrase in front of Eglon, he has nothing to lose. From his point of view, he is already in the worst of all possible situations, unable to help his people.15 Moreover his plan, that involves the use of a double polysemy, is foolproof. No matter how the king will interpret the question, Ehud will always get out unharmed. If the king to be a secret message, he will empty the room to דבר סתר considers the hear it. This would be the ideal moment to kill the king, which is exactly how it goes in the story as we have it in Judges. The kind of secret is irrelevant, although commentators have suggested various interpretations. A political message may need to be discussed in private among leaders, not in the presence of all. Likewise, a carnal secret also requires privacy.16 In both cases, what is important is that the king is interested enough to hear this message behind closed doors. In addition, even if the king did not interpret the message as such and as a result had it delivered in front of all the ones present, Ehud’s plan would not be uncovered. At that point, he has to make up a message he can deliver publicly. The king would have no reason to distrust him and Ehud would still walk out safe and sound.17

14 See Radaq on this, “in order to send out the ones around him.” See also the text ”so all those in attendance left his presence“ ,ויצאו מעליו כל העמדים עליו :itself that reads (Judg 3:19). 15 As noted by Butler, the role of Ehud as ‘savior’ was not as much to rescue but to sup port and help the people (Judges, 65–66). 16 For references on the various readings, see footnote 12. 17 This suggestion of a foolproof plan of Ehud contrasts with Amit’s perspective that “the events and the connections among them need to be explained by the repeated application of coincidence, and not by careful advance planning or calculation” and that resultantly God determines the success of human planning in the story (The Book of Judges, 173; “The Story of Ehud [Judges 3:12–30]: The Form and the Message,” in Signs and Wonders: Biblical Texts in Literary Focus, ed. Cheryl Exum [Decatur, GA: Scholars Press, 1989], 97–123). What the intentional use of polysemy 121

Of course, Ehud has every reason to believe that the plan will work and that the Moabite king will read something secretive into his words. Ehud was sent as a savior, so he should have his chance to be one, at least at some point in the story.18 Also, he was wearing a dagger under his cloak. When the opportunity came for him, he would be ready to do as expected.19 The left-handedness detail may not have necessarily made Ehud a good fighter; however, it gave him a certain advantage. With fewer people being left-handed, he would be able to surprise his opponent, just like he did by using words to get his way in. Also, if the king had suspected him, he could have checked his left side for a weapon and he would not have found one.20 As indicated above, this ties into the perfection of the plan of Ehud. a “Benjaminite.” The ,בן הימיני Furthermore, Ehud is presented as a information in itself is rather formulaic as the introduction of a new character goes hand in hand with the mentioning of his affiliation.21

Amit tries to elucidate here, is that the story thrives on a joint venture between Ehud and God with God as the stronger element to achieve the desired effect. Yet, in the analysis as described above, the plan is perfect, because of certain preconditions among which the fact that God is at Ehud’s side. If Ehud succeeds, it is partly, and maybe mostly, thanks to God’s help. Yet, on the micro level of the ambiguous words uttered in verse 19, it is Ehud who is the one acting. Nevertheless, Amit argues elsewhere that the repetition of the phrase in a variant in verse 20 creates two stages and a way out for Ehud to “withdraw[n] from his original plan [. . .] with some sort of oracular saying or fictitious secret” (The Book of Judges, 188–189). The latter statement is closer to what is argued in the current article. ,ויזעקו בני ישראל אל יהוה ויקם יהוה להם מושיע את אהוד בן גרא בן הימיני :Judg 3:15 18 “Then the Israelites cried out to the Lord, and the Lord rose up a champion for them: the Benjaminite Ehud son of Gera.” Amit has remarked that the judges were not repre sentatives of ideal leadership, rather they were “one-time liberator[s], who disappear[s] after performing their [his] act of deliverance” (The Book of Judges, 167). Notwithstanding, in this single moment of action, the judge was intended to succeed as God’s messenger and servant. As Amit notes herself, “the presentation of Ehud, already during the stage of exposition, as a deliverer who was raised up by God, creates expectations of a miracle, of a ‘great act’ or of some sort of manifestation of the Lord’s Spirit, while his becoming aware of the series of tactics carried out by Ehud is more likely to obscure the divine role in the act of delivery” (The Book of Judges, 172). ויעש לו אהוד חרב ולה שני פיות גמד ארכה ויחגר אותה מתחת למדיו על ארך :Judg 3:16 19 So Ehud made for himself a two-edged dagger, a gomed in length, which he girded“ ,ימינו on his right side under his cloak.” 20 See Butler, Judges, 59. 21 According to Boling, the reference to the Benjaminites is foregrounded by adding the definite article, as such “suggest[ing] that he is presented as typical—the Benjaminite par excellence.” He considers the ambiguity to be present in the fact that “Eglon was sin glehandedly defeated by this famous left-handed Benjaminite, as the Israelites chose to send tribute ‘by his hand’” (Boling, Judges, 86). Remarkably, Boling does not pick up the internal paradox of the ‘left-handed Benjaminite’, although he notices the descriptive way of indicating left-handedness. 122 karolien vermeulen

Yet, what makes this reference stand out is that it is immediately fol- a man impeded in/deficient with“ ,איש אטר יד ימיני lowed by the notion regard to the right hand.” So whereas Ehud is a son of the right hand, he is not at the same time, because something seems to be wrong with left,” the common word for“ ,שמאל that hand. Instead of using the word ‘left’ in biblical Hebrew, the text prefers the description that repeats the By doing so, the affiliation of Ehud as Benjaminite is affirmed 22.ימין word and recognized formally, but semantically it is denied, calling him “left” instead of “right.” The play prepares the action of Ehud in verse 19, presenting Ehud in advance as being two sides of a coin at the same time. This argument becomes even stronger when Ehud’s weapon is mentioned, ,a two-edged dagger.” As can be read in Judg 20:15–16“ ,חרב ולה שני פיות the weapon for which the left-handed Benjaminites were known was their sling: ויתפקדו בני בנימן ביום ההוא מהערים עשרים וששה אלף איש שלף חרב לבד מישבי הגבעה התפקדו שבע מאות איש בחור מכל העם הזה שבע מאות איש בחור אטר יד ימינו כל זה קלע באבן אל השערה ולא יחטא And the Benjaminites mustered on that day from the cities 26,000 men drawing a sword, apart from the inhabitants of Gibeah they mustered 700 men picked from all this people, 700 picked men deficient with regard to the right hand. Each of them could sling a stone at a hair and not miss.23 It was not a dagger and definitely not the two-edged one as mentioned here, but a sling. The change in weapon is another indication that Ehud is good at raising certain expectations which turn out to be fulfilled differently: he is a man of the right but he is also left, and he is a man of the sling, while using the dagger. To make the picture complete, the weapon

22 The unusual word choice is interpreted by Boling as a way to address and emphasize the “peculiar and unnatural” state of left-handedness (Boling, Judges, 86). The reading in a man with an emaciated“ ,גברא גמיד בידיה דימינא the Targum supports this, translating right hand.” Note that the LXX reads ἀμφοτεροδέξιον, “ambidextrous,” interpreting the left- handedness obviously as an advantage, being “with two right hands.” Other commentators, such as Butler, have remarked that whatever was intended with the description, “impeded cannot mean injured or deformed in any way,” because Ehud could have not killed the king with a single stroke, if something was wrong with his hand (Butler, Judges, 70). Moreover, his unusual appearance may have raised suspicion to begin with, which would have made as a means ימין any further actions impossible. Amit focuses on the repetition of the word to direct the readers to Ehud’s talented hand and what he may achieve with it (The Book of Judges, 179). A similar remark has been made by Luis Alonso-Schökel (“Erzählkunst im Buche der Richter,” Bib 42 [1961], 149–157). The regular word for ‘left’ is used only once in the story, in Judg 3:21. 23 Judg 20:15–16. See also 1 Chr 12:2. the intentional use of polysemy 123 is girded at his right hand side. This in itself is a practical decision; weapons are carried at the opposite side of the hand used to handle them. But then this also plays, once more, upon the confusion between right and left explored in the lines introducing Ehud.24 Another sign that Ehud may succeed in getting the king on the wrong foot, is the depiction of the king. As in the case of Ehud, the narrator drops a few hints, adding at first sight superfluous or irrelevant information. very healthy, well-formed” (Judg 3:17). How does that“ ,בריא מאד Eglon is have anything to do with the tribute that Ehud is bringing him? Is it making the king fatter or rather more avaricious? Is his appetite to be taken literally or in a figurative sense?25 In addition, there is the king’s name, Eglon, meaning “calf,” often understood as “sacrificial or fatted calf.”26 The

24 Amit suggests that the mentioning of both left-handedness and being a Benjaminite “seems to have been an idiomatic expression used to refer to daring warriors from the tribe of Benjamin. The characterization of Ehud by this idiom indicates, not only that he was a superb fighter having special talents, but is also a key to creating the hypothesis of what is about to happen: that the one about to bring a tribute to Eglon king of Moab is Ehud’s son of Gera, a daring fighter known for the use of his left hand” (The Book of Judges, 179–180). 25 On the portrayal of the king, see Butler, Judges, 70, “Eglon is thus pictured as very pleasingly plump on the one hand and as a perfect creature for sacrifice on the other.” carries mostly positive connotations when used in the בריא Stone notices that the term Hebrew Bible and that the translation ‘fat’ refers to health, beauty, and attractiveness of the animal, or human as in this case. The term is never used in the sense of ‘obese’ or to mock someone (Stone, “Eglon’s Belly and Ehud’s Blade,” 650–651). The LXX seems to adhere to a similar view, rendering the word as ἀστεῖος, “handsome,” “charming,” “refined.” Yet, the fact remains that of the fourteen occurrences in the Hebrew Bible, it is only used thrice to refer to people, so that it is more likely that the fat-cows-image is evoked in the audience’s mind than that of the slender looking young men in the book of Daniel (Dan 1:15ff). Also, Ehud does not need to be fat to be a suitable sacrifice; it suffices that he is healthy and well-formed. The positive connotation of the term does not change the pos sible sacrificial setting intended. On the contrary, it makes Ehud the ideal sacrifice. Those who read the story as a euphemistic tale of a homosexual encounter may see in the beau tiful Ehud another element that supports their interpretation. Therefore, Stone’s remark which was aimed at countering the reading of a sacrifice into the scene, is actually giving additional support to such an interpretation. as well as ,עגלון ,עגל .On the generic reading of the name, see HALOT II, 784–785, s.v 26 Alter, The Art of Biblical Narrative, 38–41. Stone also objects to this interpretation arguing that the adjective ‘fatted’ is by no means suggested by the Hebrew term. He rightly points considered to be the primary noun in the name, as a ,עגל to the many uses of the word designation for the golden calf, an idol (Stone, “Eglon’s Belly and Ehud’s Blade,” 654–655). as a עגל Out of the 41 occurrences of the word in the biblical corpus, only eight refer to sacrifice and three more occurrences can be found to use the word in a context of eating or describing the calves as well nourished. Nevertheless, the word has two major con notations, being the golden calf on the one hand and a sacrifice on the other hand. The remaining occurrences are not be categorized under one single label. Since the scene in Judges 3 has no connection whatsoever with idolatry or a golden calf, but does include another ,מנחה ;a weapon ,חרב ;also used for animals ,בריא) elements of a sacrificial setting 124 karolien vermeulen combination of the name, even in its more neutral version of “calf man” evoke a sacrificial (בריא) and the mention of the king being well-formed setting. Add to that the dagger of Ehud and the presence of other gifts ;and it seems that everybody is preparing for the king’s slaughter (מנחה) the dagger will kill the fatling.27 Moreover, the king is not presented as the smartest guy in town. He seems to be more interested in life’s little luxuries and unable to consider messages that address other issues, such as his own murder.28 Therefore, when Ehud comes at Pesilim, near Gilgal, a turning point, he has very good reasons to believe that he will be able to overpower the king and turn things around.29 When he arrives at the palace, Ehud’s kind of gift), it is far more likely that a sacrifice is evoked. Note that to Amit the golden calf connotation has cultic associations “as an object of adoration” as does the reading of a burned offering (The Book of Judges, 184). Stone formulates a complementary argument that “readers aware of the social and political context of Iron I and the literary traditions it spawned readily grasp that no such person as the contemporary cartoon of Eglon could ever have become a Moabite warlord king or even have figured believably in a fictional tale about that era. Nor would such a story inculcate admiration for Ehud, who would appear trivial compared to Deborah, Gideon, Othniel, Samson, or David, much less their Iron I literary counterparts, Achilles, Hector, Patrocles, or Diomedes” (Stone, “Eglon’s Belly and Ehud’s Blade,” 656). Yet, it seems to me that the meaning of the name as having to do something with calves is undeniable as well as the fact that there are other words present in the text that tie into a sacrificial reading. Stone is right in that they do not need to be taken pejoratively, but that does not mean that the reading should be rejected all together even with the more neutral reading of the words (Eglon as “calf man” instead of “little calf,” .(as “well-formed” rather than “fat,” see above בריא likewise 27 On the sacrificial reading of the passage, see Brettler, The Creation of History in Ancient Israel, 81–82; Alonso-Schökel, “Erzählkunst im Buche der Richter,” 149–150; Alter, The Art of Biblical Narrative, 39; and Yairah Amit, The Book of Judges, 183–184. A literary- linguistic reading of the text argues in favor of a possible sacrificial reading of the passage. Here as well, polysemy is at play, yet on the level of the larger text. Is the story about a sacrifice or is it about a political conflict or rather about a sexual encounter? The text con tains clues that point in different directions, supporting each of the readings and as such presenting itself as polysemous. 28 On a ridiculous king, see among others Jonathan Grossman, “The Use of Ambiguity in Biblical Narratives of Misleading and Deceit,” Tarbiz 73,4 (2006): 483–515 [Hebrew], and J. Alberto Soggin, Le livre des juges (Commentaire de l’Ancien Testament Vb; Genève: Labor et Fides, 1981), 49. ,the consonants are reversed—עגלון and גלגל Notice the subtle wordplay between 29 ,עגל turned over. As suggested by Butler, the name of Eglon is a phonetic pun on the word “round,” and may as such add to the play as described here as well (Butler, Judges, 69). See also Alonso-Schökel, “Erzählkunst im Buche der Richter,” 148. In addition, there is the semantic load of the term itself, “to turn.” Whereas scholars have pointed to the sculptures at Gilgal as a kind of border or evoking a religious setting that prepares the interpreta as some oracle, the wordplay suggests two things: one, that it is a דבר סתר tion of the place of turning and that Ehud turns back there, and second, that this turning involves Eglon, whose very name is evoked by the sounds of Gilgal. For the role of the sculptures at Gilgal, see among others Amit, The Book of Judges, 186–187; George Moore, A Critical and the intentional use of polysemy 125 phrase works as a magical ‘Semsi mountain open!’ It allows him to do the impossible: kill the king.

The Riddle Read and Reread

Polysemy in the Interpretation Narratologically, the polysemy is almost a deus ex machina, an unexpected and magic solution to the problem. Since it is the character Ehud, who comes up with this answer, scholars have felt the need to explain his choice. After all, Ehud was a ‘deliverer’ and thus a kind of role model. However, he defeats his opponent by means of a trick, not in a heroic fight.30 Although commentaries take different roads with regards to the overall interpretation of the passage, the majority attempts to disambiguate the double polysemy. This clarifying tendency goes from paying little to no attention to the wordplay, as if it is straight forward and does not require any explanation, to justifying the trickery as the right of a hero. In addition, the clarification goes hand in hand with a moral judgment of the character Ehud. For some, Ehud is without a doubt a hero; for others, his position is less clear. If the commentary talks about Ehud in terms of a hero, deceit is considered a skill and a tool to achieve something praise- worthy. It is often discussed at length.31 In the other cases, the artful use of language is evaluated rather negatively, which results in a tension between the supposedly heroic role of Ehud and the despicable means he uses to maintain that status.32

Exegetical Commentary on Judges (ICC 7; Edinburgh: T&T Clark, 1895), 95; Alonso-Schökel, “Erzählkunst im Buche der Richter,” 50; Gregory Mobley, The Empty Men: The Heroic Tradi tion of Ancient Israel (ABRL; New York: Doubleday, 2005), 93–96. 30 Feldman, “Ehud,” 46. 31 Boling, Judges, 85; Soggin, Le livre des juges, 47–53; Halpern, The First Historians, 39–75; Hamlin, Judges, 74–75; Susan Niditch, War in the Hebrew Bible, 106–107, 117–119; Barnabas Lindars, Judges 1–5: A New Translation and Commentary (Edingburgh: T&T Clark, 1995), 143; Robert O’Connell, The Rhetoric of the Book of Judges (Leiden: Brill, 1996), 84; Amit, The Book of Judges, 167–180; McCann, Judges, 43–45; Christianson, “A Fistful of Shek els,” 53–78; Brian Tidiman, Le livre des juges (Vaux-sur-Seine: Edifac, 2004), 97–105; Susan Niditch, Judges: A Commentary (Louisville, KY: Westminster John Knox Press, 2008), 1–44, 55–59; Stone, “Eglon’s Belly and Ehud’s Blade,” 649–663. 32 Rashi on Judg 3:19, Gen. Rab. 99.3; Ruth Rab. 2.9; Grossman, “The Use of Ambiguity in Biblical Narratives of Misleading and Deceit,” 483–515; Karel Deurloo, “Karikatuur van de vreemde koning (Ri. 3:7–3:30),” in Geen koning in die dagen, ed. Hanna Blok et al. (Baarn: Ten Have, 1982), 31–36. 126 karolien vermeulen

Language and Character Compatible Many scholars identify Ehud—although at first sight presented as the underdog approaching the Moabite king—as the hero of the story. In identifying the exact nature of Ehud’s heroism, they use multiple comparisons. Eric Christianson notes that It is at least clear to most that the Ehud story makes ample use of rhetorical devices in order to establish a character for whom violence is simply a tool with which to deliver the word of God.33 Throughout his article, it becomes clear that the figure of Ehud is very much like the hero character of the spaghetti Western, using deceit as a “special ability.”34 In the climactic showdown of Sergio Leone’s first of the dollars trilogy, A Fistful of Dollars, the Man conceals a steel shield beneath his poncho, one that he wrought himself. His opponent is unaware of this and eventually loses the battle as a result of his confusion at the Man’s surviving several rounds of rifle fire. Ehud too conceals a self-made weapon. Like the Man, Ehud has no qualms with winning the ‘shoot-out’ with deceit. Indeed, there is little that Ehud does in the story that does not involve some deception.35 Christianson subscribes to a vision that acknowledges and favors Ehud’s verbal and nonverbal tricks. As for the play with language, he states that “Ehud’s words are choice, ironic (iconic?) and as sharp as his sword.”36 The same economy of words is characteristic for heroic Western language.37 According to Susan Niditch, Ehud is not a cowboy, but a sort of Robin Hood—a social bandit who fights for justice against the mighty and powerful.38 On a metaphorical level, the confrontation between Ehud and Eglon represents an Israelite-Moabite conflict.39 The war ideology found in many stories of the book of Judges takes a particular form in this chapter: “an ideology of tricksterism. The confrontation relies upon deception,

33 Christianson, “A Fistful of Shekels,” 53. 34 Christianson, “A Fistful of Shekels,” 61. See also Boling, Judges, 85; O’Connell, The Rhetoric of the Book of Judges, 84, and Mobley, The Empty Men, 76. 35 Christianson, “A Fistful of Shekels,” 63. 36 Ibidem, 64. 37 Ibidem, 65. 38 Niditch, Judges, 3–4, 55. See also Deurloo, “Karikatuur van de vreemde koning,” 33–36, and Hamlin, Judges, 70–77. 39 Marc Brettler, “The Book of Judges: Literature as Politics,” JBL 108,3 (1989): 395–418. the intentional use of polysemy 127 an ethic entirely at odds with a soldier’s code of honor.” 40 Nevertheless, Niditch accepts and acknowledges ambiguity as part of the story, because of its recurrence elsewhere in the book of Judges.41 Note that the warlike setting also allows for a euphemistic reading of the passage. A connection between combat and sex has been found in various other narratives in the Hebrew Bible. Similarly, the writer(s) of the Ehud-Eglon episode may have “purposefully commingle[d] military death and sexual conquest.”42 Baruch Halpern proposes a third comparison. According to him, Ehud is [. . .] a seasoned samurai, or to use a modern caricature, a sort of James Bond. Bred for combat, schooled to feats of sinister valor, Ehud was precisely the man to execute the operation that Judges 3 describes.43 Halpern spends many pages on Ehud’s left-handedness and its compatibility with heroism. It turns out that the left-handers were a kind of elite troop and not the weak or bad warriors, as hinted at before.44 The ambiguous language used by Ehud is considered part of the story. In the end “Ehud’s ‘secret word’ (v.19) remained a secret, but Eglon’s ‘hidden things’ were now exposed.”45 Halpern’s way of sneaking wordplay into his own text is aimed at raising sympathy for his argument and in extension for the character Ehud. All three interpretations identify Ehud as a prototypical hero and his double tongue as a useful and skillful tool. Notice how Halpern feels the need to coin a little wordplay himself and how he literally turns the tool of the narrative into his own. Similarly, the contributions of Christianson and Niditch contain little twists as the above quotes illustrate, as if the particular language use of the character prompts them to do something comparable. Each of the scholars has a particular point to make and whereas

40 Niditch, War Ideology in the Hebrew Bible, 106. 41 Niditch, Judges, 15. 42 Niditch. Judges, 6. This interpretation is also supported by among others, Alter, The Art of Biblical Narrative, 45; Brettler, The Creation of History in Ancient Israel, 82–83; and Mobley, Empty Men, 85–86. Other stories in which war and sex are combined: Jael and Sisera (Judges 4–5), Samson and Delilah (Judges 16), the rape of Dinah (Genesis 34). Out side the Hebrew Bible, the Homeric material forms a prototypical example. See Emily Vermeule, Aspects of Death in Early Greek Art and Poetry (Berkeley, CA: University of Cali fornia Press, 1979). 43 Halpern, The First Historians, 39–75. On the James Bond reading, see also Mobley, Empty Men, 76. 44 Halpern, The First Historians, 59–60. See also Amit, The Book of Judges, 179–180, and Soggin, Le livre des juges, 47–53. 45 Halpern, The First Historians, 60. 128 karolien vermeulen the polysemy is never the starting point or the main element of interest, they all pick it up along their discourse, using it at their convenience and adding it to their list of reasons-to-consider-Ehud-a-good-guy. And as their arguments show, they succeed in doing so. The polysemous language proves that Ehud is a spaghetti western hero as well as a Robin Hood and a seasoned samurai. His weapon resembles both high tech weapons such as occurring in James Bond movies, and simple and cheap tools such as the gravel Tom Thumb found along his path.

Language and Character Incompatible Another group of scholars has a more negative take on the double polysemy. They associate it with deception, lies, and dishonesty, anything but a heroic quality. Since they also consider Ehud a good man, they are faced with a tension in the text. In their analysis of the story, they will try to solve this incompatibility by neglecting, minimizing, or rationalizing the double tongue. In general, they aim at reducing the negative connotation they see evoked by the phrase uttered by Ehud. An attitude of negligence can be found in rabbinic writings. It is striking how few notes the rabbis—usually fond of ambiguous language— have made on the verse. The verse is only quoted twice in the rabbinic corpus: once in Genesis Rabbah and once in Ruth Rabbah.46 Both show a need to reduce the negative characterization of Ehud in the story, or more precisely, the negative interpretation. In the former, the commen- בנימן זאב יטרף. :tary connects Benjamin, Jacob’s youngest son, with Ehud Although 47.מדבר בשופטין מה הזאב הזה חוטף כך חטף אהוד לבו של עגלון this midrash is traditionally highly regarded, a critical undertone emerges. The explanation itself illustrates that the immediate association of the polysemy as something despicable has to be adjusted.48 The deception is only mentioned as explanatory note, but according to the midrash it should be understood as an act of seizure, the way in which Ehud acted. It is not explicitly stated that Ehud’s intervention is partly verbal. The focus is on the result of the action: Ehud succeeds whereas the way in

46 Gen. Rab. 99:3, and Ruth Rab. 2:9. 47 Gen. Rab. 99:3: “Benjamin is a devouring wolf. One speaks about his judge (i.e., the judge descended from him). As this wolf seizes, so did Ehud seize the heart of Eglon (i.e., he deceives him).” 48 Feldman, “Ehud,” 139. the intentional use of polysemy 129 which he achieved this, is ignored.49 The reference in Ruth Rabbah talks about Eglon and his piety to God instead of about Ehud’s role.50 Here, the Moabite king (he rises from his throne, i.e., he rises for God) forms the point of focus, as such, quietly ignoring the deceitful act of Ehud.51 The lack of other mentions of the passage can be interpreted in light of the references that are made: it is better not to draw attention to the exact way in which Ehud achieved his goal of killing Eglon. A similar approach can be seen with regard to other passages including deceit committed by figures assumed to be role models for the people. One can think of the passage in Gen 12:10–20 in which Abraham calls Sara his sister which results in her being taken away to become the wife of another man.52 In the Antiquities of Josephus, the Ehud-Eglon episode is rewritten in order to minimize Ehud’s bad behavior, and consequently, rehabilitate him. This was necessary as the Romans, Josephus’ primary audience, felt strongly about deceit, as we see, for example, in Livy’s disdain for the Alban leader Mettius Fufetius, who broke his treaty with Rome (1.27–28), and for the Carthaginians, who were known for their faithlessness (fides Punica). We can see Josephus’ uneasi- ness on this matter elsewhere in his treatment of the deceitfulness of Jacob, Simeon, and Levi.53

49 Jacob Neusner, Genesis Rabbah: The Judaic Commentary to the Book of Genesis: A New American Translation (Volume III; BJS 106; Atlanta, GA: Scholars Press, 1985), 377. ויקם מעל הכסא אמר לו הקב"ה אתה עמדת מכסאך לכבודי חייך הריני :Ruth Rab. 2:9 50 And when he rose from the throne, the Holy One Blessed“ ,מעמיד ממך בן יושב על כסא ה' is He said to him, ‘You rose from your seat for my honor. By your life, I will raise up from you a son who shall sit on my throne.’” 51 Yet, Feldman explains otherwise: “Such respect for Eglon reflects the general attitude of respect which the rabbis had for rulers, simply in virtue of their office.” Feldman does not necessarily consider Ehud a better person, but according to him, the gap between the two characters is smaller. 52 Notice that in a repetition of the motive in Gen 20:1–18, the text itself already nuances the lie by having Abraham explain to the king that Sara is indeed his sister, in addition to ,And yet indeed“ ,וגם אמנה אחתי בת אבי הוא אך לא בת אמי ותהי לי לאשה :being his wife she is my sister, the daughter of my father is she, but not the daughter of my mother, and she became my wife” (Gen 20:12). This justification is not present in the other occurrences in Gen 12:10–20 and Gen 26:1–11. Yet, given their connection, it may be understood in all of the passages. Such a reading has been defended by Klaas Smelik, “Verhaal en context,” in Een patriarchale leugen: het verhaal in Genesis 12 verschillend belicht, ed. Bob Becking & Klaas A.D. Smelik (Ten Have: Baarn, 1989), 54–77. For a discussion of the motive, see among others, Cheryl Exum, “Who’s Afraid of ‘The Endangered Ancestress’?” in The New Literary Criticism and the Hebrew Bible, ed. Cheryl Exum & David Clines (Sheffield: Shef field Academic Press, 1993), 91–113. 53 Feldman, “Ehud,” 141. 130 karolien vermeulen

In order to disambiguate Judg 3:19, Josephus leaves out the second message in verse 20.54 Again, the commentator assumed a connection between the use of the double polysemy on the one hand and the moral judgment of the character on the other hand. Interestingly, Josephus’ changes were prompted by the supposed reaction of his own audience, the Romans. While various sound and structural devices are listed and praised for their sophistication in Latin rhetorical treatises, polysemy is mentioned under the heading ‘vices’ in the same works.55 The underlying idea was that a good and honest man did not need trickery to convince his audience. Aware of this sensitivity of his audience, Josephus decided to make the necessary changes. And by changing the language, he also changed the overall picture of Ehud, making him a more honorable person, at least in the eyes of the Romans. Jonathan Grossman offers another interpretation that draws on the discrepancy between Ehud’s supposed status and his linguistic behavior in the narrative. In his article “The Use of Ambiguity in Biblical Narratives of Misleading and Deceit,” Grossman relies on the psycholinguistic mechanics of the lying process: יש מן הילדים שבשעה שהם משקרים יאמרו 'אני מפתיח' במקום 'אני מבטיח', או 'אני נשבר' במקום 'אני נשבע' ובכך יבקשו למלט נפשם מן השקר, ויש מן המבוגרים שיאמרו 'אמנם הבטחתי', אך לא הבטחתי לקיים.56 Thus, since the act of lying intrinsically influences people’s utterances, confusing and confused language of a character go hand in hand with his/her motivation. Ehud is nothing more than a liar, who is tricked himself by language, as it reveals his true state of mind. This does not imply anything on Ehud’s morality; it only reveals something about him being a human and thus having to play by the rules of the human brain, whether he likes this or not.

54 Jos. Ant. 5:185–197. 55 Reth. Her. II.XXVI.40: “Item vitiosum est quod in aliam partem ac dictum sit potest accipi,” “Again, that is faulty which can be taken in another sense than the speaker intended.” Notice that the author is discussing ambiguity as faultiness in argumentation here. On ambiguity and style, see Rhet. Her. IV.LIV, Inst. 6.3.47ff. The former thinks of it as “obscuring the speech” (“quae obscuram reddunt orationem”), whereas the latter considers double entendres not suitable to orators (“non omnis eos oratoribus convenire in primis ex amphibolia”) and “rarely effective” (“raro belle respondeant”). 56 Grossman, “The Use of Ambiguity,” 484, “It happens among children at the moment of lying, that they will say: ‘I premise,’ instead of ‘I promise,’ or ‘I bear,’ instead of ‘I swear,’ and through this their minds seek to escape from the lie. It happens among adults who will say ‘I truly promised,’ but I did not promise to fulfill.” the intentional use of polysemy 131

Furthermore, Grossman deconstructs the idea of ambiguity by referring to the importance of mockery in the text. It is obvious that Ehud is going to kill the king. Any reader understands after mentioning a hidden sword and gaining favor through tribute.57 The king is the only ignorant player. According to Grossman, there is no better way to ridicule the king than by letting him be killed by a small left-handed man.58 Notice that in this interpretation neither the king nor Ehud is depicted as strong and heroic character. Ehud is a victim of his own cognition and called unimportant and even deficient, whereas the king is the laughing-stock of the story unaware and even insensible to what is going to happen to him. In the end, Ehud will be the better one of the two, because despite it all he manages to complete his mission, notwithstanding rather than thanks to his polysemous speech. The three approaches discussed here illustrate that a negative assess- ment of Ehud’s language, which naturally evokes a negative evaluation of the character, asks for an adapted interpretation in order to prevent a moral rejection of the same character by the audience. If no comments are given on the polysemy, no discussion is necessary and the attention can be drawn to what really matters, that he managed to “seize Eglon’s heart” and as such save his people after years of suppression. Changes to the text have a similar effect. Why bother the audience with polysemy if that will make it harder for them to focus on the good qualities of Ehud? And if all of this does not help, one can still play the rational card of psy- cholinguistics. People lie sometimes and that messes up their language. Ehud was a good guy who happened to be a bad liar and so his real plan shone through, when he was addressing the king. Yet, in the end, he did what he had to do and so one should ignore his slip of the tongue which was rather a slip of the brain.

57 See also Moore, A Critical and Exegetical Commentary on Judges, 95, and Deurloo, “Karikatuur van de vreemde koning,” 33. Meir Sternberg sees in the story an example of foreshadowing, so that “everything so falls into place, therefore, that Eglon is already as good as dead even before he grants the ambassador a private audience” (Meir Sternberg, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading [Bloom ington: Indiana University Press, 1985], 333–334). 58 Grossman does not follow the idea here, as developed by Halpern and others, that the left-handedness was rather a sign of an elite paratrooper than a lesser man (Halpern, The First Historians, 59–60; Marc Brettler, “Never the Twain Shall Meet? The Ehud Story as History and Literature,” HUCA 62 (1991): 285–304 [291–294]). 132 karolien vermeulen

The Functionality of the Polysemy in Interpretation As illustrated in the above overview, interpreters had and have multiple ways at their disposal to link the double polysemy to the story. Remark- ably, they all prefer to read the wordplay as deceit: Ehud used an ambiguous phrase of which he was sure the king would interpret it the wrong way. This tricksterism can be taken positively or negatively. In the latter case, commentators have undertaken various efforts to exonerate Ehud and his despicable means of double tongue. Little to no attention has been paid to the fact that the words were not necessarily a lie, or better, that they are truthful and untruthful at the same time. As argued in this article, the phrase is an attempt, an honest one, to get out of an impossible situation. Ehud could have had some idea of what the king would do, when he would utter his ambiguous words, but Ehud was not a visionary. There was no way for him to be absolutely sure that the king would misapprehend. Therefore, he keeps all his options open, which is exactly what the double polysemy does: creating possibilities, broadening the narratological outcomes, and exploring the semantic potential. Therefore, the misunderstanding of the king is twofold; it entails both a wrong choice—there is no pleasant secret intended—and the very act of choosing—it were better for him if he had not chosen at all. As such, most commentators are much more like Eglon, withholding only one option and therewith overlooking the complexity of the text. Their preoccupation with the deceptive aspect of polysemy forces them to disambiguate the phrase and prevents them from seeing the richness behind it. Although the majority of exegetes have recognized the dual message in case of Ehud’s words on the lexical level, their final claims on the narratological level are not multiple: the judge must be either a liar or a truthful man. The fact that Ehud could be both simultaneously is not considered an option. This is the more remarkable since that is exactly as far as the text itself goes. Yet, every polysemy includes the possibility of a monosemic reading and that possibility has been eagerly explored by exegetes. As a result, scholars have been able to defend opposite positions based on the same expression. This became even more so due to the moral conclusions they have drawn from the ambiguity. Adele Berlin has warned against letting the method determine the text’s formal analysis arguing that “the point of comparing [. . .] is to see how and why they may differ, not to collapse their differences into a hypothetical Vorlage.”59

59 Adele Berlin, “A Search for a New Biblical Hermeneutics: Preliminary Observations,” in The Study of the Ancient Near East in the Twenty-First Century: The William Foxwell the intentional use of polysemy 133

In the case of polysemy, a similar guideline can be issued to accept the variety of readings instead of ruling it out by means of a sophistic manipulation—by the interpreter—of the double tongue in the text.

Secret Words in Action: A Conclusion

Based on the previous analysis, one can conclude that the ambiguity in Ehud’s story has a double role to play. It serves both the characters in the story and the interpreters outside the narrative world. On the one hand, ambiguity is a tool of the character Ehud. It enables him to pursue his murder plan, despite the fact that the king seems to be surrounded by guards protecting him.60 Whether Eglon was really mislead or whether his curiosity overruled his rational thinking, is not made clear. Nonethe- less, Ehud has Eglon’s full attention and the power balance in the story shifts. The ambiguity solves the impasse of the episode. How other than by double entendre would the servant have succeeded in his plans to kill the king? On the other hand, the same device plays an important role in the interpretation of the verse and the broader story. Commentators have connected the double polysemy ([un]consciously?) to their appreciation of Ehud, and have used it to justify their exegetical claims.61 For all interpretations, it counts that they have turned the narrative’s tool into their own. The double tongue is indeed deceitful to the extent that it gives Eglon as well as the interpreter a false feeling of coming to the right interpretation. It is more honest, when it is viewed as creating possibilities. Therefore, ,might be better viewed from Ehud’s position דבר סתר what is hidden in who oversees the various readings and knows that it is exactly in the plurality of meaning that his way out may emerge.

Addendum on the Position of Translations toward Ambiguity

Note that the approach of commentators might not be that different from translators with regard to the words of Ehud in Judg 3:19. Whereas the latter had to choose one reading over another because of the very nature of translation, the

Albright Centennial Conference, ed. Jerrold Cooper & Glenn Schwartz (Winona Lake, IN: Eisenbrauns, 1996), 195–207 (202). ”.All that stood by him“ ,כל העמדים עליו :Jugd 3:19 60 61 Even Stone, who opens his article with counterarguments on the “morality of regi cide,” does nothing more than rehabilitating Eglon. The overall discourse is still heroic in nature, and discusses likewise morals (Stone, “Eglon’s Belly and Ehud’s Blade,” 649). 134 karolien vermeulen former seemed to have felt that urge as well, though they could have kept multiple readings. Interestingly, translators who had the possibility of retaining a double meaning did not necessarily do so. Although not cognate, the LXX renders the secret thing with an ambivalent lexeme: λόγος μοι κρύφιος πρὸς σέ βασιλεῦ.62 The Greek λόγος covers both saying and matter.63 However, it is uncertain whether this is a conscious rendering of the wordplay or a matter of coincidence, since the chosen word is the common in Greek. The apposite κρύφιος, “hidden,” has דבר denominator for the Hebrew been attested in Prov 9:17 as well. In the plural, the word is rendered as “secrets” in Psalm 9 (title line). Again, the Greek translation uses a word that seems to cover both Hebrew meanings: the word/matter can be hidden (e.g., under the cloak) or be secretive (e.g., the fact that Ehud will kill him).64 a secret“ ,פתגמא דסתרא אית לי למללא עמך מלכא The Targum with its Aramaic דבר thing/word I have to tell you, O King,” renders the polysemy of the Hebrew immediately מלל Yet, the addition of a verb of saying 65.פתגם with the ambiguous undoes this. The language offers the possibility to retain the double reading, but the Targumist chose to clarify the text. This tendency will reoccur in most modern translations of the verse.66 From the Vulgate on, translations use either “thing” (or any synonym: “something,” “case,” “object” . . .) or “word” (“message,” “errand” . . .). The latter is the most favored option. The Dutch Statenvertaling forms one of few exceptions. Maar hij zelf keerde wederom van de gesneden beelden, die bij Gilgal waren, en zeide: Ik heb een heimelijke zaak aan u, o koning ! dewelke zeide: Zwijg ! En allen, die om hem stonden, gingen van hem uit.67

62 LXX Judg 3:19. 63 LSJ, 862–863, s.v. λόγος. Greek makes a distinction between words and matters spoken of, but the word λόγος does not cover concrete objects or things, contrary to the Hebrew .דבר 64 LSJ, 804, s.v. κρύφιος. See also Takamitsu Muraoka, A Greek-English Lexicon of the Sep tuagint (Leuven: Peeters, 2009), 415–416. ,(See also Willem F. Smelik, The Targum of Judges (Leiden: Brill, 1995 .פתגם .CAL, s.v 65 371–373. 66 In the course of this article, I have consulted the following translations in English: KJV (1611), Young’s Literal Translation (1898), Douay-Rheims American Edition (1899), JPS (1917), NIV (1965), NASB (1995), and ESV (2001); in French: La Bible de Jérusalem (1956), Chouraqui (1974–1977), Nouvelle Edition Genève (1979), Traduction Oecuménique de la Bible (1988), and Bible en français courant (1997); in German: Luther’s Bible (1545), Buber-Rosenzweig (1925), Schlachter Version (1951), Einheitsübersetzung (1980), and Neue Evangelische Uberträgung (2009); in Dutch: Statenvertaling (1637), Leidse Vertaling (1912), Willibrordvertaling (1995), Nieuwe Bijbelvertaling (2004), and Naardense Bijbel (2004). The translations in this sample show the tendency discussed above, which is inherent to the nature of translation to interpret. The clarifying aspect occurs especially in cases of ambiguity. 67 Statenvertaling Judg 3:19. English translation of Statenvertaling: “I have a secret thing/ word for you, O King.” the intentional use of polysemy 135

While the word ‘zaak’ is the semantic equivalent of the English ‘thing,’ ‘case,’ the Dutch word covers both material as well as verbal things and as such it equals of the Hebrew source text. This will be the more probable, if דבר the polysemous one takes into account verse 20, in which the phrase of verse 19 is nearly repeated, I have a ‘thing’ of God for you.”68“ ,דבר אלהים לי אלך :with only a slight variation ,Thus .דבר In this case, the Dutch translator uses the word ‘woord’ (“word”) for he makes deceit a human, but not a divine characteristic. The change in word choice for the rendition of the same Hebrew word is something that is typical of the Statenvertaling. It shows that the translators did not have a problem with the deception of Ehud, as long as he kept God out. Ehud was only allowed to lie on instead of אלהים his own account. Nevertheless, the repetition of the verse with away. After all, Ehud is sent by דבר does not take the ambiguity of the word סתר God as savior and even when God did not support the bloodshed and whatever else Ehud pulls out of Eglon’s stomach when he kills him, Ehud may still consider -which he hap ,דבר אלהים his act of saving his people as an act for God and thus a pens to fulfill by means of a sword, indirectly a “thing from God.” Linguistically as דבר speaking, the Dutch translators could have rendered both occurrences of polysemes, yet, the addition of the word ‘God’ prompted them to not do that the second time as to not connect the sword and the murder it stands for, to God. It is remarkable that almost all modern translations, except for the above mentioned Statenvertaling (and its revised version of 2010), opt for the ‘word’ aspect This can be motivated by its primacy in dictionaries and as .דבר of the lemma such by its frequent occurrence in the biblical Hebrew corpus.69 Furthermore, the אלהים occurrence of an almost identical version of verse 19 in verse 20—where supports such an interpretation. Taking into account that the—סתר replaces “thing” is apparently also from God, theology and the human sense for reason lead to a reading of a “word” rather than a “something.” It would be remarkable to have Ehud call the dagger coming from God while that same God is condemn- ing murder and bloodshed elsewhere.70 Also, God is not the kind of divine being delivering weapons to his people from time to time, at least not in the material- ized form of short swords, as is the case here. As Klein points out Eglon is led to anticipate a divine revelation, a ‘secret word/thing of Elohim’, while Ehud actually has a deadly secret of human origin. Beside the obvious disparity, Ehud’s substitution of ‘Elohim’ for ‘secret’ implies that the two words are correlative, which is only part of the truth. A secret may be good or bad, but god is good; unlike the willful withholding of a secret, god is unknowable. The divine name is used as a lure.71

68 Judg 3:20: “A word of God I have for you.” .דבר .HALOT I, 211–212, s.v 69 ”.you shall not murder“ ,לא תרצח :See among others, Exod 20:13 and Deut 5:17 70 71 Klein, The Triumph of Irony in the Book of Judges, 38. 136 karolien vermeulen

Yet, be it as it may, Ehud could have still maintained his double tongue even in verse 20, killing Eglon with God’s regards.72 In light of the analysis of the polysemy as a way of creating possibilities and broadening instead of limiting the semantic load of the story, a polysemous translation is preferable in order to let the audience, not able to read the source text, experience the wordplay. And if this audience feels like choosing one particular reading over another, they can rightly do so. According to Amit, that is exactly their job as readers, “wavering, hesitating, deciding, in brief: experiencing the events described—the more the meanings and messages of the work become their [his] own conclusions, like one who has learned because he experienced it on his own flesh, and was convinced.”73 The experience of ambiguity and polysemy is in this respect not different from the involvement of the reader in the actions; making sense of the polysemy is an action in itself and one that, contrary to the narrative itself, cannot move forward without the reader’s active participation. In order to make this happen, a translation should at least offer its reader the opportunity to choose for him/herself and make this process possible. In the absence of equivalent polysemes, in itself the most realistic outcome, translators can make use of annotated translations, such as is customary in the JPS translation, or apply backslashes in between the different readings of the word. It may be worth it to consider other typographical ways of presenting translated polysemous words, something which is unexplored at the time.

72 According to Amit, the two utterances are indeed different despite their obvious sim ilarities. They have a different function. “Following the first statement, the king dismisses all of his attendants from the room. His return from the sculptured stones leads the king to believe that Ehud indeed has a secret message to tell him. In the second stage, Ehud drew close to the king and repeated his words, substituting the word ‘from God’ for ‘secret,’ and omitting the address, ‘O king.’ This omission highlights the importance of God, who is the true king, and of the ‘message’ about to be said in his name” (The Book of Judges, 188). 73 Amit, The Book of Judges, 168. Bridging the Gap between Linguistics and Literary Studies of Ancient Biblical and Jewish Texts: A Proposal Exemplified by a Study of Gen 9:8–17

Ellen van Wolde

Abstract

,in relation to the deity and to the clouds קשת Because Gen 9:8–17 uses the word the inference has been made in biblical scholarship that the text refers to a rainbow. The plausibility of this inference is tested in this article. Attention is given to the various linguistic aspects of this word in the Hebrew Bible and to the specific textual composition of Gen 9:8–17 as well as to the broader ancient Near Eastern framework established by comparative literary and iconographic evidence. The designates in Gen 9:8–17 a warrior’s קשת conclusion is reached that the word bow which represents both the deity’s might and power as well as his willingness to transfer his power over the earth to those living on it.

There are billions of neurons in our brains, but what are neurons? Just cells. The brain has no knowledge until connections are made between neurons. All that we know, all that we are, comes from the way our neurons are connected. Tim Berners-Lee, inventor of the web In 1906 Sir Charles Sherrington published The Integrative Action of the Ner vous System, which was a collection of ten lectures delivered two years before at Yale University in the United States.1 In this monograph, Sher- rington summarized two decades of painstaking experimental observations and his incisive interpretation of them and it is no exaggeration to say that Sherrington’s book changed the subsequent course of neurophysiology. He was one of the first to describe the human brain as “an enchanted loom where millions of flashing shuttles weave a dissolving pattern, always a meaningful pattern, though never an abiding one.” Since then, neurophysiology, neurobiology, and related disciplines have revealed the importance of neurons, the nerve cells in our brains specialized in connection and communications. We know now that our brains are complex networks in which new nodes preferentially attach to existing nodes that

1 Charles S. Sherrington, The Integrative Action of the Nervous System (New Haven, CT: Yale University Press, 1906). 138 ellen van wolde already have many nodes. When we acknowledge that connectivity characterizes the human brain, and all kind of human interactions including their literary products or texts, the question arises for our study of biblical and ancient Jewish literature: How to acknowledge ancient literary texts in their connectivity, examine them as meaningful patterns that developed out of human cognition in interaction with its cultural and natural environment? And how do we attach new nodes to these networks of existing nodes? The aim of this article is to consider the connectivity of words, grammar, and texts in close relation to cognition and to exemplify this cognitive linguistic approach in a study of Gen 9:8–17, the so-called rainbow-text of God’s covenant with Noah and the earth.

Bridging the Gap between Linguistics and Literary Studies: A Cognitive Linguistic Approach

The goal of a cognitive linguistic approach is to study meaning embedded in language, culture, and cognition by exploiting verifiable semantic methods for the analysis of linguistic expressions. In contrast to other, non-cognitive types of semantic research which is lexicographically (language-internally) oriented, cognitive linguistics examines words, concepts, and texts a. as embedded in cognition, that is to say, in semantic relationships (i.e., language-internally), in cognitive configurations (i.e., figuring in experi- ential, conceptual, and social routines), and in metonymical and metaphorical networks of meaning; b. in relation to the world views and cultural categories of the times in which they functioned; c. in relation to the communicative contexts of use. These basic tenets explain why cognitive linguistics cannot accept a dictionary-view of meaning, but takes an essentially encyclopedic view of meaning in which even the meaning of common, everyday terms is seen as supported by a vast network of interrelated knowledge. The cultural concerns of cognitive linguistics are obvious: the study of dictionary and encyclopedia as closely related phenomena is, in fact, a study of the rich repository of cultural knowledge. Consider, for example, the concept of the word ‘island.’ In our common understanding, an island is a mass of land completely surrounded by water. In a sense, an island is nothing more than ‘land.’ The land is what the word profiles, or designates. However, it is not the land as such, but linguistics & literary studies of ancient biblical & jewish texts 139 the land enclosed by water which is designated. In other words, the notion of surrounding water is intrinsic to the concept [island], in the sense that the island cannot be conceptualized without reference to a water mass. If there were no surrounding water, there would be no island. Conse- quently, the word ‘island’ designates a profile-base relation, which itself presupposes the broader cognitive domain of the earth’s geographical features.2 In an expert view on the earth, that is to say, not from a common sense perspective but from a geological perspective, an island is a mountain on the sea-floor or on the ocean-floor with its top above the water line. Hence, the term ‘island’ profiles a mountaintop above the water line and includes in its base an under-sea mountain, and this profile-base relationship figures in the cognitive domain of historical geology. This example elucidates how a word’s meaning is closely related to the cognitive domain on which it stands out. A cognitive domain is defined as any knowledge configuration that provides the context for the conceptualization of a semantic unit. In this definition, a distinction is made between the domain against which concepts take shape and the base on which an entity is profiled. The base of an expression is the conceptual content that is inherently, intrinsically, and obligatorily invoked by the expression. It comprises the full array of conceptual content that it specifically evokes and relies upon for its characterization. A cognitive domain is a more generalized ‘background’ knowledge configuration against which conceptualization is achieved. The semantic unit ‘island’ profiles the conceptual entity [mass of land] and this profiling takes place against a conceptual base [surrounding water] and the profile- base relation is what constitutes the semantic value of a word. The cognitive domain in which this relation functions is [geography]. However, the cognitive domain so far described as [geography] is commonly not figuring as the single background of the profile-base relation, but is part of a complex of related domains. For example, to understand the concept of the Hawaiian Islands, the concept of Eyjafjallajökull, and the concept of Moruroa means something completely different. In the first case, the Hawaii case, the cognitive domain [geography] is prototypically extended with the cognitive domain [holiday]; in the second case, the cognitive domain [geography] is prototypically extended with the cognitive domain [volcanic activity] and, since its eruption in 2010 with

2 John R. Taylor, Cognitive Grammar (Oxford Textbooks in Linguistics; Oxford: Oxford University Press, 2003), 199. 140 ellen van wolde

[disruption of flights and holidays]; and in the third case, the cognitive domain [geography] is prototypically extended with the cognitive domain [nuclear testing]. The usage of the word ‘island’ depends, therefore, on multiple domains that figure in a specific communicative context and these domains constitute together the configuration of knowledge. Typically, these cognitive domains overlap and interact in numerous and complex ways. This matrix of cognitive domains constitutes the general schema against which a word stands out as its instance. In Figure 1, the bold circle represents the semantic unit profiling an entity (profile P), the box represents the base (base B), and the D with single or double accents stand for a cognitive domain, and the complex of D’–D’’’ for the matrix of cognitive domains. It is clear that cognitive linguistics is a usage based approach. It is an approach to language that takes into consideration the various components of cognition that should provide us with an instrument for interpreting language in a way that can reveal the thought that is implicated in it. The approach presented here is adapted from Ronald Langacker,3 and

d’’

d’ P

d’’’

Figure 1

3 Ronald W. Langacker, Foundations of Cognitive Grammar, Volume 1: Theoretical Prereq uisites (Stanford, CA: Standford University Press, 1987) and idem, Foundations of Cognitive Grammar, Volume 2: Descriptive Application (Stanford, CA: Stanford University Press, 1991). linguistics & literary studies of ancient biblical & jewish texts 141 elaborated and applied on biblical concepts and texts in more detail in my book Reframing Biblical Studies.4 One of the fundamental assumptions of this approach is that language is used within culturally determined categories of conceptualization. Language, as a cultural given that people use, is not only shaped by the mind but itself shapes the mind. And acts of language and acts of cognition are events that take place in contexts of conceptualization—in what are above described as the profile-base relationships in cognitive domains. In order to understand an act of language, one must be familiar not only with the language being used; one must be aware of the social and/or literary context in which the language is used, and one must know as much as possible about the cultural and conceptual realities that are embodied and evoked by that language. Hence, meaning is considered to be both conceptual and contextual. In a textual usage event or text, language units are not merely set alongside each other, but are integrated into larger, hierarchically arranged compositions. These literary relations are as determinative for the mental image of a linguistic unit as the linguistic relations and their function in the usage event. One of the outstanding features of Langacker’s Cognitive Grammar is that he is able to explain the interdependence of the linguistic and literary features through the concept of valence relation. He defines an item’s valence as its disposition to combine with other items. In chemistry it has long been recognized that the valence properties of atoms—their combinatory potential—can only be understood and explained with reference to their internal structure. The matter is not quite so clear-cut in linguistics owing to the prime importance of conventional- ity. How a symbolic unit interacts with others cannot be predicted in any absolute way from its internal structure alone, because at any one time the established conventions of a language strongly sanction only a relatively small proportion of the potentially available combinations. Nevertheless, it is the internal structure of component expressions that determines the vast range of potential combinations from which the selection is made, as well as the likelihood that particular sorts of combinations will actually be conventionally exploited. Valence relations between atoms are based on the sharing of electrons. The valence relations uniting linguistic expressions depend in similar fashion on the sharing of elements. It is only by virtue of having certain substructures in common that two component expressions can be integrated to form a coherent composite expression. To the extent that we regard the component structures as distinct and separable entities, we can speak of correspondences between their shared substructures, that is to say, between

4 Ellen J. van Wolde, Reframing Biblical Studies: When Language and Text Meet Culture, Cognition, and Context (Winona Lake, IN: Eisenbrauns, 2009). 142 ellen van wolde

certain substructures within one and those substructures within the other to which these are construed as being identical. Such correspondences are found at both the semantic and the phonological poles, and they constitute the one invariant feature of valence relations.5 When two or more linguistic units combine to form a more elaborate expression, they enter into a grammatical valence relation, in which an item’s valence is defined as its disposition to combine with other items. The structures that combine are called component structures or semantic substructures; the integrated entity that results, is called the composite structure.6 Linguistically, words are viewed as distinct and separable semantic substructures. Literarily, the sharing of corresponding elements in the semantic substructures allows for their integration into composite structures of meaning. in Gen ’מצבה Consider, for example, the sentence ‘the stone placed as a 28:18. The phrase represents an integrated conception of two components, ”,put“ ,שים Schematically, the verb type .שים and the verb מצבה the noun “place,” “set,” can have either a horizontal orientation, “put down,” “lay down,” or a vertical orientation, “set up.” Because of the semantic sub- -in a spatial domain, this stand מצבה structure of a vertical orientation of ,There .שים ing stone can enter in Gen 28:18 into a valence relation with and comes to מצבה merges its specification with the noun שים the verb share the same element of vertically oriented space; hence, “set up.” The valence relation between the instances depends, therefore, on the sharing of corresponding elements. Based on this cognitive linguistic approach, an integrated linguistic and literary analysis of Gen 9:8–17 will be offered of What do they profile; which) ענן and קשת a. the meanings of the words notions do they inherently include; in which cognitive domains do they figure?), b. relevant aspects of grammar and words combinations in clauses, such as ,נתן ב- and קום and the verbs אתכם and בין the prepositions c. textual patterns of meaning based on semantic substructures that are the base of a literary composite meaning structure.7

5 Langacker, Foundations of Cognitive Grammar, Volume 1, 277–278; bold letters are his. 6 Ibid., 277. 7 The following analysis of Gen 9:8–17 is earlier published in VT 63 (2013): 124–149, under the title “One Bow or Another?: A Study of the Bow in Genesis 9:8–17.” I am grateful to Brill Publishers for their permission to re-publish the analysis of Genesis 9 in the pres ent volume. linguistics & literary studies of ancient biblical & jewish texts 143

The Meaning of Words

?in Gen 9:8–17: A Rainbow קשת The Word is generally understood to designate a rainbow in Gen קשת The word 9:8–17.8 Textual arguments provided to support the rainbow option are a. god sets the bow in the clouds in verse 13, b. the bow appears in the clouds in verse 14, c. god sees the bow in the clouds in verse 16. Intertextual support for this view is found in Ezek 1:28, where the bow is said to shine in the clouds on a day of rain, and in Sir 43:12–13 and 50:7, where the Greek word τόξον, “rainbow” functions analogously to the word in Ezek 1:28. According to this view, this word expresses the notions קשת of [arc], [colors], and [sky],9 while including the notions of [sunshine] and [rain]; it figures in the cognitive domain of [meteorology].10 Since Julius Wellhausen, many scholars have wondered about the word’s meaning as ‘rainbow.’ The reason for doubt is the fact that out of times designates 72 קשת its 76 occurrences in the Hebrew Bible, the word unequivocally a warrior’s bow (or, rarely, a hunter’s bow). Pieter De Boer also poses another pressing question, namely why the Hebrew Bible does not have a distinctive term for a rainbow, and he suggests that it is because

8 See BDB, 906; HALOT III: 1155; NIDOTTE III: 1004; ThWAT VII: 223; Benno Jacob, Das erste Buch der Tora: Genesis (Berlin: Schocken Verlag, 1934; E.T. New York: Ktav Pub lications, 1974), 255; Umberto Cassuto, A Commentary on the Book of Genesis, Part I: from Adam to Noah Genesis I–VI/8; Part II: From Noah to Abraham: Genesis VI/9–XI/32 (Original Hebrew edition: 1944–1949; E.T. Jerusalem: The Magnes Press, 1961–1964), II, 136; Claus Westermann, Genesis, 1. Teilband: Genesis 1–11 (BKAT I/1; Neukirchen-Vluyn: Neukirchener Verlag, 1974), 632–635; Victor Hamilton, The Book of Genesis: Chapters 1–17 (Grand Rap ids, MI: Eerdmans, 1990), 317. See also Fabry in ThWAT VII, 218–225, for a description of in Gen קשת the long tradition of Judaic and Christian interpretations in which the word 9:12–17 is understood as a rainbow. 9 Conceptual contents are conventionally expressed by terms in small capital letters in between square brackets. 10 In cognitive terminology, the word ‘rainbow’ profiles the conceptual entities of [arc], [sky] and [colors]; this is called its ‘profile’. At the same time, the word ‘rainbow’ cannot be conceptualized without reference to weather conditions. If there were no sunshine or rainy weather, there would be no rainbow. Consequently, the word ‘rainbow’ inherently, intrinsically, and obligatorily invokes the notions of [rain] and [sunshine]; this is called its ‘base.’ The profile-base relation is what constitutes the semantic value of a word, which itself presupposes the broader cognitive domain of [meteorology], in which a ‘cogni tive domain’ is defined as any knowledge configuration that provides the context for the conceptualization of a semantic unit (for an extensive description and examples see Van Wolde, Reframing Biblical Studies, 51–103). 144 ellen van wolde rainbows in the ancient Near East, or in the present day Middle East, occur very rarely.11 Bioclimatically, Israel is divided into three zones: a subhumid zone with an annual rainfall of between 1000–400 mm, a semiarid zone in which the annual rainfall ranges between 400 and 200 mm, and an arid zone with 200–25 mm of rain.12 Assuming that many biblical texts were formed, edited, and transmitted in the semiarid zone of ancient Israel, the low annual rainfall would indeed explain the rarity of the meteorological phenomenon of a rainbow. Israel’s geographical latitude of 31º30´ N, in which the sun is positioned high in the sky, would also account for the fact that a rainbow is seldom perceivable. Does the rarity of the appearance of a rainbow in (ancient) Israel and in the Near East provide a plausible explanation for its infrequency in the language and literature, and in the iconography of the Middle East?13 That is quite possible. The only occurs קשת בענן thing we know for sure, however, is that the collocation only twice in the Hebrew Bible, namely here in Gen 9:14–16, where it is first introduced by God as “my bow that I set in the clouds” in verse 13 the bow that is in the“ ,הקשת אשר יהיה בענן ביום הגשם ,and in Ezek 1:28 clouds on a rainy day.” So, it certainly is not a conventional or fixed-word combination in the Hebrew Bible. Furthermore, apart from the rarity of the meteorological phenom- both ,בענן and קשת enon and of the linguistic collocation of the terms conceptual and textual elements seem to exclude the rainbow option. The conceptual content of words depends on the linguistic construal of meaning expressed in lexical items and on their cultural and communicative contexts of use. All languages construe experiences, ideas, artifacts, et cetera differently. The optic meteorological phenomenon that is called in English ‘rainbow,’ in German ‘Regenbogen,’ in Swedish ‘regnbåge,’ in Dutch ‘regenboog,’ and in Italian ‘arcobaleno,’ all build their linguistic expressions on the shape of this phenomenon as an arc and on its relation to rainy weather. French, Modern Hebrew, and Greek, on the other hand,

11 Pieter A.H. de Boer, “Quelques remarques sur l’arc dans la nuée (Genèse ix 8–17),” in idem, Selected Studies in Old Testament Exegesis (OTS 27; Leiden: Brill, 1991), 129–138, especially 134. 12 http://www.science.co.il/reports/biological-diversity-israel-1997.pdf, 20. 13 Scenes picturing thunder and rain on Syrian and Mesopotamian seals abound, but images of the rainbow have so far not been found. Cf. Maurits van Loon, “The Rainbow in Ancient West Asian Iconography,” in Natural Phenomena: Their Meaning, Depiction and Description in the Ancient Near East, ed. Diederik J.W. Meijer (Amsterdam: North Holland, 1992), 149–168. “The rainbow is a natural phenomenon that one would expect to encounter on the monuments of Western Asia, but that has remained elusive so far” (149). linguistics & literary studies of ancient biblical & jewish texts 145

and ουράνιο קשת בשמים ’,designate the same phenomenon as ‘arc-en-ciel τόξο, respectively. These languages base their linguistic expressions on the shape of an arc and on its place in the heaven or sky. Another Greek term ίριδα, Spanish ‘arco iris,’ and Portuguese ‘arco-íris’ construe this phenomenon in relation to its shape as an arc and in relation to its colors. Thus it becomes clear that linguistic expressions in different languages may be based on distinct elements and have different motivations. Still the conceptual contents of these linguistic expressions is the same: they all designate the concepts of [arc], [colors], and [sky], and include the notions of [sunshine] and [rainy weather], and understand it in the cognitive domain of [meteorology].14 In addition, language users will employ these linguistic units (i.e., fixed combinations of linguistic expression and semantic content) in various contexts in distinct ways. For example, the conceptual meaning attributed to a ‘rainbow,’ ‘Regenbogen,’ ‘arco iris’ on many Christian websites, represent God’s peaceful covenant with the world, whereas in the name-giving customs of primary schools in the Netherlands―of 123 schools, De Regenboog is the most common name for a primary school in the Netherlands [2009]―the rainbow represents an open climate for all ethnic and cultural groups, for all pupils with different talents and characters. In short, the conceptual content of [rainbow] and English קשת בשמים expressed in French ‘arc-en-ciel,’ Modern Hebrew ‘rainbow’ is partly determined by language conventions which select some conceptual characteristics in their combination with linguistic expressions, and is partly based on a vast complex of experience and (culture- based) knowledge that resonates in the minds of language users. קשת Returning to biblical Hebrew, the conceptual content of the words are [bow] and [clouds], in which [bow] might metonymically be ענן and and the question whether ענן extended to [arc]. The meaning of the word or not it may entail the concept of [rain] should be briefly considered. Robert Scott gives a survey of all words used to designate clouds in the expresses cloud or mist in ענן Hebrew Bible, demonstrating that the word general, cloud mass, cloud cover or cloud-stuff as extended rather than defined, but not ‘a cloud’ and seldom if ever cloud-bringing rain.15 The on the other hand is used to designate a distinct cloud, in most עב word

14 That is to say, someone who perceives and conceptualizes a rainbow will always include the notions of [arc], [colors] and [sky], but will express it according to the lin guistic conventions of the language he or she uses. 15 Robert B.Y. Scott, “Meteorological Phenomena and Terminology in the Old Testa ment,” ZAW 64 (1952): 11–25, especially 24–25. 146 ellen van wolde cases a rain-cloud (frequently plural). Edmund Sutcliffe demonstrates is used for water-carrying clouds.16 And finally, Mark עבים that the word is עבים or עב Futato confirms the previous conclusions that only the term designates the cloud ענן used to refer to rain clouds,17 whereas the word mass, cloud cover or undifferentiated cloud.18 Therefore, the conceptual ,[can be described as [clouds] and [cloud cover ענן content of the word while including the notions of [vastness of space] and [expanse].19 This in Gen 9:13–16 entails the content of ענן makes it implausible that the word .קשת rain], and therefore it cannot have this element in common with] designate in Gen 9:13–16 קשת בענן and קשת In sum, the lexical items the concepts of [bow] and [clouds], whereas the notion of [colors], [rain] and [sunshine] are neither present nor implicated by the text. קשת or קשת This makes it, from a conceptual point of view, unlikely that .designates a rainbow בענן Textual elements confirm this conclusion. First, it should be pointed out that people who tend to think that ‘rainbow’ makes sense in this text, because it follows after rain and flood, should be aware that the ‘rainbow’ appears quite a long time after the raining. The rain begins in Gen 7:12 and ends in 8:2. The flood is mentioned in Gen 9:11 and 9:15. However, according to the narrative, about a year elapsed from the end of rain in Gen 8:2

16 Edmund F. Sutcliffe, “The Clouds as Water-Carriers in Hebrew Thought,” VT 3 (1953): 99–103. 17 NIDOTTE III, 465–466. 18 In ancient Mesopotamia, rain is related to the Higher Heaven and Anu the god of High Heaven is responsible for falling of the rain; the Middle Heaven or Ešarra is the level of the winds and the storms of Marduk, and the Lower Heaven or visible sky contains the sun, moon and stars, and constellations. Hence, the winds in the Middle Heaven are con sidered to blow on a higher level than the level of the stars, that form the Lower Heaven; cf. Wayne Horowitz, Mesopotamian Cosmic Geography (Mesopotamian Civilizations; Winona Lake, IN: Eisenbrauns, 1998). Also in the Hebrew Bible, rain is connected with the highest heaven and with the Hebrew God who is responsible for the falling of the rain. Two texts are considered to be positioned above the sky or (עבים) seem to imply that rain clouds ;(עבים) namely Job 26:8, “He [God] wrapped up the waters in his clouds ,(ענן) cloud cover does not burst under their weight,” and Nah 1:3, “His [God’s] way (ענן) yet the cloud cover ”.are the dust of his feet (ענן) is in the storm and the whirlwind and clouds profiles the conceptual entity of [clouds] and ענן In cognitive terminology, the word 19 [cloud cover]; further research may elucidate whether it can also express the notion of [sky]. At the same time, this word ‘clouds’ or ‘cloud-cover’ cannot be conceptualized with out the notions of [vastness of space] and [expanse]; this is its ‘base.’ The profile-base relation [clouds], [cloud cover] / [vastness of space], [expanse] is what constitutes which itself presupposes the broader cognitive domain ענן the semantic value of the word of [ancient meteorology]. linguistics & literary studies of ancient biblical & jewish texts 147

,in 9:13.20 Also in the immediate context קשת בענן to the display of the only the flood is mentioned. Second, in Gen 9:8–17 God sets the bow in the sky as a sign for himself, a sign of the covenant between him and all life on earth: “when it appears, I will remember my covenant” (verses 14–15), “when the bow is in the clouds, I will see it and remember the covenant” (verse 16). The text does not say that this bow is a sign for humanity or even that it is visible to human beings.21 Since the story is narrated in two direct discourses introduced by “God said” in verse 12 and by “God said to Noah” in verse 17, we share God’s perspective only. And while sharing his point of view, we come to know that God sees the bow as a reminder of in verses 12, 13 and 17). Third, if the אות הברית his covenant (three times expressed the idea of a rainbow and if this bow functioned as קשת word a token that in the future would remind God of his covenant, what then is the conceptual connection between his bow, i.e., his rainbow, and his covenant? What distinctive features cognitively link these items together? That is to say, in order to function as a sign, one might expect the rainbow to have some element in common with a covenant. Not the colors, the rain, or the sunshine, because they do not figure in the text. The only element that comes to mind is the arc-like shape, but what does this arc-like form have in common with the covenant? And an even more pressing question, why would a rainbow be called ‘his’ rainbow?22 meaning קשת To conclude this section on the possibility of the word “rainbow,” we must acknowledge that its common meaning in the Hebrew Bible is “warrior’s bow” and that there would have to be strong evidence otherwise. However, the above mentioned קשת in Gen 9:8–17 to read

20 If we escape a little from the narrative to the real world, imagine a rainbow appear ing a year after the rain stopped! 21 If, in fact, the rainbow was seldom seen in ancient Israel, it could hardly have been thought of as a token that God might have regularly noticed, which would make the mean ing of ‘rainbow’ even more implausible. 22 Another question might be posed. Today, we know that a rainbow does not actually exist at a particular location in the sky, but depends on the observer’s location and the position of the sun. All raindrops refract and reflect the sunlight in the same way, but only the light from some raindrops that reach the observer’s eye constitute the rainbow for that observer. (For further information see: http://eo.ucar.edu/rainbows.) As a result, a deity who is envisioned as living above the atmosphere would be unable to see a rainbow at all. Although people in the ancient Near East did not have this scientific knowledge, they would have been aware of the fact that the perception of a rainbow is dependent on the observer’s location and the position of the sun. If the deity is thought of as living above the heavenly expanse (compare Ezek 1:26), and the sun, moon, and stars are conceived as moving below the heavenly vault, then it would be impossible for the deity to see a rainbow. 148 ellen van wolde meteorological, linguistic, conceptual, and textual issues do not confirm in Gen קשת this exceptional or unconventional meaning of the word -designates a rain קשת בענן or קשת and make it implausible that 17–9:8 bow in this text.

?in Gen 9:8–17: A Warrior’s Bow קשת The Word as a warrior’s bow קשת Some scholars have opted for the meaning of קשת in Genesis 9.23 Linguistically, the argumentation that the word expresses a warrior’s or a hunter’s bow is a strong one. Out of its 76 uses in the Hebrew Bible, this meaning is obvious in 72 texts.24 Apart from the in Ezek 1:28 in קשת three uses in the text under analysis and the use of relation to a rainy day (followed by Ben Sira in 43:11 and 50:7), all other in the Hebrew Bible clearly indicate the meaning of קשת uses of the word a warrior’s bow (rarely a hunter’s bow). In the context of fighting or war, this word appears in the Hebrew Bible with its prototypical meaning of [warrior’s bow], and this meaning necessarily includes the notions of [arrows], [weapon of attack], and [enemy], and is functioning in the cognitive domain of [war]. ,occurs seven times: three times in our text קשת In Genesis, the word twice in the sense of a warrior’s bow, and once in the meaning of a hunter’s bow, namely in Gen 27:3, where Isaac says to Esau, “Take your gear, your quiver and bow, and go into the open and hunt me some game.” In addition, in Gen 21:16, [Hagar] “went and sat down at a distance a bow- refers to either a warrior’s bow or a hunter’s bow; this קשת ”,shot away ”an “archer , קַ ּׁשָ ת is equally true for Gen 21:20, where Ishmael is called a warrior’s bow,” are Gen 48:22“ ,קשת or “bowman.” The two clear uses of and 49:24. In Gen 48:22, the dying father Jacob speaks to Joseph saying, “And now, I assign to you one portion more than to your brothers, which I wrested from the Amorites with my sword and bow” (NJPS translation). ”,from the hands of the Amorites (לקח) Literally, the text speaks of “taking ,קשתי my sword,” and“ ,חרבי and together with the word combination of “my bow,” it is obvious that the context is one of fighting or war, that the

23 De Boer, “Quelques remarques sur l’arc dans la nuée”; Erich Zenger, Gottes Bogen in den Wolken: Untersuchungen zu Komposition und Theologie der priesterschriftlichen Urge schichte (SBS 112; Stuttgart: Verlag Katholisches Bibelwerk, 1983); Udo Rüterswörden, “Der Bogen in Genesis 9: Militärhistorische und traditionsgeschichtliche Erwägungen zu einem biblischen Symbol,” UF 20 (1988): 247–264; Christoph Uehlinger, “Das Zeichen des Bun des,” BiKi 44 (1989): 195–197. 24 DCH VII, 339–340. linguistics & literary studies of ancient biblical & jewish texts 149 described weapons are weapons of attack, and that these weapons are closely connected with its possessor, who defeats his enemies. As a consequence of his powerful victory, Jacob was able to apportion the land to Joseph’s sons. In Gen 49:23–24, Jacob continues his speech to Joseph saying, “Archers bitterly assailed him [Joseph]; they shot at him and harried him. Yet his bow stayed taut, and his arms were made firm by the hands ”,his bow“ ,קשתו of the Mighty One of Jacob” (NJPS translation). The word to stay,” “abide,” “sit,” plus“ ,ישב is used here in combination with the verb ,permanent,” “enduring”: his bow abode as an enduring“ ,באיתן the adverb were made firm by the (זרעי ידיו) firm one, and the arms of his hands of Jacob. So here again, the bow is closely (אביר) hands of the Mighty One linked to its owner as a representation of his enduring power. We may warrior’s bow,” this“ ,קשת conclude that in Genesis’ two uses of the word word figures in the cognitive domains of [war] and [dominion] and designates a [warrior’s bow], while including the notions of the warrior’s [victorious power] and [might].25 Culture, place, and time-bound information related to the way wars are fought, the structure of an army or troop, the person and function of an archer, the production and use of weapons, the development of warfare, and so on, might enlighten us about the configuration of knowledge in which this bow functions. From texts in the Hebrew Bible, we learn that the bow, bowstring, and arrows are used as a weapon of attack by powerful fighters—mainly kings, commanders or other powerful persons—and often figure in a war-winning situation.26 In the conventional conceptualization of the series of actions and events in ancient warfare (i.e., in the prototypical scenario of warfare),27 the bow-strike seems to mark the decisive blow in the final stage of war. This is visible in the books of Samuel and Kings, because here the bow is used by kings or their archers when describing their victory over the enemies.28 A comparable use of the word

[profiles in these two Genesis texts a person’s [warrior’s bow קשת Hence, the word 25 and includes in its base the concepts of [victorious power] and [might]. 26 Fabry in ThWAT VII, 222. 27 This is called a ‘prototypical scenario’, that is, the most representative series of actions, events, or behaviors in a culture that constitutes the content of an idea. Such a scenario is a reflection of the conventional behavioral sequences in a culture, of the pat terns of actions, situations or events. 28 See 1 Sam 31:3 and its description of the final blow the Philistine archers strike on ,קשת Saul when hitting him with their bow; 2 Sam 1:18–27 and David’s ‘Song of the Bow’ or a song of military power and glory of Israel; 2 Samuel 22, where David praises Yhwh, who has saved him from the hand of his enemies, saying in verse 35, “who trained my hands for the battle, so that my arms can bend a bow of bronze.” Here the bow represents the victory 150 ellen van wolde

is noticeable in the Psalms, where metaphors depict God as a heav קשת enly warrior who fights with warrior attributes. Based on a close study of these Psalms, Martin Klingbeil29 shows how God is characterized as being actively involved in the actual attack on an enemy in a war-winning situation and how this functions as an expression of his might and supremacy.30 And he concludes: “Although Yahweh acts as a warrior he never becomes the warrior-god exclusively, he rather remains the supreme god displaying warrior characteristics, but not in the form of mere emblem which would designate him as the war-god.”31 Because the bow is such a distinctive mark of warfare and of the winner who demonstrates his rule and dominion, it comes as no surprise that the end of the war is represented by the “breaking of the bow.” See, for example, in Ps 46:10–11: “He puts a stop to wars throughout the earth, breaking the bow, snapping the spear, consigning weapons to flames. ‘Desist! Realize that I am God! I dominate the nations; I dominate the earth’ ” (NJPS translation). Also in Isa 41:2–3, where Cyrus’ bow defeats the enemies and brings them down to dust, the text refers to both Cyrus’ sword and bow, his might, as well as to Yhwh who has helped him in his victory. We can, therefore, draw the designates קשת conclusion that throughout the Hebrew Bible the word prototypically a [warrior’s bow], while including the notions of [weapon of attack], [arrows], and [enemy], as well as those of [power], [might] and [supremacy]. Thus it becomes clear that it figures not in one, but in two cognitive domains, namely [war] and [dominion].32 It is used in a prototypical scenario of fighting events in which the bow strike marks the

of David as a warrior. Similar war-winning situations are expressed by strikes of victorious bows in 1 Kgs 22:34 (the bow of an archer of the king of Aram kills Ahab, so that he dies of its wounds); 2 Kgs 9:24 (Jehu kills king Joram and thus becomes king); 2 Kgs 13:14–18 (Elisha tells King Joash of Israel to take his bow to defeat Aram, which he does while exclaiming in verse 17: “An arrow of victory for Yhwh! An arrow of victory over Aram!”). 29 Martin Klingbeil, Yahweh Fighting from Heaven: God as Warrior and as God of Heaven in the Hebrew Psalter and Ancient Near Eastern Iconography (OBO 169; Fribourg/Göttingen: Universitäts-Verlag/Vandenhoeck & Ruprecht, 1999). takes (מיתר) See, for example, Ps 21:13, where God as a warrior with his bowstring 30 aim against the faces of the enemies. 31 Klingbeil, Yahweh Fighting from Heaven, 302. 32 In the Hebrew Bible, the bow is not used with a connotation of [protection]. This a weapon ,(מגן) conceptualization of the bow stands in sharp contrast to that of a shield of defence that figures in the cognitive domains of [war] and [protection]. See, among others, Ps 18:35–36: “[the God] who trained my hands for battle; my arms can bend a bow of bronze. You have given me the shield of your protection; your right hand has sustained me, your care has made me great” (NJPS translation). linguistics & literary studies of ancient biblical & jewish texts 151 decisive blow in the final stage of war, and thus comes to stand for the [victorious weapon] of the winning warrior.

Iconography: ANE Pictures of a Deity and His or Her Warrior’s Bow Iconographic information may help us to build up a mental image of ancient Near Eastern people and their visualizations of gods or goddesses with a bow. The iconographic data presented in this section are based on Klingbeil, who examined ninety three ANE pictures that visualize God as warrior or as God of heaven. He distinguishes four motif groups in the images of gods as warrior: (1) the smiting god, (2) the god in archer scenes, (3) the god with the spear, and (4) the god in arms. He concludes, It is interesting to note that the active warrior scenes, i.e., images on which the god is actively involved in warrior activities like smiting, shooting, thrust- ing, are largely limited to the end of the Late Bronze Age and the beginning of the Iron Age. [. . .] The passive warrior scenes, i.e., the god in arms motif, have been limited to the Neo-Assyrian period, while only one sample comes from the Syro-Palestinian region.33 Seven Neo-Assyrian cylinder seals which could be dated to the late 9th and early 8th century bce, offer scenes that show either the gods Ninurata and Adad or the goddess Ištar carrying bows crosswise over their shoul- ders or pointing a bow.34 All of them show worshippers facing the god or goddess who are raising their hands in adoring gestures toward the god. On one seal, Ištar holds the bow in her hand, while facing a female adorant.35 On another seal, an armed bearded god (Ninurta or Adad) is holding a bow in his hand.36 Klingbeil describes another series of five pictures with the weather-god in arms or the sun-god in arms. The first is a limestone stela from Babylon (dated 823–811 bce) that presents a scene with three figures.37 On the left, the goddess Ištar raises her one hand in a gesture of blessing, while the other hand is extended toward the front, holding the tip of a bow and ring. In the middle, the god Adad is holding a bundle of lightning in both

33 Klingbeil, Yahweh Fighting from Heaven, 195–196. 34 Klingbeil, Yahweh Fighting from Heaven, 188–196, presents and describes the seven seals in seven Figures 21–27. Only Figure 21 contains a deity pointing a bow (ibidem, 188–189). 35 Ibidem, Figure 26, described on 194. 36 Ibidem, Figure 27, described on 195. 37 Ibidem, Figure 84, described on 252–253. 152 ellen van wolde hands. On the right, a worshipper who is dressed like an Assyrian king is facing him. The relief on the famous ‘broken obelisk’ from Nineveh shows a personification of the winged sun disk and is dated to the 10th century bce.38 The scene shows the Assyrian king (?) holding ring and rod in one hand, while the other is open in a gesture of receiving adoration. In front of him, four enemies are visible in a submissive posture, having raised their hands in a gesture of adoration. They have a ring through their nose to which a rope is attached which is held by the king. However, the most interesting aspect of the image is the top part of the scene. Above the captives, a winged sun disk is visible. From the wings, two hands are coming forth, one holding a bow as if to hand the weapon to the king, while the other is extended with an open palm toward the king in a blessing gesture. The god represented by the image is the Assyrian solar deity Šamaš. In presenting the bow to the king, he is aiding the monarch in warfare and the conquest of his enemies, thus transferring military power to the ruler. A more anthropomorphized variant of the motif can be found on a bronze disk with unknown provenance which can be dated to the 9th century bce.39 The center image consists of three figures: two giants are kneeling on stylized mountains supporting the wings of a sun disk from which the upper body of a male god is emerging. The god is holding a bow in front of him, while the other hand is raised and extended toward the front in a blessing gesture. The connection of the sun disk with the god with a bow appears to identify him as a solar deity with military connotations which is especially apparent with the association of the motif with scenes of war and victory. The fourth picture is the well-known glazed tile from Aššur, found in the Anu-Adad-Temple, dating to the time of Tikulti-Ninurta II (890–884 bce), now in the British Museum.40 The sun disk is depicted as encompassing

38 Ibidem, Figures 87 and 88, described on 257–259. The original is housed in the British Museum, BM 118898, and has been published in ANEP 300, no. 400. 39 Klingbeil, Yahweh Fighting from Heaven, 259, only reproduces the center image in the line drawing of Figure 89 and describes it on 259–260. The design is mostly Assyrian but exhibits also Urartian elements, e.g., the form of the chariots and dress of the drivers. 40 BM: 115706. This tile has been repeatedly published: in black-and-white in ANEP, 536, in color in Walter Andrae, Coloured Ceramics from Ashur And Earlier Ancient Assyrian Wall-Paintings from Photographs and Water-Colours by Members of the Ashur Expedition organised by the Deutsche Orient-Gesellschaft (London: Kegan Paul, Trench, Trubner & Co. Ltd., 1925), Plate 8, 27, which is reproduced in André Parrot, Nineveh and Babylon (London: Thames and Hudson, 1961), Figure 282. It is extensively studied in Ruth Mayer- Opificius, “Die geflügelte Sonne: Himmels- und Regendarstellungen im alten Vorderasien,” linguistics & literary studies of ancient biblical & jewish texts 153 the winged god completely, while there are rays or flames of fire depicted within the sun nimbus. With his hand the god is holding a bow which he has stretched to its limits, pointing at an imaginary or at least unseen enemy, since the scene has been broken off on the right side. A chariot scene is visible below, of which only the head of the charioteer and the upper part of the horse’s head are visible. Around the winged sun disk, there are stylized clouds with raindrops suspended from the upper border below an inscription. The association of the god with rain-clouds demonstrates his identification with a storm- and weather-deity, while the wide wings symbolize the dark thundering heaven. A rather intricate combination of warrior and god of heaven attributes is present in this scene. It creates a complex image of the god fighting from heaven with meteorological weapons. As for the identification of the winged god in the sun disk, we would follow Mayer-Opificius, who has convincingly shown that the god represents Šamaš rather than Aššur. A relief from Nimrud from the North-West palace of Asshurnasirpal II (883–859 bce) shows another solar god with a bow and arrow.41 Here the winged sun disk is depicted as a ring which has been placed over god’s upper body and the god is aiming a bow toward the right side at an enemy, while bowstring and arrow are clearly distinguishable. This ancient Near Eastern pictographic material gives us additional information on how people may have thought of a warrior god. If they had viewed him in images comparable to those of the solar deity (Šamaš), who is holding a bow in front of him while the other hand is blessing, or who presents his bow to the king and is thus aiding the monarch in warfare and the conquest of his enemies, they might have thought of the deity in terms of someone who is in charge and who is transferring his power to the ruler. Or if they had viewed him in images comparable to those of the weather-god Adad, the god Ninurta, or the goddess Ištar, who are represented as powerful deities carrying bows crosswise over their shoul- ders, they might have thought of them as deities whose weapons testify to their power. The conceptual content that all these pictures share with each other is that the god with the warrior’s bow expresses [victorious power] and [might] in the cognitive domain of [dominion]. In addition, some pictures of the sun-god with a bow indicate a [transfer of

UF 16 (1984): 189–236, especially Figure 25, and in Klingbeil, Yahweh Fighting from Heaven, Figure 90, described on 260–261. 41 Klingbeil, Yahweh Fighting from Heaven, Figure 91, described on 261–262. 154 ellen van wolde power], whereas other pictures of Adad, Ninurta and Ištar with a bow apparently function in the cognitive domain of [cult] as is clearly visible in the depicted adorant-worshippers.

Intertextuality: Mesopotamian Texts and Astronomy Intertextual relationships of the biblical flood story with Mesopotamian texts are equally instructive. Stéphanie Anthonioz demonstrates that in the available ancient Near Eastern flood stories, the flood is described as a great battle between divine beings and the winds of the flood in which the bow is the crucial instrument with which the deity or flood hero defeats the dangerous and bad powers of nature.42 In The Myth of Anzû, Ninurta is the hero who fights seven fights and defeats all seven bad winds. At the heart of the story describing the turning point in the fight (Tablet II 53–60), the text describes how Ninurta defeated the flood. The powerful deity “drew his bow (qaštu) and armed it with an arrow, from the middle of the bow he sent his arrow.”43 Thus he calmed the flood and made it powerless. In Ninurta and the Stones, a similar fight against the winds and the monster of the deluge is pictured, in which a long bow again takes up a crucial position. In Enuma Elish (IV 35–50), Marduk’s fight against Tiamat is described: 35 il façonna l’arc, révéla son arme, 36 Il fit chevaucher la flèche, il plaça la corde, 37 il brandit la masse, (la) saisit de sa droite, 38 Il suspendit à son côté l’arc et le carquois, 39 Il plaça l’éclair devant lui, 40 Il emplit son corps de flammes ardentes, 41 il fit le filet (pour y mettre) Tiamat intacte au-dedans, 42 Il prit les quatre vents, rien d’elle ne (pourra s’en) sortir, 43 Le vent du sud . . .44 Marduk’s competency as an archer brings Tiamat to an end, and his victory is marked by the verb nâhu: it is not the sea that is brought to rest,

42 Stéphanie Anthonioz, “Noé ou le répos du guerrier,” RB 117 (2010): 185–199, especially 199. 43 See Amar Annus, The Standard Babylonian Epic of Anzu (SAACT 3; Helsinki: The Neo-Assyrian Text Corpus project, 2001), 23–24, and Anthonioz, “Noé ou le répos du guer rier,” 191. 44 Translation by Philippe Talon, ed., The Standard Babylonian Creation Myth Enūma Eliš (SAACT 4; Helsinki: The Neo-Assyrian Text Corpus Project, 2005), cited by Anthonioz, “Noé ou le répos du guerrier,” 197. linguistics & literary studies of ancient biblical & jewish texts 155 but the hero, the Lord himself. Anthonioz concludes that the story of the flood with the protagonist and hero Noah opens and closes the story with this same concept of [rest]. In the section immediately following our text in verses 12–17, Noah receives the rest of a warrior who gets a vineyard after finishing his active service as the hero of the flood.45 Various Mesopotamian texts narrate how the bow, after being used by deities as a successful weapon of attack, becomes a victorious sign of the war won over the enemies and how it is set in the sky as the Bow Star, qaštu. The most important testimony of this view is Enuma Elish VI 80–94, where Anu, the deity of the Upper Heavens, presents his bow to the other gods and praises its power; he calls the bow his daughter and gives it three names: “long bow,” “conqueror,” and “Bow Star,” and he lets it appear in the sky as a constellation. Finally, he gives this bow a place in the assem- bly of the gods. 80 Les grands dieux, à cinquante, prirent place 81 les dieux des destins, à sept, fixèrent les sentences. 82 Le Seigneur présenta alors l’Arc, son arme, et le jeta devant eux, 83 ils virent le filet qu’il s’était fait, les dieux, ses pères. 84 ils virent l’Arc, combien était parfaite sa forme, 85 ses pères louèrent les actes qu’il avait accomplis. 86 Anu l’éleva et prit la parole dans l’assemblée des dieux, 87 ayant embrassé l’Arc, (il dit): «C’est bien ma fille!» 88 de l’Arc, il nomma ainsi les noms: 89 «Long bois», que ce soit le premier; que le deuxième soit «Conquérant», 90 son troisième nom est «Etoile de l’Arc», il la fit apparaître dans le ciel, 91 il fixa sa position parmi les dieux, ses frères. 92 après qu’Anu eut décrété les destins de l’Arc, 93 il instaura un trône royal, élevé parmi les dieux, 94 et Anu le fit s’asseoir, lui, dans l’assemblée des dieux. <. . .>?46 Enuma Elish VI 80–94 shows us the process of metaphorical conceptualization of qaštu, “bow.” This word begins its life in the cognitive domain of [war], where it expresses a [warrior’s bow]. It is then extended to signify a [sign of victory], still figuring in the same domain. Finally, it is transferred to the target domain of [astronomy] and [deities], and comes to express a bow or an arc-like constellation, the [Bow Star].

45 Anthonioz, “Noé ou le répos du guerrier,” 199. 46 Talon, The Standard Babylonian Creation Myth Enūma Eliš, 101. 156 ellen van wolde

Based on this text, Horowitz describes the Bow Star as the final addition to the universe in Enuma Elish and explains that in Astrolabe B (KAV 218 B I 14–16) the Bow Star is identified with Ištar of Elam and is called the daughter of Anu.47 Also the astronomical text 5R 46 23 identifies the Bow Star as Ištar of Babylon. In Mesopotamian iconography, the bow is clearly the standard attribute of Ištar. Walker mentions an Old Babylo- nian prayer to the “gods of the night,” which contains a list of stars, and qaštum the Bow Star is one of them.48 Rochberg mentions the Babylonian namburbû ritual, which is a ritual against the “evil of the bow,” which includes offerings to be made to the god Ea and the goddess Ištar.49 There- fore, the semantic content of [warrior’s bow] is valid here, but it is metaphorically extended to the constellation Bow Star.50 The Bow Star Qaštu is identified by Erica Reiner and David Pingree as the Canis Maior.51 This conspicuous constellation lies mainly just south of the Milky Way between Orion and the long train of bright stars from the old Argo Navis. The constellation is dominated by the bright star Sirius, the brightest appearing star in the sky. It is also known that the ancient Egyptians based their calendar on its yearly motion around the sky, so that it is certain that both in ancient Mesopotamia and in ancient Egypt this Bow Star was an important constellation.52

’Warrior’s Bow‘ קשת Arguments Given so Far for the Word in קשת Let me evaluate the arguments in favor of the view that the word Gen 9:8–17 designates a [warrior’s bow]. The first argument is a linguistic one. In 72 out of 76 occurrences in the -certainly designates a weapon of attack while includ קשת ,Hebrew Bible ing the notions of the warrior’s [victorious power] and [might]. The

47 Cf. Horowitz, Mesopotamian Cosmic Geography, 124. 48 Cf. Christopher Walker, “The Myth of Girra and Elamatum,” AS 33 (1983): 145–152, especially 146. 49 Cf. Francesca Rochberg, The Heavenly Writing: Divination, Horoscopy, and Astronomy in Mesopotamian Culture (Cambridge, NY: Cambridge University Press, 2004), 56. 50 Horowitz, Mesopotamian Cosmic Geography, and Rochberg, The Heavenly Writing, in Gen 9:12–17 from ‘warrior’s bow’ to the קשת extend therefore the meaning of the word constellation Bow Star. 51 Erica Reiner & David Pingree, Babylonian Planetary Omens, Part 2 (Malibu, CA/ Groningen: Undena/Styx Publ., 1981), 11. 52 According to Rochberg, The Heavenly Writing, 56, n. 31, a new identification of Qaštu as Canis Minor (with Procyon, the eighth brightest star in the sky) by Hermann Hunger & David Pingree, Astral Sciences in Mesopotamia (Leiden: Brill, 1999) replaces the earlier identification. linguistics & literary studies of ancient biblical & jewish texts 157

in its common, prototypical meaning of a warrior’s bow figures קשת word in the cognitive domains of [war] and [dominion]. 53,קשת The second argument is a textual one. In the biblical texts with is used in a prototypical scenario of fighting-events in קשת the word which the bow strike expresses the decisive blow in the final stage of war, and thus comes to stand for the [victorious weapon] of the winning warrior. The third argument is a cultural one. In the ancient Near Eastern mind, a bow may be carried by humans or deities. Iconography has elucidated how deities are pictured with bows. It thus became clear that all discussed pictures are sharing the same conceptual content: the scene with a god armed with a warrior’s bow all express [victorious power] and [might] and stands out against the cognitive domain of [dominion]. In addition, some scenes with the solar god with a bow also express a [transfer of power], whereas others with the deities Adad, Ninurta and Ištar with a bow predominantly function in the cognitive domain of [cult], as is clearly visible in the depicted adorant-worshippers. The fourth argument is an intertextual one. Various Mesopotamian texts narrate how the bow, after being used by deities as a successful weapon of attack, becomes a victorious sign of the war won over the enemies of the flood. Here again, the conceptual content of the deity’s warrior-bow is [victorious power] and [might], and figures in the cognitive domains of [war against the flood] and [dominion]. The final argument is based on astronomy. In various Babylonian texts, it is told how the divine bow as a victorious sign of the war won over the enemies is set in the sky as the Bow Star, qaštu. Thus, this constellation, which is equally important in ancient Mesopotamian and Egyptian astronomy, represents [victorious power] and [might], and figures in the cognitive domain of [dominion]. These distinct data sets demonstrate that the prototypical meaning of a warrior’s bow which might be carried by humans or deities, is a weapon of attack, and they also illustrate which ideas are closely connected with it, namely [victory], [enduring power] and [might], concepts that not only figure in the cognitive domain of [war] but also in the cognitive domain of [dominion].

53 Gen 48:22 and 49:23; 1 Sam 31:3; 2 Sam 1:18–27; 2 Sam 22:35; 1 Kgs 22:34; 2 Kgs 9:24; 2 Kgs 13:14–18, in the metaphors of God as warrior in the Psalter and in prophetic texts such as Isa 41:2–3. 158 ellen van wolde

Word Combinations in Clauses in Gen 9:8–17

I have set my bow in the clouds,” in“ ,את קשתי נתתי בענן In the verbal clause specific because) קשתי relates the specific lexical unit נתן verse 13, the verb which appears to ענן of the pronominal suffix ‘my’ in “my bow”) to the word be a general, unrelated, and extensive cloud mass or cloud cover. Semanti- expresses the act through which an object or נתן ב- cally, the collocation of matter is set in motion “to place something in.”54 The difference between verse 13, on the one hand, and verses 14 and 16, on the other hand, is that in verse 13 the deity moves his bow on a path that starts in his hands and ends in the clouds, whereas in verses 14 and 16 the bow is not moved anymore, but is actually defined by its place in the clouds. In all three verses, the temporal relation between something very specific, God’s bow, and something .ב- general, the cloud cover, is marked spatially by the preposition .בענני ענן על הארץ Verse 14 presents the unique word combination Nowhere else in the Hebrew Bible is it stated that God “clouds clouds” or “brings in clouds,” and it stands out as extraordinary. What does it mean? First, it is God who is bringing the cloud mass, “my clouding clouds.” Second, God’s bringing of clouds is explicitly related to the appearance of the bow, “my bow and my bringing of clouds.” Third, these clouds are related to the earth: “my bringing of clouds over the earth.” The conceptual content of the which was described as [clouds] and including the notions of ענן word [vastness of space] and [expanse], is now contextually specified: it derives over the earth.” In other“ על הארץ its specification from the clouds’ position words, the meaning components [vastness of space] and [expanse] in the -enter into a valence relation with the meaning components [sur ענן word over the earth,” and indicate“ על הארץ face] and [vastness of space] in that the clouds are conceived relative to the entire earth. The semantic components of verses 13 and 14 build up into a more complex meaning structure. The divine actions of “setting” and “clouding” are both spatially oriented. The setting of his bow in the clouds in verse 13 shows that God separates from his bow and that he places it somewhere between his dwelling place (the heaven) and the earth. Verse 14 shows that this place is the clouds that are presented as covering the entire

54 Cf. Casper J. Labuschagne in THAT II, 117–141, especially 119, and Michael A. Grisanti in NIDOTTE III, 205–211. To put it in cognitive linguistic terms, in verse 13 God initiates an action of setting, which builds in its valence structure a relation to a place left and a goal in” itself expresses the notion of a container: it designates“ ב- aimed at. The preposition designates a spatial נתן בענן the fact of finding oneself in or within a place. The collocation movement that the subject, God, performs on his bow along a path, which localizes the clouds as the final position to which the subject orientates his movement that culminates inside the clouds. linguistics & literary studies of ancient biblical & jewish texts 159

earth. Therefore, the bow is spatially positioned in the intermediate space between the deity and the entire earth. And it is this bow that is said to represent Gods covenant between himself and the earth. An even wider network of semantic relations becomes visible when we take into account the entire section of Gen 9:8–17 in Figure 2. The following correspondences between the meaning components form a clear combinatory pattern: (1) There are only two subjects of action, viz. “I” in verses 9–13a, 14a, 15, 16b, 17a, and “my bow” in verse 13b or “the bow” in verses 14b and 16a, suggesting that “my bow” or “the bow in the clouds” is the representative of the deity who is speaking in the first person. (2) There is the fivefold use of the preposition with pronominal suffix with you.” The deity relates his action to Noah, his family and“ אתכם his descendants; this is the main partner, while the other partners are share the נתן ב- and קום conceived in relation to them. (3) The verbs same spatial concept of ‘setting up’ or ‘establishing’ and are set in similar syntactic constructions. In the first clause, the active participle of the look, I am establishing my“ ,אני הנני מקים ,’verb is linked to the subject ‘I ”I am giving the sign of the covenant“ ,אני נתן covenant” (verse 9a) and (verse 12a), while in the second clause the qatal verb indicates the actual

Partners Objects Subject and Actions Gen 9 Verse 9 אֲ נִ֕ י הִ נְנִ֥י �מֵקִ ֛ ים אֶת־ּבְרִיתִ ֖ י אִּתְ כֶ ם וְאֶ ֽת־זַרְ עֲכֶ ֖ם אַ ֽחֲרֵיכֶ ֽ ם 10 וְאֵ֙ת ּכָל־נֶ ֤פֶ ׁש הַ ֽחַ ּיָה֙ ֣ ראֲׁשֶ אִּתְ כֶ֔ ם . . . ּובְכָל־חַ ֥תּיַ הָאָ ץ֖רֶ אִּתְ כֶ ם . . . לְ לכֹ֖ חַּיַ ֥ת הָאָ ֽרֶ ץ 11 הֲ ִ ק ֹמ תִ ֤ י אֶת־ּבְרִיתִ ֖ י אִ ּתְ כֶ ם 12 אֲנִ ֣י נֹתֵ֗ ן אֹֽות־הַּבְרִ ית ּבֵינִי֙ ּובֵ ֣ינֵיכֶ֔ םּובֵ ֛יןּכָל־נֶ ֥פֶ ׁש חַּיָ ֖ה ֣ ראֲׁשֶ אִּתְ כֶ֔ ם 13 נָתַ יּ֖תִ ּבֶ ֽעָ נָ ן אֶת־קַׁשְּתִ֕ י קַׁשְּתִ֕ י הָ ֽ יְתָה֙ לְא֣ ֹות ּבְרִ ֔ ית ֖יּבֵינִ ּובֵ ֥ ין הָאָ ֽרֶ ץ 14 ּבְ עַ ֽ ְ נ נִ ֥ י עָנָ �֖ן עַל־הָאָ ֑רֶ ץ נִרְ ֥ אֲתָ ההַּקֶ תׁ֖שֶ ּבֶ ֽעָ נָ ן 15 וְזָכַרְּתִ֣י אֶת־ּבְרִיתִ ֖ י אֲׁשֶ ר ֣ ּבֵ ינִי֙ ּובֵ ֣ינֵיכֶ֔ ם ּובֵ֛ין ּכָל־נֶ ֥פֶ ׁש חַ ּיָ ֖הּבְכָל־ּבָׂשָ ֑ ר 16 וְהָ ֥ יְתָ ההַּקֶ תׁ֖שֶ ּבֶ ֽעָנָ �֑ן ּורְ אִיתִ֗ י הָ לִ זְ ּכ ֹ ר֙ ֔ ּבְרִ ית עֹולָ֔ ם ּבֵ֣ין אֱֹלהִ֔ ים ּובֵין֙ ּכָל־נֶ ֣פֶ ׁש חַ ּיָ֔ ה ֖ רּבְכָל־ּבָׂשָ אֲׁשֶ ֥ ר עַל־הָאָ ֽרֶ ץ 17 הֲ קִ ֹ֔ מ תִ י אֹֽות־הַּבְרִית֙ ּבֵינִ֕ יּובֵ ֥ ין ּכָל־ּבָׂשָ ֖ ר אֲׁשֶ ֥ ר עַל־הָאָ ֽרֶ ץ

Figure 2 160 ellen van wolde

I set“ ,נתתי ב- I establish my covenant” (verse 11a) and“ ,הקמתי ,execution my bow in the clouds” (verse 13a). The participle expresses an a-temporal relation and thus creates the perception of duration in which the end point is not included; it characterizes the deity relative to his covenant and to the sign of the covenant. Conversely, the verbal clauses with the finite qatal forms narrate the actual start of the establishing of the covenant and of the setting of his very own fighter’s bow as a sign of this covenant; they express temporal processes in which the beginning and the share the זכר and ראה endpoint of the action are included. (4) The verbs similar concepts of seeing directly with the eyes or indirectly in the mind, both in the first person. All four verbs in (3) and (4) represent the spatial, conceptual, and perceptual perspective of God. (5) The repetitive use of the preposition ‘between’ is striking: it is used with regard to the covenant ,in verses 13, 16 בין ובין of one partner with another partner (expressed by in verses 12 בין ובין ובין and 17) or with two other partners (expressed by amounts to twelve. In בין and 15), so that the total sum of the preposition case of a partnership between two partners, these are “Me”/”God” and “(all life on) earth”; in case of three partners, it is between “Me” and “you” and .designates the intermediate space בין all life with you.” The preposition“ The covenant and the bow as the sign of the covenant are localized in space between “Me”/”God” and “(all life on) earth” or between “Me” and כל אשר and ענן על הארץ you” and “all life with you.” (6) The phrases“ -share the concept of ‘the earth’ as well as ‘extensiveness’ and a cer הארץ is used to designate ענן tain vagueness, which also explains why the term the extensiveness of the cloud covering the earth. Note also the seven- all” and “entire” in verses 12 (once), 15 (three times), 16“ ,כל fold usage of everlasting” in verses“ ,עולם twice), and 17 (once), and the double usage of) 12 and 16. This meaning structure explains why verse 14a explicitly states that God is bringing the clouds “over the earth.” It is the individuality of the bow and the individuality of the covenant as well as the superiority of the powerful party that stands over against the extensive earth as a whole. When the deity sees his powerful weapon set in the clouds, he will remember his covenant with all those covered by the clouds.

The Composite Meaning Structure in Gen 9:8–17

All these correspondences together build up into a composite meaning complex which is made possible by the sharing of base components. Figure 3 summarizes the meaning components and the bold print refers to the shared base components. A short explanation will follow. linguistics & literary studies of ancient biblical & jewish texts 161 vv. 9–11 [setting up] קום [proximity] את [my bond] בריתי ‘I’ — [sharing of power] — ‘with you and all life on earth’ v. 12 [setting] נתן [intermediate space] בין [mediation] [bond] אות הברית ‘I’ — [mediation] — ‘between me, you, all life with you’ v. 13 [setting] נתן [container] ב- [my warrior’s bow] קשתי ‘my bow’ — [spatial transfer of bow] — ‘in the clouds’ [being] היה [my warrior’s bow] קשתי [intermediate space] בין [mediation][bond] אות ברית ‘my bow’ — [token/mediation] — ‘between me and the earth’ v. 14 [clouding] ענן [earth]ארץ [clouds] ענן ‘I’ — [cover/expanse] — ‘over the earth’ [appear] ראה [container] ב- [warrior’s bow] קשת ‘the bow’ — [spatially transferred] — ‘in the clouds’ v. 15 [remember] זכר [intermediate space] בין [my bond] בריתי ‘I’ — [power shared] — ‘between me, you, and all life’ v. 16 [being] היה [container] ב- [warrior’s bow] קשת ‘the bow’ — [spatially transferred] — ‘in the clouds’ [see] ראה [remember] זכר [intermediate space] בין [enduring bond] ברית ‘I’ — [power transferred] — ‘between God and all life on earth’ v. 17 [setting up] קום [intermediate space] בין [sign of bond] אות הברית ‘I’ — [power transferred] — ‘between me and all life on earth’ Figure 3 162 ellen van wolde

בריתי ,הקמתי / אני מקים ,Verses 9, 10 and 11 contain three main elements covenant,” (used twice) expresses a bond“ ,ברית The word .אתכם and between a superior party and an inferior party, a partnership between a lord and his vassals. It designates [hierarchy] and [supremacy] as well as [bond] and [sharing of power], while it functions in the cognitive expresses the spatial קום domains of [dominion] and [peace]. The verb concept of [setting up] or [establishing] and indicates the start of [with,” (five times) describes [proximity“ ,את this bond. The preposition and [togetherness]. Thus it becomes clear that these three elements share the base components of [sharing] and [togetherness] in relation to power. In verses 12 (“I am giving the sign of the covenant”) and 13a (“I set my -is used, but the grammatical con נתן bow in the clouds”) twice the verb -is not collocated with the prepo נתן struction differs. In verse 12, the verb the sign“ אות הברית ,but merely combined with a direct object ,ב- sition of the covenant,” so that it indicates the regular meaning of ‘giving’ and expresses the concept of an [offer]. The addressee or recipient of this present is not mentioned (the text does not say “I give the sign to you”) but is attached to the object itself (“I give the sign of the covenant between :[is [mediation אות me and you”). The conceptual content of the word expresses that a characteristic feature אות הברית the word combination of the covenant is present, which makes a human or divine being realize contains a similar בין that the bond is actually happening. The preposition notion, since it designates [intermediate space]. Whereas verses 9–11 -and thus expresses the togeth את with the preposition הברית collocate in verse 12 includes בין erness of the covenant partners, the preposition the notion of distance between the covenant partners who are positioned ,נתן אני הברית vis-à-vis the space that separates them. The three elements .[share, therefore, the base component of [mediation בין and ,אות ”,in“ ,ב- is collocated with the preposition נתן In verse 13a, the verb ענן describes a spatial movement. The word נתן בענן את קשתי so that designates [clouds] and [cloud cover] and includes the notion of [intermediate space] between heaven and earth. In verse 14, these clouds are defined vis-à-vis the earth. So, what started as a position in- between-heaven-and-earth is subsequently defined as a position related to the earth. In addition, the notion of [extensiveness] included in the God .ארץ cloud cover] corresponds with that of [expanse] in] ענן word moves his bow along a path, while taking his hand or at least his place in the (upper) heavens as its starting point, and the clouds are localized as the final position to which he orientates his movement that culminates linguistics & literary studies of ancient biblical & jewish texts 163 inside the clouds. Therefore, these words express the [spatial transfer] of God’s [bow]. In verse 13b the bow set at a distance becomes “the sign of the covenant prototypically designates a קשת between me and the earth.” The word [warrior’s bow] used in the cognitive domains of [war] and [dominion] in relation to the victorious winner of that war, and entails meaning components such as [victorious weapon of attack], [enemy], [victory] as well as [power], [might] and [supremacy]. The covenant designates [hierarchy] and [supremacy] as well as [bond] and [sharing of בין designates [mediation] and the preposition ,אות power]. The word [intermediate space]. It thus becomes clear that all these words share the base components of [spatial transfer] and [power]. From verse 14 onwards, the time frame of the narration changes. The events in verses 14–17 are told from a point in the future in which the seeing of the sign of the covenant makes God realize what actually happened in the past. Verse 14a first depicts the circumstances (“whenever I bring the base component ארץ shares with ענן clouds over the earth”) in which of [expanse]. The appearance of the bow in verse 14b shows that God is already separated from his weapon and that it has already been spatially transferred to the clouds which are defined vis-à-vis the earth. Therefore, the words in this verse share the base component of [spatially transferred], a transport in place that is situated in the past. -remem“ ,זכר In verse 15 we get access to God future thoughts. The word ber,” designates [mediation] between present and past which takes place in someone’s mind. The bow in the clouds God is seeing makes the past present, so that he realizes that he did share his power with the partners expresses [hierarchy] and [supremacy] as well ברית on earth. The word -designates [inter בין as [bond] and [sharing of power], and the word mediate space], so that we can conclude that the base components amount to God’s realization of his sharing of בין and ,ברית ,זכר shared by power in the past, hence [power shared]. The same future standpoint is shared in verse 16, but now the lens is zooming out, presenting a kind of overall view, in which God for the first time does not speak about ‘I’ and ‘you,’ but the text describes the covenant partners as ‘God’ and ‘the earth.’ Verse 16 opens with the presence of the bow in the clouds and God’s perception of it. The bow’s place in the clouds entails that it has been set at a distance from God, and since the clouds have previously been defined in relation to the earth, this position in the clouds relates the bow to the earth too. God’s warrior-bow represents his enduring power, so that the event, in which the bow has 164 ellen van wolde been transferred from him to the clouds, indicates that his power (over the earth) is transferred to the earth’s inhabitants. occurs again. Whereas in verses 9–11 קום Finally, in verse 17 the verb is collocated with the covenant in order to express את the preposition in verse בין the togetherness of the covenant partners, the preposition 17 includes the notion of distance between the covenant partners while they are positioned vis-à-vis the space that separates them. God’s remem- brance of the sign of the covenant brings to mind both the distance and the bridging of this distance by the sign, thus representing the actual [transfer of power] that took place in the past. Above, I asked what the conceptual connection would be between God’s rainbow and his covenant, and now I ask the same question about his warrior’s bow. The correct answer is that the distinctive features that link God’s fighting bow and his covenant are the conceptual contents of [power] and [sharing of / transfer of power].

The Context of Genesis One can raise the question how God’s transfer of power represented by the bow in the clouds may fit the textual context of Genesis 9 and other texts in Genesis it is referring to. Due to space reasons, my answer must be short. In Gen 9:1–7, prior to mention of the covenant, three topics are presented: (1) at the beginning and the end of the episode God blesses Noah and his sons, which he describes as being fertile and filling the earth; (2) closely linked to this blessing and spreading over the earth are the of the other living beings on earth toward (חת) and dread (מורא) fear humans; (3) these living beings are said to be “delivered into your hands” Clearly, God gives the human race the power to monitor 55.(נתן בידכם) the earth, although their dominion is restricted by some rules: they are not allowed to eat flesh with its life-blood in it or to shed the blood of a human being. Regardless these restrictions, from now on all animal life will fall under human control, and all green plant life will too. The reason for this is given in verse 6: “For in his image did God make human beings.” This ‘image of God’, as well as the blessing and multiplication on earth, and the dominion of the earth in Gen 9:1–7 refer back to 1:26–28, where .רדה ב- the rule over the earth is expressed by the verb

.(נתתי לכם את כל) ”See also Gen 9:3 “I give you everything 55 linguistics & literary studies of ancient biblical & jewish texts 165

humankind will both rule and dominate ,(רדה ב-) In the idiom of verse 26b with an enormous power. [. . .] as ‘image’, the human race will embody and -itself is an evoca רדה .assert the power of its referent over the natural world tive verb. One nuance bears upon mastery, especially as an expression of implies a relationship between victor and רדה [. . .] .victory or punishment vanquished. Another nuance bears upon the identity of the victorious party. can express the power that a king wields רדה ;That part is often royal [. . .] over his subjects [. . .]. In this sense, it is a royal prerogative [. . .]. For the in Gen 1:26, then, these two nuances suggest that humankind רדה reading of is empowered to hold dominion over the world and rule its inhabitants as a king. Through its ‘image’, the human race will master the world as a majes- tic, executive, and triumphant power.56 The topic our text shares with Gen 1:26–28 and 9:1–7, is human dominion over the earth. However, what was expressed symbolically as ‘the image of God’ and ‘dominion’ in Gen 1:26–28 and 9:1–7 is expressed metonymically in 9:8–17.57 Metonyms can be either real or fictional concepts representing other concepts either real or fictional, but they must serve as an effective and widely understood second name for what they represent, which is unquestionably true for the private weapon of God. The bow, the entity that is intimately associated with God, represents God’s might and enduring power, and his transport of the bow in the clouds over the earth represents his transfer of power over the earth to the human beings on earth. This bow is not a weapon of protection, but a weapon of attack of a victorious fighter and God’s abdication from his weapon can serve as an effective metonymic concept of his handing over power.

Conclusion: The Bow in Genesis 9

in relation to the deity and to קשת Because Gen 9:8–17 uses the word the clouds, the inference has been made in biblical scholarship that the text refers to a rainbow. The plausibility of this inference is tested in this article. Linguistic, textual, iconographic, intertextual, and comparative literary קשת data and arguments are given in favor of the view that the word

56 Cf. W. Randall Garr, In His Own Image and Likeness: Humanity, Divinity, and Mono theism (CHAN 15; Leiden: Brill, 2003), 155–157. 57 Cf. Taylor’s definition of metonymy: “The process by which an expression which ‘basically’ designates entity e, comes to be used of an entity closely associated with e, within a given domain, as in I (=my name and phone number) am in the telephone book.” (Taylor, Cognitive Grammar, 590). 166 ellen van wolde designates a [warrior’s bow] and figures in the cognitive domains of [war] and [dominion], while including conceptual components such as [weapon of attack], [arrow], [enemy], [victorious weapon of a winning warrior] and [power]. The analysis of the composite semantic structure of Gen 9:8–17 showed is closely linked to its owner as the victorious deity and his קשת that God’s weapon represents his enduring power. God’s establishing of his covenant -in combination with the preposi ברית with those living on earth and his -express [bond], [sharing] and [togetherness]. The position את tion ing of his warrior’s bow in the clouds that are covering the earth and its describe בין collocated with the preposition אות הברית function as the the [distance] between God and the earth as well as the [transfer of power]. In other words, the establishing of the covenant stands for God’s willingness to share his rule over the earth with those living on it, while his setting of the bow in the clouds stands for God’s actual transfer of his power over the earth to those living on earth. Whether or not God’s setting a bow at a distance implies the notion of [bow star] is impossible to say. It is, however, very well possible that God placing of his weapon-bow in the sky is comparable to similar behavior of deities in ancient Mesopo- tamian literature. Unfortunately, no further indications are found in the text of Genesis 9. It is the individuality of the bow and the individuality of the covenant as well as the superiority of the powerful party that stands over against the extensive earth as a whole that characterizes this text. The powerful deity transports his weapon of attack into the clouds over the earth as a sign of his covenant with the human beings, as a sign of his abdication of his weapon of attack and a transfer of power. With the handing over of his mighty weapon, he demonstrates that he will never again attack and destroy all living beings on earth. He will never again use a flood to exercise power. From now on, the descendants of Noah, the human race and the living beings with them, are made responsible for the dominion over the earth. The ‘Literary’ in Trickster Narratives: A Comparison of Gen 27:1–39; 29:15–30; 31:19–55; Joshua 9 and 2 Sam 3:26–27

Anne-Laure Zwilling

Abstract

Biblical texts offer quite a number of stories of trickery, humbug or deception. One can find there a broad panel, going from the simple trick to the elaborate stratagem and from the instant deception to the long-planned ruse. The way the information is given also differs: the deceptions can be recalled briefly in a few sentences, or they can be narrated at length with numerous details provided inside a sophisticated story frame. Through the study of different examples, I will try to discover which relation can be perceived between the complexity of the trick and the mode of its rendering. This will help to establish that, as regards trickster narratives, one can indeed speak of ‘literary.’

Two stories of deceit can be found in the Jacob cycle, in the book of Gen- esis. In Gen 27:1–39, Jacob takes Esau’s place to receive the blessing from Isaac. In Gen 29:15–30, Laban gives Jacob Leah instead of Rachel for the marriage. Both texts belong to the cycle of Jacob, one as well as the other narrate a substitution of persons, and twice the youngest takes the place of the eldest with Jacob being one of the main characters. There are indeed many similarities in these two stories, as both episodes relate a deceit on identity, with Jacob being involved both times. Jan Fokkelman underlines that in each story exchanges take place “in the darkness of night and behind a veil”1 while Yair Zakovitch points out that they both take advantage of blindness, either actual or due to the absence of light.2 How- ever, despite the similarities, others, such as Zvi Jagendorf, underline that

1 Johannes Fokkelman, Narrative Art in Genesis: Specimens of Stylistic and Structural Analysis (SSN 17; Assen: Van Gorcum, 1975), 129; Zvi Jagendorf, “ ‘In the Morning, Behold, It Was Leah:’ Genesis and the Reversal of Sexual Knowledge,” Proof 4 (1984): 187–192. 2 Jacques Vermeylen, “Le vol de la bénédiction paternelle: Une lecture de Gen 27,” in Pentateuchal and Deuteronomistic Studies: Papers read at the XIIIth IOSOT Congress Leuven 1989, ed. Christian Brekelmans and Johan Lust (BEThL 94; Leuven: University Press, 1990), 23–40; Yair Zakovitch, “Through the Looking Glass: Reflections / Inversions of Genesis Stories in the Bible,” BibInt 1 (1993): 139–152. 168 anne-laure zwilling

Genesis 27 and Genesis 29 have a very different rendering and that their discourses differ greatly.3

Explaining the Differences in Discourse

Taking into account the way the trick is narrated in these texts, I have argued previously that there is a reason for such a difference, which is the position of Jacob.4 This theory was based on a narrative-critical approach, which seems the most suitable manner to draw attention to the difference in the number of characters and the relationship between them. I will recall here briefly the major elements of this analysis. The first is the number of characters: Genesis 27 (hereafter named the blessing episode) displays four strong characters: Isaac, Rebecca, Jacob, and Esau. The episode can be divided in seven scenes, each one involving two members of the family who all interact with one another, each one of them having a place in the plot and a definite role: Rebecca plots, Jacob helps. They both partake in the design and achievement of the plot. Isaac is lured into believing that Jacob is Esau, and Esau has to go without his blessing; thus, Isaac is deceived and Esau is wronged. In Genesis 29 (the wedding episode), the different roles are held by characters who each assume several roles: Laban alone organizes and achieves the plot, and Jacob claims to have been both cheated and wronged.5 Rachel and Leah are silent and one knows nothing of their relationship to Laban—besides being his daughters. The reader is not told why Laban substitutes Leah to Rachel or how he managed to bring Jacob to believe that it was indeed Rachel that he was marrying. No one knows the part played by either daughter in the plot. There are thus two strong characters (Jacob and Laban) and two mute characters (Rachel and Leah) in the wedding episode. Not only do the two episodes differ in the number and style of the characters, but they also vary in the management of time. In the blessing episode, the trick is

3 The difference between story and discourse used here is based on the definitions given by Seymour Chatman, Story and Discourse: Narrative Structure in Fiction and Film (Ithaca, NY: Cornell University Press, 1980). Jagendorf, “ ‘In the Morning, Behold, It Was Leah,’ ” 187–192. 4 Anne-Laure Zwilling, “Le narrateur ‘intrigue’ son lecteur: dire ou taire les modalités de la ruse, comparaison des récits de Gen 27 et 29,” in L’intrigue dans le récit biblique: qua trième colloque international du RRENAB, Université Laval, Québec, 29 mai–1er juin 2008, ed. Anne Pasquier, Daniel Marguerat, and André Wénin (Louvain: Peeters, 2010), 171–180. 5 Zakovitch claims that this exchange is a punishment of Rebecca more than of Jacob. Zakovitch, “Through the Looking Glass,” 139–152. the ‘literary’ in trickster narratives 169 intertwined with the initial contract between Isaac and Esau, which is to provide food in order to receive the blessing. One could say that the actual events are inserted in what should have been, the discourse becoming a reflection of the trick: The simple story The trick inserted 3–4: Isaac asks Esau for a meal 5b: Esau sets out to answer Isaac’s demand 7–10: Rebecca explains her plot to Jacob 11–13: disagreement and answer 14–17: preparation of the trick 18–29: implementation of the trick 30–31: Esau has finished doing what Isaac requested 32: disclosure of the trick Many repetitions can be found in the text: Rebecca repeats to Jacob what Isaac has asked for (verse 7), after that she tells him what should be done. Then Jacob does it and next tells Isaac what he has done. Finally, Esau again repeats what has been done (verse 30). All this renders the story quite long, close to the rhythm of the action it recalls. In Genesis 29 on the contrary, the exchange is nearly immediate. It is expressed in the text by a brief and simple change of name, Laban bringing Leah and not Rachel: “and in the morning, behold, it was Leah” (Gen 29:25). The wedding episode is very short and has a quick pace; the exchange of brides is narrated as to give the impression of a magical trick performed. There is no suspense there, no expectation for the reader. On the contrary, in the blessing episode, Isaac expresses his doubts several times, creating a sharp contrast between the energy displayed to try and really know who he is dealing with, and his inability to recognize his own son.6 This incompetence is all the more striking for all the information is given to him, mostly through the senses:7 the text mentions touching, eating, hearing, and smelling.8 Isaac is blind, but his ‘night’ is full of information. Furthermore, all this is fully knowledgeable to the reader. He receives information provided by the narrator on the blindness of Isaac, on the fact that Rebecca is listening, and on the whereabouts of Esau.9

6 Edwin M. Good, Irony in the Old Testament (BLS; Sheffield: Almond Press, 1981), 99, underlines the irony of this suspicion of Isaac in Gen 27:18, 20, 21, 22, 24, and 26. 7 Jagendorf, “ ‘In the Morning, Behold, It Was Leah,’ ” 190. 8 Luis Alonso Schökel, Dov’è tuo fratello? Pagine di fraternità nel libro della Genesi (Biblioteca di cultura religiosa; Brescia: Paideia Editrice, 1987). 9 Narrator is here understood in the meaning given by narrative criticism and does not intend to debate on the actual author of the text. 170 anne-laure zwilling

The means of the trick are explained, Jacob’s concerns are expressed, and Isaac’s doubts are exposed at length. In the wedding episode, by contrast, there is nothing in the ‘night’ of Jacob. This story is designed to surprise the reader who can only, if he is careful, be aware that the name Leah is used instead of Rachel in verse 21.10 No other warning is yet given and no reaction is expressed by Jacob (“he goes to her,” verse 23). Verse 25, in its simplicity, comes as totally unexpected. This certainly explains the tre- mendous amount of subsequent writing in which commentators display a noteworthy imagination to explain how Jacob could have mistaken Leah for Rachel.11 The obvious reason for such a contrast in writing seems the part played by Jacob in the trick. Should the reader have a full understanding of the plot, instead of being left with the question: ‘how did that happen?’ he would then be in a position superior to that of Jacob. This is not the intent. Jacob is the hero of the cycle and the narrator wants to draw the reader to his side. In short, when Jacob is deceiving someone, the reader learns everything on how it is done; but when Jacob is cheated, the reader is fooled just like Jacob.

A Literary Device . . . Explained?

The explanation of this literary device fits perfectly for the wedding and the blessing episodes of Genesis 27 and 29. Both being located in the cycle of Jacob, they are two excellent illustrations of a trick being rendered in one case with many explanations and details, in the other nearly without any information. They illustrate two opposite renderings of deceits and fit perfectly well in the perspective of building the character of a hero in the broader story. One can wonder, though, whether this use of the trick as a means to build a character can be found elsewhere in the Bible and whether that would consequently result in a literary device typical of trickster stories. Is it possible to construct a general theory of the depiction of deception in the Hebrew Bible?

10 Jacob asks Laban in verse 21 to give him ‘his wife.’ This is followed in verse 23 where Laban is said to take his daughter Leah and bring her to Jacob. 11 In addition to the major commentaries of this episode, see also for instance Cath erine Chalier, Les matriarches: Sarah, Rébecca, Rachel et Léa (Paris: Cerf, 1991); James A. Diamond, “The Deception of Jacob: A New Perspective on an Ancient Solution of the Prob lem,” VT 34 (1984): 211–213; Carey Ellen Walsh, “Under the Influence: Trust and Risk in Biblical Family Drinking,” JSOT 90 (2000): 13–29. the ‘literary’ in trickster narratives 171

Establishing a definite theory would necessitate a study of all the tricks and deceits mentioned in the Bible.12 There are, however, a very great number of them. Even if one excludes simple lies, such as pretending that one’s wife is one’s sister for instance, or basic tricks, such as Abimelech asking his young sword carrier to stab him so that he wouldn’t have been killed by a woman (Judg 9:54), the number of biblical stories of tricks and deceits remains high. And even if one also leaves aside more elaborate techniques, such as the one Jacob uses to augment the amount of striped cattle, there remains quite a lot.13 A great many stories are episodes of wars and battles in which some claim to retreat in order to attack (Judg 8) or pretend to come in peace before killing each other, such as Jehu in 2 Kings 9. In the book of Genesis, everybody knows about the snake deceiving the woman in chapter 3 or the two stories just studied. One can also find in chapter 31 Rachel fooling Laban who is searching his teraphim. In Genesis 34, the sons of Jacob get the Hivites to be circumcised and attack them while they are still in pain, whereas in Genesis 37, Jacob’s sons make him think Joseph is dead and later, Joseph pretends not to speak Hebrew while overhearing every word they say (Gen 42:23). That is far from being all: in the book of Joshua for example, one reads how the men of Gibeon fooled Joshua into believing they lived far away (Josh 9:1–27); in the book of Judges, Ehud fooled the king of Moab into getting rid of his guards so that he could kill him (Judg 3:15–30). Jael, wife of Heber, fools and kills Sisera (Judges 4), Gideon led the Madianites to believe that his small force was much larger (Judg 7:15–23), Delilah seduced Samson into telling her the truth about his weakness (Judg 16:4–30), and the Benjaminites hide to abduct the daughters of Shiloh (Judges 21). In the first book of Samuel, we hear about Michal helping her husband David to escape from Saul, letting him out through a window, then tricking Saul’s men into thinking that a teraphim in her bed is actually David (1 Samuel 19). In the second

12 On this topic, see Shimon Bar-Efrat, Narrative Art in the Bible, trans. Dorothea Shefer- Vanson (JSOT.SS 70; Sheffield: Almond Press, 1989); Kathleen Anne Farmer, The Trickster Genre in the Old Testament (Southern Methodist University Dissertation, 1978); Good, Irony in the Old Testament; Susan Niditch, Underdogs and Tricksters (San Francisco/New York: Harper and Row, 1987); Naomi Steinberg, “Israelite Tricksters, their Analogues and Cross- Cultural Study,” Semeia 42 (1988): 1–13. See also Kathleen M. Ashley, “Interrogating Bibli cal Deception and Trickster Theories: Narratives of Patriarchy or Possibility?” Semeia 42 (1988): 103–116; Mieke Bal, “Tricky Thematics,” Semeia 42 (1988): 133–155; André Wénin, “Le jeu de l’ironie dramatique dans les récits de ruses et de tromperies,” in L’intrigue dans le récit biblique, 159–170. 13 Scott B. Noegel, “Sex, Sticks, and the Trickster in Gen. 30:31–43,” JANES 25 (1997): 7–17. 172 anne-laure zwilling book of Samuel, Amnon fools Tamar into coming into his bedroom (2 Sam 13:1–14). In the book of Kings, one can find the story of Solomon deceiving two women to get the truth about the identity of a child (1 Kgs 3:16–28), and Jezebel leading people into believing wrongly that Naboth had cursed God and King Ahab (1 Kgs 21:1–16). This list is certainly not exhaustive and one has to admit that it is impossible to deal with all these texts in one article. Therefore, and in the perspective of opening the discussion, I will narrow the search to five episodes in this article which aims only at being a trial, a tentative comparison, a work in progress. I have chosen, in addition to the blessing and the wedding episode of Genesis mentioned above, three texts of very different styles: a long story from the book of Joshua (Joshua 9); a short micro-episode from the story of Joab in 2 Sam 3:26–27; and a third text from Genesis, the episode of Rachel and the teraphim (Gen 31:19–55).

Tentative Study of Texts

The Trick of the Gibeonites ( Joshua 9) One of the chapters of the Bible recalls the story of Joshua and the men of Gibeon (Joshua 9). Having heard “how Joshua had taken Ai, and had utterly destroyed it, doing to Ai and its king as he had done to Jericho and its king,” these men wanted to make a treaty with Joshua.14 This being possible only with people who do not belong to the chosen land, they pretend to come from far in order not to be considered as such. They utilize disguise (worn clothes and shoes) and artifacts (moldy bread and burst wineskins) to support their claim of having traveled a long journey. The narrator provides a judgment on their acts: at the beginning of this with“ ,בערמה episode, the narrator warns the reader that these men act cunning.” He also mentions the error of the men of Joshua: they did not “ask directions from the Lord (verse 14).” When the ruse is eventually dis- why did you lie to us“ ,למה רמתם אתנו covered, Joshua asks the men (verse 22)?” Throughout the story, the trick is explained and displayed as such. It comes as no surprise for the reader, who knows from the beginning that this is a trick, explained first (verses 4–5) before being carried out. The knowledge of the reader is superior to that of Joshua and his men and the little suspense found in the text lays in the success of the trick.

14 Josh 10:1. the ‘literary’ in trickster narratives 173

When the men of Joshua seem suspicious and wonder if these unknown people could be fellow kinsmen, which would forbid any covenant, the reader already knows the answer to that. The Gibeonites nevertheless reinforce their artifice in language. Upon meeting the men of Israel, they lie, saying twice that they have “come from a far-away country (verses 6, 9),” and emphasize their disguise through language: “See, our clothes and sandals are worn-out (verse 13).” They expose their pretense story, from the initial idea of making a covenant to the sham preparation of their journey, in a long speech (verses 9–13), recalling what they have done (verses 5–6); thus, their cunning is both shown and told. The reader, here, is really led to understand how the congregation was fooled. The trick is more or less described as “fail proof.” Plenty of information on the means and organization of the trick is provided, as in the first story examined above of Jacob stealing the blessing. This was, as we have established, because the narrator had organized the episode for the reader to be on the same side as the hero. Jacob being the trickster, the reader knows a lot about the achievement of the stratagem. Here, however, the trickster is not the hero of the story, which clearly is a role for Joshua. One does not find in this story a clear condemnation of the act of the Gibeonites. Although they are not killed, their fate is not depicted very positively: they are to remain servants or slaves. This is repeated three times (verses 21, 23, 27). Their trick is not judged positively: in the eyes of the congregation, they should have been killed, and there even are complaints against the leaders because they are not put to death (verse 18). This story of deceit is comparable to the blessing episode as regards the amount of information provided for the realization of the trick. But the trickster is clearly not the hero of the story. Like in the wedding episode, the victim of the trick is the hero, here Joshua.15 The two first stories were opposed: in the blessing episode, in which a lot of information is given, the hero is the trickster and there is no clear disapproval of the deceit. On the contrary, in the wedding episode, no information on the trick is to be found; the hero is the victim. The Gibeonites’ episode takes part of both: the hero is the victim as in the latter, but all information on the means of the trick is given as in the former.

15 I use here the word ‘victim’ without moral connotation, in the sense of the person the trick is played upon. 174 anne-laure zwilling

The Murder of Abner (2 Sam 3:26–27) The two verses of 2 Sam 3:26–27, recalling the murder of Abner (here after the murder episode), constitute a very short episode, part of a longer story in which four characters are involved: Joab, the messengers, David and Abner. Of these, the messengers and David have very little importance. The messengers are only mentioned. David is said to be absent: things take place “out from his presence.” Although he is the hero of the broader story, in this short episode the hero is Joab, David “did not know about it” (2 Sam 3:26). In contrast, Joab is strongly present with many active verbs: he comes out from David’s presence, gives orders, pulls Abner aside and kills him. Abner, on his side, is weakly present. He is dealt with like an object: the messengers come after him, bring him back, and he is stabbed. The only active action he takes, if one can say so, is to die. All information is given in the text by showing: the reader can see Joab sending messengers, taking Abner, striking him; then Abner is shown dying. No other information is given. For instance, no mention is made of a weapon carried or used by Joab. Neither does the narrator say what is going on inside Joab’s mind, his intention. The narrative is short, quick, and efficient, just as the killing is. The attention of the reader is focused on Joab and Abner only, with the text gradually leaving out first David then the messengers. The focus goes from the public space (the middle of the gateway) to the private, reaching even into the stomach of Abner. The few words describing the killing are the apex of the episode, exhibiting at the same time the true moment of action and the closest intimacy between Abner and Joab. In fact, all characters gradually disappear from the story, leaving only Abner and Joab. This construction displays how personal the matter is. The reason is revealed and explained in the following verse: the killing is a revenge for the death of Asahel. However, this murder has come unexpected. The reader is, like Abner, taken by surprise, his awareness is no greater than the one of the other characters, Abner and David. The little information given is provided to speak“ ,לדבר אתו בשלי by the narrator who offers information such as with him in secret,” although this does not actually reveal the intent of the killing but only indirectly highlights one of its means.16 In the verses just preceding, Joab has met David and spoken to him about Abner, saying that this man came only to deceive David (verses 24–25). Maybe this could be understood as a warning for the reader, but this seems unlikely. Indeed, if deceit is expected, it is on the part of Abner. In the end, the reader is left

16 2 Sam 3:27. the ‘literary’ in trickster narratives 175 out of Joab’s project, just as David is.17 The reader will finally find himself in the same situation as the one David ends in: as it happens often in life, everything seems clear afterwards. When the killing is done, one thinks that it could have been foreseen. Joab, one of the heroes of the chapter, is the trickster here. Very little is said about the way he manages to dupe Abner. In this way, this episode works in contrast with the Gibeonites episode. The first two episodes could be opposed: in the blessing episode, the hero is the trickster and all information on the trick is given; in the wedding episode, the hero is the victim and no information is provided. The Gibeonites episode, as said above, provides a lengthy description of the trick like the blessing episode, but the hero is the victim like in the wedding episode. This fourth episode, the murder of Abner displays the reverse method: the hero is the trickster, but no information is given on the means of the trick.

Rachel and the Teraphim (Gen 31:19–55) In the last text taken into account in this article, Rachel is hiding the teraphim she took from her father Laban. While he searches for the idols that she has hidden in her saddle, she explains that she cannot stand up .(I have the way of women” (Gen 31:35“ ,דרך נשים לי ,because, says she Laban, as such prevented from searching the entire tent, is unable to find the teraphim and loses face before Jacob. The stratagem is intricate. Rachel obviously tricks Laban by lying to him, but she also deceits Jacob, who was unaware of the theft and unknowingly will take a high risk, going so far as to promise that “anyone with whom you find your gods shall not live” (Gen 31:32). The reader, however, is aware of all this. His knowledge is superior to the one of Laban and of Jacob; he knows the moves of Rachel even though he ignores her motives. The reader is also on the side of Rachel as the search carried out by Laban comes closer and closer to her (into one tent, then a second, then a third, then finally into Rachel’s tent, verse 33). In the course of the story, the nearer Laban gets, the more information the reader receives on the hiding place of the teraphim. He first learns that Rachel has taken and hidden them (verse 19), then that she is actually sitting on them—they

17 The opinion on the responsibility of the crime is divided. Some, such as James C. Vanderkam (“Davidic Complicity in the Deaths of Abner and Eshbaal: A Historical and Redactional Study,” JBL 99,4 [1980]: 521–539) believe that David is guilty. For others, such as Jean-Claude Haelewyck (“La mort d’Abner: 2 Sa 3,1–39: Étude littéraire et historique,” RB 102,2 [1995]: 161–192), Joab is the sole responsible. 176 anne-laure zwilling couldn’t be closer to her (verse 34). Therefore, the more Laban approaches her, the closer he comes to the teraphim. The reader learns about the trick and follows the increasing threat both through showing and telling. Being aware that Rachel indeed has Laban’s idols in her possession, the reader is put in a place similar to Rachel’s situation: both know where the tera phim are and wonder if Laban will find them or not. Many details are given, for instance when Laban is described going from tent to tent, then “feeling all about the tent” (verse 34), an action that will be described later by Jacob saying that Laban “has felt through all my goods” (verse 37). A kind of comic situation is built here, the thriving knowledge of the reader that Laban is being fooled opposed to the growing uselessness of Laban’s search. This culminates in Rachel addressing Laban very politely, calling him “my Lord” and presenting excuses for not saluting him most respect- fully, while at the same time making a fool of him (Gen 31:35). The knowledge of the reader is great and the suspense is high also. In this episode, the reader is aware of the trick from the start; he is drawn to the side of Rachel as the suspense in the story is increasing, when Laban is approaching the teraphim he is looking for. In such a detailed rendering of the scene, the scant information on Rachel’s move draws attention to the simplicity of the means. The scene here intends to depict an absence of preparation, the trick resembling more an instinctive cunning move from Rachel than a carefully prepared one. As many commentators have noticed, the narrator seems to imply that Rachel has done nothing more than using “the way of women.”18 This, indeed, unveils the core of the trick, which rests on double entendre. The “way of women” that Rachel calls forth can refer to the impurity of the woman, which would prevent Laban from touching her or even coming close to her. It could also be the cunning traditionally attributed to women.19 This double entendre is a game played—not on the reader, however, but on Laban. The reader is hence again brought to side along with Rachel, both the trickster and the hero of the episode. The

18 On the ‘way of women,’ see Danna Nolan Fewell and David M. Gunn, “ ‘The Way of Women,’ ” in Gender, Power, Promise: The Subject of the Bible’s First Story? (Nashville: Abingdon Press, 1993). On the more general question of women and trickster theories, see Ashley, “Interrogating Biblical Deception and Trickster Theories;” Nelly Furman, “His Story Versus Her Story: Male Genealogy and Female Strategy in Jacob Cycle,” Semeia 46 (1989): 141–150; Jo Cheryl Exum and Johanna W. H. Bos, eds., Reasoning with the Foxes: Female Wit in a World of Male Power, Semeia 42 (Atlanta, GA: Scholars Press, 1988), especially Esther Fuchs’ article: “For I Have the Way of Women: Deception, Gender, and Ideology in Biblical Narrative,” Semeia 42 (1988): 68–83. 19 Ashley, “Interrogating Biblical Deception and Trickster Theories.” the ‘literary’ in trickster narratives 177 entire construction of this episode in which the trick is very simple brings the reader to see things from Rachel’s point of view. Information is given on precisely how Rachel tricks Laban. This leads the reader to follow the progress of the deceit and draws him to the side of the trickster.

Conclusion

Having taken into account these five different trickster stories, it becomes obvious that the narrators do not always use the same literary technique. Firstly, in definitive, no link can be established between the complexity of the plot and the degree of sophistication of the discourse. The very simple “snatch and hide” of Rachel in the teraphim episode is the pretext to a rather long story, whereas in the murder episode, the possibly complex means to bring Abner back to David’s camp to speak with Joab are expressed in one verse. In another text, an exchange which may seem somewhat difficult to achieve brings a man in love to sleep with another woman than the beloved one he had worked for seven years. The inter- change itself is performed in a wink. On the other hand, the quite rudimentary disguise of the Gibeonites is rendered at length. No link can be established neither between the hero of the story or the one of the episode, and the rendering of the trick. Both in the blessing and in the wedding episode, the narrator has used all means to draw the reader to take sides with the hero. In the blessing episode, fully explaining and describing the means of the ruse bring the reader to the side of Jacob the trickster. In the wedding episode, on the contrary, the reader is left without a clue in the same situation as Jacob: “In the morning, behold, it was Leah.” These two situations appear at first to set a rule, but the analysis of other trickster narratives shows that this is not the case. The opposite construction can also be found: in the Gibeonites episode, the description of the trick is complete, but the reader is put on the side of the victims. In the murder episode, however, although there is no information on the means of the trick, the reader is on the side of the trickster. Each one of the ruses, in the selection analyzed here, displays a specific literary organization. The modalities of the trick, the complexity of its rendering in the story and the knowledge of the reader are characteristic to each one of them. As concerns the modalities of the trick, the texts that we are discussing can be arranged from the simplest trick with Abner being pulled aside by Joab in the murder episode to the most complex with Jacob cooking and disguising himself as Esau in the blessing episode. 178 anne-laure zwilling

Between these, one can find Rachel doing quite a basic move, simply shov- ing the teraphim in her saddle, while the Gibeonites need to use disguise and artifacts. This is explained in a story far longer than necessary to the sole understanding of the trick; whereas the modalities of the possibly complex exchange of the bride in the wedding episode are largely left to the reader to guess. Finally, the knowledge of the reader is nearly always superior to that of the victim, the only text where it is clearly not the case being the murder episode. There is, in conclusion, no definite literary design for the rendering of the trick which determines the combination of modalities, rendering and awareness of the reader chosen by the narrator. However, one should not jump to the conclusion that this combination happens by chance. Quite the contrary, the narrator combines together the different possibilities—for instance to inform the reader or not, provide a short or detailed account of the trick for instance—in order to draw the reader either to the side of the victim or to the side of the trickster. The narrator chooses and organizes the rendering of the trick as a literary device, aiming at bringing the reader to take sides. As explained in a previous article,20 the lengthy explanation of the trick combined with a great knowledge of the reader and a slow narrative pace in the blessing episode brings empathy towards Jacob and draws the reader to the side of the hero. He is here the trickster. In the wedding episode, however, the total surprise of the victim shared by the reader brings again the reader on the side of the hero—who this time is the victim. These two literary combinations do not, however, exhaust the possibilities. In the murder episode, where again the reader is completely surprised, this very surprise brings the reader to share the emotion of the trickster more than of the victim. The Gibeonites narrative is quite different, explaining several times the modes of the trick. It is destined here to allow the reader to understand how the Israelites were tricked and to render acceptable the covenant. The explanation given here draws to the side of the victim. Finally, in the teraphim episode, the awareness provided to the reader builds the empathy with Rachel and draws to opt for the trickster. In each narrative, the different modes of management are arranged in order to allow the narrator to bring the reader either to the side of the trickster or to the side of the victim. The various combinations encountered testify that this arrangement has nothing to do with chance. They demonstrate, if need be, the importance of the ‘literary’ in the trickster narratives.

20 Zwilling, “Le narrateur ‘intrigue’ son lecteur,” 171–180. the ‘literary’ in trickster narratives 179

Summary Table

Genesis 29 2 Sam. 3:26–27 Genesis 31 Genesis 27 Joshua 9 Marriage Murder of Teraphim Esau’s blessing The Gibeonites Rachel / Leah Abner Aim Overcome Esau / Unclear Treaty with Kill Abner Get away with blessing Joshuah the teraphim Modalities Language Lie No / Lie None Lie / Twist ommission Preparation Long None (unclear) Long Send for Abner Short Artefacts Meal None Many Maybe None (“secretly”) Disguise Animal’s skin None Yes None None Comparse Rebecca None (unclear) Many men None None Support of the Elaborate trick Quick action Elaborate trick Quick action Double trick entendre Discourse Reservations Refusal at first No ? No No Presentation Yes No Yes No No Repetition Yes No Yes No No Characterization By Isaac By Jacob By narrator and No No Joshua Showing / telling Showing & Showing Showing & Showing Showing & telling telling telling Roles Hero Jacob Jacob Joshuah Joab (David) Rachel Victim Esau Jacob Joshuah Abner Laban Trickster Jacob Laban Gibeonites Joab Rachel

Reader Expectation Strong Absent Medium Weak Strong (questions of (Jacobs’ Isaac) words) Knowledge > victim (Unclear) ? > victim < victim > victim Reader is with Trickster Victim Victim Trickster Trickster Call to the Empathy Surprise Explanation Surprise Empathy reader

Part Two

Other Ancient Jewish Writings

“Friends Hearken to Your Voice”: Rabbinic Interpretations of the Song of Songs

Tamar Kadari

Abstract

The article argues that, in order to fully reveal the richness and multifaceted nature of the rabbinic exegesis of the Song of Songs, one needs to use tools taken from literary studies. The sages regarded the Song of Songs as a poem and were attuned to its poetic qualities. The article identifies three key parameters that should be considered in the study of midrashim on Song of Songs: The number of speakers in the song, the identity of the poetic speaker and to whom he is addressing his song, and the situation of utterance: what event is described in the Song of Songs according to the rabbis and what is the atmosphere that dominates therein. The different midrashic examples illustrate that the usage of these parameters reveals a wide range of exegetical approaches. They determine not just the content and mood of the Song, but also the nature of its holiness, status and function. Furthermore, as I demonstrate, the rabbis were not blind to the poetic character of the Song of Songs; rather, their midrashim reinforce and expand upon the unique stylistic features that characterize the biblical song.

Introduction

The inclusion of the Song of Songs within the biblical canon was not a simple or self-evident process. The controversy regarding the sacred status of the scroll and its canonization emerges from a number of tannaitic sources,1 including the oft-quoted statement of Rabbi Akiva that “the Song

1 M. Yad. 3.5; b. Meg. 7a; Avoth de Rabbi Nathan, version A:1 ed. Schechter, 2. On the meaning of defiling the hands and the issue of canonization, see bibliography in Menaḥem Haran, The Biblical Collection: Its Consolidation to the End of the Second Temple Times and Changes of Form to the End of the Middle Ages (Hebrew; Jerusalem: Magnes Press, 1996), 202 n. 3; Shamma Friedman, “ ‘The Holy Scriptures Defile the Hands’: The Transforma tion of a Biblical Concept in Rabbinic Theology,” in Minḥah le-Nahum: Biblical and Other Studies Presented to Nahum M. Sarna in Honor of his 70th Birthday, ed. Mark Brettler & Michael Fishbane (Sheffield: JSOT Press, 1993), 117–132; Sid D. Leiman, The Canonization of Hebrew Scripture: The Talmudic and Midrashic Evidence (New Haven: Connecticut Acad emy of Arts and Sciences, 19912), 188 n. 489, who questions whether there even was an act of canonization. 184 tamar kadari of Songs is the holy of holies.”2 Against this background, it is of particular interest to examine the interpretations of the Song of Songs during the rabbinic period. Once the Song of Songs had become an integral part of the Holy Scriptures, it was used by the rabbis in various midrashic techniques (petiḥtaot, petirot, parables etc.) and these were incorporated in different rabbinic collections. A special collection was devoted to the midrashim on the Song of Songs, known as Midrash Song of Songs Rab- bah (ShirR).3 One of the questions that engage academic scholars, is whether the rabbinic interpretation of the Song of Songs differs from that of other biblical works. Does the manner in which they interpret its verses reflect a confrontation with the unusual nature of its contents, i.e., love songs, and the concrete physical descriptions that appear in the text? Rabbinic interpretation of the Song of Songs has been the focus of a number of studies. Attention was drawn towards the allegorical nature of the midrashim on the Song of Songs. Scholars have seen the allegorical interpretation as a way of removing “moral obstacles” caused by the literal interpretation of this book.4 They emphasized that, besides resolving the ethical-educational problem, the allegorical interpretation also gave the Song of Songs a heightened value: Whereas the other books of the Bible do indeed proclaim the bond of love between Israel and the Lord, only the Song of Songs is a dialogue of love, a conversation between man and God that gives religious faith a kind of intensity no other form of expression can.5 The allegorical interpretation was also the main focus of studies addressing the relation between early Jewish and Christian exegesis of the Song of

2 M. Yad., ibid. (MS. Kaufmann); t. San. 12.10: “R. Akiva said: One who raises his voice has ,(כמין זמר) in the Song of Songs in the banquet house and makes it into a kind of song no portion in the World to Come.” 3 Two other compositions from a somewhat later period are: Midrash Zuta al Shir ha- Shirim, ed. Solomon Buber (Vilna, 19252) (= Aggadat Shir ha-Shirim, ed. Solomon Schech ter [Cambridge: D. Bell, 1896]); and Midrash Shir ha-Shirim, ed. Eleazar Gruenhut, et al. (Jerusalem: Ketav-yad va-sefer, 1995). 4 Yitzḥak Heinemann, Darkei ha-Aggadah (Hebrew; Jerusalem: Magnes Press, 19703), 155–157. 5 Gerson D. Cohen, “The Song of Songs & the Jewish Religious Mentality,” in Studies in the Variety of Rabbinic Cultures (Philadelphia: Jewish Publication Society, 1991), 3–17 (12). As to Cohen, the allegorical interpretation was the reason for including the Song of Songs in the canon. rabbinic interpretations of the song of songs 185

Songs.6 Scholars noted the resemblance and differences between the uses of allegory and discussed the influence between them. As to these scholars, the identity of the female figure—the people of Israel or the Church—stood at the very center of the Jewish-Christian polemic. In addition to their involvement in the allegorical interpretation, scholars were also concerned with the mystical interpretation of the book. According to them, the detailed physical descriptions and expressions of love found in the Song of Songs were understood by certain rabbis as portrayals of the human soul’s longing for closeness to God, to enter into the mystical garden of the Pardes and to acquire knowledge of the divine chariot.7 According to Scholem and Lieberman, alongside the accepted allegorical interpretation of the Song of Songs, there also existed an ancient esoteric midrash of the Song of Songs 5:10–16, which included descriptions of the divine body (Shi’ur Qomah).8 In the present article, we shall examine the rabbis’ interpretation of the Song of Songs from a new perspective. It is my position that the common division into such categories as peshat (literal meaning), allegory and mysticism are too general and do not properly represent the wide range of exegetical approaches used by the rabbis in their interpretation of the

6 Thus, for example Isaac Baer, “Israel, the Christian Church and the Roman Empire” [Hebrew], Zion 21 (1956): 1–49; Ephraim E. Urbach, “The Homiletical Interpretations of the Rabbis and the Expositions of Origen on Canticles, and the Jewish-Christian Disputation,” in Studies in Haggadah and Folk-Literature, eds. Joseph Heinemann & Dov Noy (ScrHie 22; Jerusalem: Magnes Press, 1971), 247–275; Reuven Kimelman, “Rabbi Yohanan and Origen on the Song of Songs: A Third-Century Jewish Christian Disputation,” HTR 73 (1980): 567– 595; Marc Hirshman, A Rivalry of Genius: Jewish and Christian Biblical Interpretation in Late Antiquity (Albany, NY: State University of New York Press, 1996), 67–82. For a discussion of the differing nature of Christian and Jewish allegory, see Daniel Boyarin, Intertextual ity and the Reading of Midrash (Bloomington, IN: Indiana University Press, 1990), 105–116. In a recent article, Alon Goshen-Gottstein “Polemomania: Methodological Reflection on the Study of the Judeo-Christian Controversy between the Talmudic Rabbis and Origen over the Interpretations of the Song of Songs” JS 42 (2003–4): 119–190 [Hebrew], Goshen criticizes the ease with which certain scholars find echoes of the Jewish–Christian polemic in the exegesis of the Song of Songs. Lately David Stern, “Ancient Jewish Interpretation of the Song of Songs in a Comparative Context,” in Jewish Biblical Interpretation and Cultural Exchange: Comparative Exegesis in Context, ed. Natalie Dohrmann & David Stern (Phila delphia, PA: University of Pennsylvania Press, 2008), 87–107 (103–107), suggests using new categories: exoteric and esoteric, which touch upon the practice of reading, its purpose and function. 7 Urbach, “The Homiletical Interpretations,” 249–250. 8 Gershom Scholem, Jewish Gnosticism: Merkaba Mysticism and Talmudic Tradition (New York: The Jewish Theological Seminary of America, 1965), 38–40; Saul Lieberman, “Mishnat Shir ha-Shirim,” appendix D, in Gershom Scholem, Jewish Gnosticism, 123; and see the opposing view by Daniel Boyarin, “Two Introductions to the Midrash on the Song of Songs” [Hebrew], Tarb. 56 (1987): 479–501 (497–499). 186 tamar kadari

Song of Songs. In order to illuminate the rabbinic exegesis of the Song of Songs, I have used tools taken from literary studies. The use of literary tools in order to analyze ma‘asi ḥakhamim and exegetical narratives has witnessed considerable growth over the past thirty years.9 Literary analysis quite befits the midrashic material on the Song of Songs. The scroll opens with the heading: “The Song of Songs which is Solomon’s.” After all, the opening words “the Song of Songs” attribute the material to a specific literary category: that of poetry.10 As I will demonstrate below, many sayings can support the fact that the rabbis understood the Song of Songs as poetry. Such as that of Rabbi Judah ,(זמרני) son of Ilai in Midrash Song of Songs Rabbah 2.16.1: “He sings of me Or the saying by Rabbi Yitzḥak in Midrash ”.(זמרתיו) and I sing of him Song of Songs Rabbah 1.2.1: “The Song of Songs [. . .] that song which was And Midrash Song of Songs .(השרים השוררים) ”said by the singing singers Rabbah 1.1.10: “Rabbi Jonathan says: He (Solomon) wrote Song of Songs .(דברי זמר) ”first [. . .] When a man is young he composes songs Some literary theoreticians who examine the process of reading and the activity of interpretation note that the reading of a poem differs from that of a story, play or letter. A poem arouses certain expectations on the part of the reader that influence the manner of his reading, his relation towards the sequence of words, and the insights that emerge from this reading.11 Accord- ing to Culler, the poem is understood by its readers as a complete work that can provide all the information necessary for its understanding. During

9 Jonah Fraenkel was the first to analyze ma’aseh ḥakhamim using close reading and tools of new critics, see Jonah Fraenkel, Studies in the Spiritual World of the Aggadic Nar rative (Hebrew; Tel Aviv: Hakibbutz Hameuchad, 1981); Jonah Fraenkel, The Aggadic Nar rative: Harmony of Form and Content (Hebrew; Tel Aviv: Hakibbutz Hameuchad, 2001); and recently, Inbar Raveh, Fragments of Being—Stories of the Sages: Literary Structures and Worldview (Hebrew; Or Yehuda/Beer Sheva: Dvir, 2008). For application of Fraenkel’s method in the exegetical narrative, see Ofra Meir, The Homiletic Story in Genesis Rabbah (Hebrew; Tel Aviv: Hakibbutz Hameuchad, 1987); Joshua Levinson, The Twice Told Tale: A Poetics of the Exegetical Narrative in Rabbinic Midrash (Hebrew; Jerusalem: Magnes Press, 2005); ibid., “Dialogical Reading in the Rabbinic Exegetical Narrative”, PT 25 (2004): 497– 528. Levinson investigates the reading dynamics of the rewritten Bible as a ‘dialogical read ing’ of two distinct types of discourse, narrative and exegetical. Applying critical notions as intertextuality on midrash, see Boyarin, Intertextuality. For the use of contemporary literary theory on midrashic studies and its impacts, see David Stern, Midrash and Theory: Ancient Jewish Exegesis and Contemporary Literary Studies (Evanston, IL: Northwestern University Press, 1996), who also brings a thorough review on the midrash–theory con nection, 1–13. ”.is the term for both “song” and “poem שיר The Hebrew word 10 11 Jonathan Culler, Structuralist Poetics: Structuralism, Linguistics and the Study of Lit erature (London: Routledge, 19927), 161–162. rabbinic interpretations of the song of songs 187 the course of reading a poem, the reader attempts to resolve a number of questions, such as: Who is the poetic speaker? What is his mood? Whom is he addressing? What is the dramatic situation of the Song?12 The answers to these questions occur during the course of the interpretive process and serve as a kind of fictional construct that the reader creates during the course of reading the poem.13 I shall attempt to demonstrate that, when they set about interpreting the Song of Songs, the rabbis asked themselves these very questions. Through the answers they provided, the rabbis acquired exegetical keys to the understanding of the Song.14 As I shall show below, the positions of the rabbis vary, and give different answers to each of these questions. This article is divided into three parts. In the first part, I shall examine the number of speakers in the Song: Is the Song a conversation among a number of participants or is it a single voice of one speaker only? The second part will focus upon the identity of the speaker: Who is the poetic speaker? And to whom is the Song addressed? The third part will deal with the situation of utterance: we shall ask what event is described in the Song of Songs? What is the psychological state of the speaker and the atmosphere prevailed in the Song? We shall begin each part with a theoretical literary discussion; thereafter, we shall examine the rabbis’ approach to these issues in their exegesis to the Song of Songs. For the sake of clarity, in each part of the paper, I shall focus upon one particular aspect. However, I would argue that to fully understand the rabbinic exegesis of the Song of Songs, one needs to relate to all these questions together: How many speakers are there in the poem? Who are

12 Culler, Structuralist Poetics, 165. Such arguments were later also brought up by theo reticians of the ‘New Criticism’ (a movement in literary theory), that grasped the poem as a self-contained, self-referential aesthetic object. See the chapter “The Heresy of Paraphrase” by Cleanth Brooks in his book The Well-Wrought Urn: Studies in the Structure of Poetry (New York: Reynal and Hitchcock, 1947), 192–214. 13 This is true of every literary work regardless of its genre, but appears most intensely in a poem. 14 Origen brings a tradition in the name of the Hebrews, that resembles the inspired scriptures to a house with locked rooms, in each of them lies a key to another. Rav Sa’adia Gaon, a few hundred years later, writes that the Song of Songs resembles locks to which the keys have been lost. See Jay C. Treat, “Lost Keys: Text and Interpretation in Old Greek Song of Songs and Its Earliest Manuscript Witnesses” (PhD diss., University of Pennsylva nia, 1996), XV; Boyarin, Intertextuality, uses the same term in chapter 7 “The Song of Songs Lock or Key,” where he defines the Song of Songs as a hermeneutic key to the hermetic Torah. In my perceptive, the rabbis grasped the Song of Songs as an obscure text, as did Rav Sa’adia, and they provided keys to unlock it at the beginning of ShirR on the opening verses. 188 tamar kadari they? What is the poetic situation? These various aspects complement one another, and one may find them in various combinations in Rabbinic exegesis on the Song of Songs. By asking the questions which the rabbis’ themselves posed in relation to the biblical text, we shall uncover a richer and more multifaceted picture than that presented thus far.

The Number of Speakers

T.S. Eliot distinguishes among three kinds of voices in the poem: the voice of the poet talking to himself (lyric poetry), the voice of the poet addressing his audience, and the voice of the poet when he attempts to create a dramatic character addressing another imaginary character (dramatic poetry or poetic drama).15 Examination of rabbinic midrashim on the Song of Songs suggests that one of the questions that engaged them when interpreting this book, was the number of poetic speakers: does one hear in the Song of Songs the voice of a single speaker addressing a certain listener, or is one dealing with a dialogue or conversation among a number of participants? Con- frontation with this question already appears in tannaitic sources, in Tosefta Sotah (Lieberman ed.) 9.8: In similar fashion, one says, “Under the apple tree I awakened you” (Songs 8:5)—thus said the Holy Spirit. “Set me as a seal upon your heart” (Songs 8:6)—said the congregation of Israel. “For love is fierce as death” (ibid.)—said the nations of the world. Three utterances (devarim) alongside one another. The Tosefta identifies three speakers in the sequence of verses in the Song of Songs: the Holy Spirit, the People Israel, and the nations of the world. Thus, according to this approach, the Song of Songs reflects a conversation among a number of participants, which might be called dramatic poetry. Dramatic poetry is characterized by the fact that the author has “divided loyalties.” He needs to divide the poem among the characters, and to vary the style in accordance with the speaker from whose mouth the words are spoken.16 David Stern has noted that in the biblical text the heart of the tension of the Song appears in the binary relationship between the lover and

15 T.S. Eliot, “The Three Voices of Poetry,” On Poetry and Poets (London/Boston: Faber and Faber, 1957), 96–112 (89). 16 Eliot, On Poetry and Poets, 94–95. rabbinic interpretations of the song of songs 189 the beloved. Yet in rabbinic midrash and in Origen’s interpretation of the Song, this is turned into a triadic relationship: the husband, the wife, and the mistress. The tension is between the latter two and concerns the question whom the husband truly loves.17 The example above can support Stern’s distinction. The Tosefta wishes to emphasize that a direct conversation occurs only between God and His people, while the nations of the world can merely witness it from afar. Those verbs that are uttered in the first and second person—“I awakened you” and “Set me”—are attributed to the two lovers who are expressing their love for one another. The last sentence, “for love is fierce as death” is seen as a general statement uttered by an observant bystander. More- over, in the voices of the two lovers one hears closeness and longing for one another: “Under the apple tree I awakened you,” “Set me as a seal upon your heart,” while in the words of the nations of the world a certain note of jealousy is sounded from the mouth of an outside observer: “For love is fierce as death” and the end of the same verse: “jealousy is as severe as Hell.”18 The understanding of the Song of Songs as a dialogue between two lovers (without taking into account the observer from the side), follows likewise from the words of the tanna, Rabbi Judah ben Ilai, in Midrash Song of Songs Rabbah 2.16.1, “My beloved is mine”: Rabbi Judah son of Ilai said: He sings of me, and I sing of him. He praises me and I praise him. He calls me, “My sister, my love, my dove, my perfect one” (Songs 5:2) and I say to him, “This is my beloved and this is my friend” (5:16). He says to me, “You are beautiful, my love” (4:1), and I say to him, “You are beautiful, my beloved, truly lovely” (1:16). He says to me, “Happy are you, O Israel! Who is like you” (Deut 33:29). And I say to him, “Who is like you, O Lord, among the gods?” (Exod 15:11). He says to me, “And who is like Thy people, like Israel, a nation one in the earth” (2 Sam 7:23). And I unify His Name twice every day, “Hear O Israel, the Lord our God, the Lord is One” (Deut 6:4).19

17 Stern, “Ancient Jewish Interpretation,” 106. 18 This tendency is emphasized in the Mekhilta de-Rabbi Ishmael, Masekhta de Shirata, Beshalaḥ 3 (ed. Horowitz-Rabin), 127. Cf. below, p. 185. These texts include martyrological allusions. See, Daniel Boyarin, “Martyrdom and the Making of Christianity and Judaism,” JECS 6 (1998): 562–577. 19 Cf. Shir Zuta 1.15 (Aggadat Shir ha-Shirim, Schechter, 18). 190 tamar kadari

Rabbi Judah ben Ilai locates certain verses in the Song of Songs that repeat one another as a kind of echo. Such repetition is among the chief characteristics of poetry, and may be the reason for which he refers to them as a song. The words are uttered once by the lover and again by the beloved; thereby, the lovers enumerate the virtues of one another. According to Rabbi Judah, the repetition indicates that the lover and the beloved listen to one another, responding to one another using the same linguistic patterns: “my love”—“my beloved.” “You are beautiful”—“You are beautiful.” After two examples from the Song of Songs, Rabbi Judah brings verses from other biblical books: “Who is like you”—“Who is like you,” “a nation one in the earth”—“the Lord is One.” These verses are now, too, understood as a dialogue between two lovers. The juxtaposition of these examples sheds light on the difference between the Song of Songs and other hymns: in the latter examples, the words of each speaker are taken from a different context and even from a different biblical book, making the dialogue between them indirect. The Song of Songs, by contrast, differs from all other biblical songs in that it includes both the words of the lover and those of the beloved, who speak with one another in a continuous exchange. This unique aspect of the dialogue in the Song of Songs is also mentioned in Midrash Song of Songs Rabbah 1.1.11: In all other hymns [in the Bible] either the Almighty praises Israel or Israel sings the praises of the Almighty [. . .]. However, only here [. . .] their praise to God is answered by praise to them. Testimonies to the understanding of the Song of Songs as a dramatic song may also be found among non-Jewish exegetes of the same period. Origen, a third-century church father living in Alexandria and Caesarea, wrote two separate compositions on the Song of Songs.20 The first was a set of homilies that were preached in church at Caesarea in the period 239–242 ce. The second was a commentary in ten books that was begun in Athens in 245 ce and completed (books 5–10) in Caesarea in the years 246–247 ce. In the prologue to his Commentary and in the opening of the first Homily, Origen explains that the Song of Songs is a marriage-song (epitha- lamium) presented in the form of a drama. He identifies four speakers in this drama: the bride, the bridegroom, the maidens, and the friends of the bridegroom, and assigns them their lines throughout. Origen treats this

20 Joseph W. Trigg, Origen (London/New York: Routledge, 1998), 1–66. rabbinic interpretations of the song of songs 191 drama as a single literary unit, though he breaks it into segments created by the imagined dramatic situation.21 Additional indication to the understanding of the Song of Songs as a drama appears in two manuscripts of the Septuagint from the fourth century ce, the Codex Alexandrinus and the Codex Sinaiticus. They assign the verses to various characters by noting the speakers in red ink, thereby presenting the dramatic reading.22 From the examples above, it is evident that the perception of the Song of Songs as poetic drama is very ancient. This is attested both in Jewish and in Christian works from the third century ce onwards. It continues to appear in amoraic material and in later sources as well.23 Alongside the approach which regards the Song of Songs as a dramatic poem, there is another view in which the Song expresses the voice of a single speaker.24 In order to illuminate the different opinions regarding the voice of the single speaker, we shall turn to our second parameter, identifying the poetic speaker.

Who is the Poetic Speaker?

The theoretician and literary scholar Jonathan Culler writes that the poetic speaker is a figure created by the reader’s imagination during the process of reading. His identity is shaped by means of information that the reader gathers during the course of reading the poem. Such a procedure is a device whose implications must be incorporated within our interpretation of the poem, forcing us to construct a fictional situation of utterance, including details and allusions related to the speaker’s mood, activity, and setting. The reader, according to Culler, is aware that the

21 Origen: The Song of Songs: Commentary and Homilies, trans. and annot. Ruth P. Law son (New York: Newman Press, 1975), 21–22. 22 Treat, “Lost Keys,” 399–407. 23 For example, in ShirR 1.9.1: “The nations of the world say to Israel, ‘What is your beloved more than another beloved?’ How is your God [to be different from] all other gods? Who is your patron from other patrons? And Israel answers them, ‘My beloved is radiant and ruddy.’ He was radiant in Egypt, and ruddy in Egypt [. . .] Radiant at the Sea [. . .] and ruddy at the Sea [. . .].” 24 Eliot, “Three Voices,” 96, compares this to a sermon or story with a conscious social purpose. In order to create communication, the poet addresses his audience using his own voice, or one of the characters voices. 192 tamar kadari poetic speaker is a fictional construction, one that is necessary in order to understand and interpret the poem.25 A rabbinic dispute regarding the poetic speaker in Song of Songs appears in Midrash Song of Songs Rabbah 1.2.1. Rabbi Judah ben Simon said: It was said at Sinai. “The Song of Songs”—the song recited by the upright (yesharim) at Sinai, as is said, “He is a shield to those who walk in uprightness” (Prov 2:7). Rabbi Yitzḥak said: Song of Songs was recited at the Sea—that song which was said by the singing singers (ha-sharim ha-shorerim), of which it is said, “the singers in front, the minstrels last” (Ps 68:26). It was taught in the name of Nathan: The Holy One blessed be He in the glory of His greatness said it, as is said, “The Song of Songs, which is Solomon’s” (Songs 1:1)—the king to whom peace pertains (sheha-shalom shello). Rabban Gamaliel said: The ministering angel said it. “The Song of Songs” (ibid.)—the song of the ministers (shir hasarim), the song that was said by the supernal ministers.26 This passage deals with the essential question: Who is the poetic speaker in the Song of Songs? Underlying their question is a sense that the rabbis saw the identification of the speaker as a substantive factor in understanding the Song. To answer this question, the rabbis refer to the words “the Song of Songs,” which they interpret in various ways. These words, which serve as the book’s title, are understood as a kind of heading to the work as a whole, expressing its nature and its contents. In the dispute cited above, all four rabbis regard the Song of Songs as a song recited by a single speaker (or a group of speakers). They disagree, however, as to the identity of this speaker: the upright ones at Sinai (Israel), the singers who sang at the Sea (Israel), God, or the ministering angels. The first two statements, attributed to the amoraim Rabbi Judah ben Simon and Rabbi Yitzḥak, express the position that the speakers in the song are Israel, referred to either as the “upright” or the “singers.” The difference between their positions relates to the situation of utterance. According to Rabbi Judah, Song of Songs depicts the revelation at Sinai,

25 Culler, Structuralist Poetics, 165–166. 26 These positions appear as part of the dispute “Where was it said?” further discussed on p. 202. One may distinguish between two primary blocks of material: the dispute about “Where was it said?” (four positions) and the dispute about “Who said it?” (four positions), see Tamar Kadari, “On the Redaction of Midrash Shir Ha-Shirim Rabbah” (Hebrew; PhD diss., Hebrew University of Jerusalem, 2004), 115–118. rabbinic interpretations of the song of songs 193 because of which Israel are referred to as “the upright.” In contrast, Rabbi Yitzḥak thinks that Song of Songs portrays the splitting of the Sea of Reeds, at which Israel sang the Song of the Sea, for which reason they are referred to as “singers.” I will discuss the difference between these two approaches further on.27 The two latter sayings are attributed to tannaim. Rabbi Nathan thinks that God himself is the speaker in the song, as stated at the end of the verse, “the Song of Songs, which is Solomon’s” (Song 1:1). Shelomo (Sol- omon) is a term used to refer to the Almighty, the king of peace. And according to Rabban Gamaliel, the ministering angels are the speakers in the song—“the song of the ministers,” i.e., the supernal ministers. We can conclude that there are two principled positions with regard to poetic speaker in the Song of Songs. The two amoraim represent Song of Songs as an earthly song of Israel, as opposed to the two tannaim who see it as a heavenly song of God or of the angels. In the following section, I will examine each of these approaches and attempt to discover the worldview which lies behind them.

An Earthly Song of Israel The perception of the Song of Songs as an earthly song, recited by the people of Israel at the splitting of the Sea or at the revelation at Sinai, is consistent with the allegorical-historical approach. According to this view, the song is not concerned with the love of a man and woman in the literal sense, but rather describes the love between God and the people of Israel as expressed at a specific historical moment. The allegorical interpretation regards the Song of Songs as an extended metaphor; the characters and events described therein signify an entire sequence of events and ideas. The speaker is the congregation of Israel, who describes her long- ings for her beloved, namely, the Holy One blessed be He, in a well-known historical event. The identification of the speaker with Israel is associated with the historical-allegorical approach. There are many examples in which verses from the Song of Songs are interpreted with the heading “The Congre- gation of Israel said before the Holy One blessed be He [. . .].” Thus, for example, in Midrash Song of Songs Rabbah 2.9.1:

27 See below, “The Situation of Utterance”, p. 202. 194 tamar kadari

“My beloved is like a gazelle” (Songs 2:9). R. Yitzḥak said: The Congregation of Israel said before the Holy One blessed be He, Master of the Universe, You say to us come. You come to us first. “My beloved is like a gazelle” (ibid.): Just as the gazelle jumps and skips from one mountain to another, from one valley to another, from one tree to another, from one hut to another, from one fence to another, so did the Holy One blessed be He jump and skip from Egypt to the Sea, and from the Sea to Sinai [. . .].28 There is an array of rabbinic midrashim that begin with the heading, “The congregation of Israel said [. . .]”, as noted and collected by Wilhelm Bacher.29 This heading also appears in the passage from Tosefta Sotah cited above. The perception of Song of Songs as an earthly song is thus common to all those midrashim in which the speaker or one of the speakers is the people of Israel—this, in contrast to those statements which see Song of Songs as the song of the angels or of God, recited in the supernal realms.

A Heavenly Song of the Angels According to the sayings of Rabban Gamaliel and Rabbi Nathan, the Song of Songs is a heavenly song. In the parallel to the above passage in Shir ha-Shirim Zuta 1.1, this statement appears in expanded form and with the names of the rabbis interchanged: “The Song of Songs” Rabbi Nathan said: The ministering angels said it, as is said, “The Song of Songs”—the song of those who sing before the Holy One blessed be He, the King of Kings, and say: “Holy, Holy, Holy.” Rabban Gamaliel said: The Holy One blessed be He says it, as is said: “The Song of Songs”—the most superb of all songs. And it was said by the patriarchs, and the righteous, and the prophets, and the ministering angels. And who said it? He to whom peace pertains, He who behaves with peace towards all His creatures.30

28 Hebrew text from Genizah fragment Oxford–Bodleian 2669/4, published by Tamar Kadari, “Two Genizah Fragments of Shir HaShirim Rabbah,” Kobez Al Yad 20 (2011): 3–47 (40). For the explanation to this passage as suggested by Saul Lieberman, see Lieberman, “More about Oaths and Vows in Israel,” Tarb. 27 (1958): 183–189 (187) [Hebrew], that refers to the parallel in Pesikta de Rav Kahana 5:8. 29 Wilhelm Bacher, Erkhei Midrash (Hebrew; Tel Aviv: Ahdut, 1923), s.v. “Knesset Yis rael”, 206–208. There are also numerous other examples that appear without this heading. Parallel to this, it would be a desideratum to gather all those places beginning with the heading, “Said The Holy One blessed be He [. . .].” 30 A similar version was extant before R. Moses Teko, Ketav Tamim, ed. Joseph Dan based on MS Paris H711 (Jerusalem: Merkaz Dinur, 1984), 40: “At the beginning of Midrash Shir ha-Shirim, R. Nathan says: ‘The Song of Songs’—the ministering angels said it, those that sing before the King of Kings, the Holy One blessed be He, saying: Holy, Holy, Holy.” R. Eliezer of Worms, in Sefer Ha-Rokeah, writes in his interpretation rabbinic interpretations of the song of songs 195

According to the first approach, the Song of Songs is the song that the ministering angels sing before the Holy One blessed be He. Its content is that of a heavenly song, similar to the Qedushah (doxology) recited in heaven: “Holy, Holy, Holy.” Further evidence for the view that Song of Songs is the song of the angels is preserved in a piyyut (liturgical poem) by Yannai, a sixth century poet, in a qerovah for Passover devoted entirely to the Song of Songs: And we shall be to you a holy people / like the holy erelim [a kind of angel] / and sing in a holy song / for all the songs are holy of holies / for with them the holy Seraphim shall sanctify You / like the secret of the holy Elim / and the holy people / who sanctify Your name in holiness.31 In this strophe, Yannai describes the Song of Songs, regarded as the holy of holies, as one of the versions of the Qedushah recited in Heaven, drawing a comparison between the songs of the Erelim and Seraphim in Heaven and the Qedushah recited in the synagogue by the people of Israel, “the holy people.” At the root of this approach seems to lie the assumption that words of praise directed towards the beloved, including the detailed physical descriptions in Song of Songs, are in fact descriptions of the Holy One blessed be He. They are possible, because the angels envision the Almighty in His full beauty and splendor. There are only a few extant traces of the view that the Song of Songs is the song of the angels. Some indirect evidence for the existence of this exegetical approach may be found in those midrashim that use verses from the Song of Songs to express the distance and tension between the angels and the people Israel. The angels in heaven imitate the praises uttered by Israel and, on the other hand, at times the people of Israel attempt to resemble the angelic hosts.32 Rabbi Akiva may have had this line of interpretation in mind, when he stated that “the Song of Songs is the Holy of Holies”—and it was not for naught that he used a linguistic phrase reminiscent of the Qedushah.

of Songs 1:1: “There are three songs here, corresponding to Holy, Holy, Holy” (MS. Oxford 720, Neubauer Catalogue 1576). 31 Maḥzor Piyyutei R. Yannai le-Torah ule-Mo’adim, ed. Zvi Meir Rabinowitz (Vol. II; Jerusalem: Bialik, 1985–87), 288, and his notes to line 167. 32 The comparison between the army of Israel and the army of angels also appears in ShirR 2.7.1. The praises of Israel to God (reading of Shema, reading the Torah and prophets) serve as an example for the angels in ShirR 8. 13.1. 196 tamar kadari

A Heavenly Song of the Holy One Blessed Be He The final position in the dispute in Midrash Song of Songs Rabbah 1.2.1 is one suggesting that Song of Songs is the song sung by God Himself: They taught in the name of Rabbi Nathan, The Holy One blessed be He, said it in His great Glory, as is said: “The Song of Songs, which is Solomon’s”—of the King of peace. Rabbi Nathan interprets the opening verse of the book as an explicit attribution of authorship: “The Song of Songs, which is Solomon’s.” He inter- ,שלומו :prets the name Shelomo as it sounds, but with a change in spelling as if it were a possessive noun: “his peace” or “He to whom peace pertains.” Peace belongs to God; He is referred to as “He who makes peace.”33 In a similar fashion, we read in Shir ha-Shirim Zuta 11: “Who said it? He to whom peace belongs, who brings peace upon all His creatures.” By means of this simple exegetical move, Rabbi Nathan removes the Song of Songs from its declared author, King Solomon, and turns it into a work written by God. It should be emphasized that we are not speaking here of an interpretation with only a limited or local significance. The change in the heading has far-reaching implications for the understanding of the book as a whole. This understanding is also reflected in the discussion in b. Shevuoth 35b: Every “Solomon” in the Song of Songs is sacred, as is said, “The Song of Songs, which is Solomon’s” (1:1)—a song to the King, to whom peace pertains. Apart from “O Solomon, you may have the thousand” (Songs 8:12) according to him; and “the keepers of the fruit two hundred” according to the rabbis. And there are those who say that this too is profane: “Behold, it is the couch of Solomon” (3:7). “Against fears at night”—from fear of Ashmedai.34

33 See also Mekhilta de-Rabbi Ishmael, Masekhta de-Pisḥa, Bo 14; Sifra, Shemini 1.15–16, and more; and there may also be an echo of this in Ben Sira 47:25 (ed. Segal), 327; and see Origen, Song of Songs Commentary, 51: “It is, I think, unquestionable that Solomon is in many respects a type of Christ, first in that he is called the Peaceable [. . .].” 34 Rashi on Songs 1:1 interprets all appearances of the word ‘Solomon’ as alluding to God, even those verses concerning which there is a dispute. The Targum generally inter prets Shlomo in a literal way, as referring to King Solomon (with the exception of Songs 8:11). According to Raphael Loewe, “Apologetic Motifs in the Targum to The Song of Songs,” in Biblical Motifs: Origins and Translations, ed. Alexander Altmann (Cambridge: Harvard University Press, Mass. 1966), 159–196 (163), this was a conscious ideological decision; Rashbam followed in the wake of the Targum. rabbinic interpretations of the song of songs 197

The general rule cited in the Babylonian Talmud relies upon the exegesis of the word ‘Solomon’ as it appears at the title verse of the book in the interpretation of Rabbi Nathan. However, he cites the dispute related to a number of verses which are nevertheless understood as relating to King Solomon. It would seem that this approach claims that the Song of Songs as a whole, as the song of God himself, is sacred, with a number of exceptions. This discussion thus limits somewhat the all-encompassing statement of Rabbi Nathan. Another dispute, regarding the terms ‘of the king’ and ‘of King Solomon’ :appears in Midrash Song of Songs Rabbah 1:1.11 ,(במלך שלמה and במלך) Rabbi Yudan and Rabbi Levi said in the name of Rabbi Joḥanan: Wherever it says in this book, “of King Solomon,” it speaks of King Solomon. When it says “the King” by itself, Scripture speaks of the Holy Name.35 Our rabbis said: “of King Solomon,” this speaks of the King of Peace, “the King” by itself, this alludes to the congregation of Israel. This controversy likewise restricts the words of Rabbi Nathan, since according to Rabbi Joḥanan only those places where it says “the King,” refer to God. This is in contrast to the rabbis, who hold that only the phrase “of King Solomon” relates to Him. The understanding of Song of Songs as a song in which the speaker is God, elicits a new question: to whom does He sing his song? The question of the addressee did not arise so long as we were speaking of a dramatic song, in which a number of different speakers, whose identity was known, participated. The transition to one speaker, particularly when speaking of Almighty God, raises the question to whom He addresses himself in his song—and here too we may find more than one answer.

God Sings to His Beloved, the Congregation of Israel In several midrashim the verses of the Song of Songs appear as literally emanating from God’s mouth. God, by uttering these verses, sings the praises of Israel. Thus, for example, in Midrash Song of Songs Rabbah 2.14.1: “O my dove in the clefts of the rock” (Songs 2:14). What is meant by “My dove in the cleft of the rock”? Rabbi Joḥanan said: The Holy One blessed be He said: I call Israel a dove [. . .]

i.e., the Ineffable Name, referring here to God. According to Lev ,בנקיבות :In Hebrew 35 shall surely die.” My thanks to (ונוקב שם ה') he who profanes the name of the Lord“ :24:16 Prof. Daniel Sperber for his explanation of this difficult word. 198 tamar kadari

And in Midrash Song of Songs Rabbah 4.4.2: “Your cheeks (rakatekh) are like halves of a pomegranate” (Songs 4:3). The emptiest one among you (reikan) is filled with Torah like this pomegranate; and there is no need to add, “behind your veil (tzamatekh)” (ibid.)— regarding the modest ones among you, the veiled ones (metzumatin) was among you. And in Midrash Song of Songs Rabbah 4.9.1: “You have ravished my heart (libavtini), my sister, my bride” (Songs 4:9). The Holy One blessed be He said: You had one heart in Egypt, and have given me two hearts (levavot). “You have ravished me with a glance of your eyes” (ibid.)—with the blood of the Passover and the blood of circumcision. “With one jewel of your necklace” (ibid.)—this is Moses, the unique and strong one among the tribes. According to this approach, the Almighty sings the Song of Songs to and about the people Israel. Unlike the position noted earlier, there is no dialogue here, but only a song in which there is one speaker.36 This approach has significant consequences for the sanctity of Song of Songs and for the status of the book. If Song of Songs reflects the words of the living God, its sanctity is beyond all doubt.37

God Sings to His Beloved—the Torah The People of Israel is not God’s only possible beloved. In a number of sources the Song of Songs is God’s love song to the Torah. The comparison of the Torah to a woman is a pervasive image in Rabbinic literature, but its appearance in the context of interpretation of the Song of Songs gives it a special significance and sheds light on the book from a new and surprising angle.38 This approach may already be identified in tannaitic sources, in Mekh- ilta de-Rabbi Ishmael, Masekhta Baḥodesh, Yitro 8 (Lauterbach ed., 264): Therefore the Ten Commandments were given, five on one tablet and five on the other tablet—thus in the words of Rabbi Hanania ben Gamaliel. But

36 For further examples, see ShirR 4.4.9; ShirR 4.8.1; ShirR 4.12.1. 37 This was evidently the approach of the redactor of ShirR to the sanctity of Song of Songs. See, Tamar Kadari, “ ‘Behold a Man Skilled at his Work’: On the Origins of the Pro ems Which Introduce Song of Songs Rabbah,” Tarb. 75 (2005–6): 169–172 [Hebrew]. 38 See Arthur Green, “Bride, Spouse, Daughter: Images of the Feminine in Classical Jewish Sources,” On Being a Jewish Feminist, ed. Susannah Heschel (New York: Schocken Books, 1995), 248–260; Elliot R. Wolfson, Circle in the Square: Studies in the Use of Gender in Kabbalistic Symbolism (Albany, NY: State University of New York Press, 1995), 1–28. rabbinic interpretations of the song of songs 199

the rabbis say: Ten on this tablet and ten on the other tablet, as is said, “These words the Lord spoke [. . .] and He wrote them upon two tablets of stone” (Deut 5:19), and it says “Your two breasts are like two fawns, twins of a gazelle” (Song 4:5). In this midrash, those verses from Song of Songs describing the beloved serve as a vehicle for the physical depiction of the Torah. According to the view of the rabbis, just as two twin fawns are identical to one another in all things, so were the tablets entirely identical, with all ten of the commandments written on each one. In this midrash, the beloved is the Torah. Another tannaitic testimony to the identification of the Torah with the beloved in the Song of Songs appears in Tosefta Kippurim 2.15 (Lieberman ed., 238): The two poles of the ark stuck out of the ark, reaching up to the veil, as is said, “And the poles were so long [. . .]” (1 Kgs 8:8). Could it be that they ripped the veil? The Torah says, “But they could not be seen from outside” (ibid.). Could it be that they were not visible from within it? Scripture says, “The ends of the poles were seen [from the holy place]” (ibid.). From this, one may say that the poles were long and reached the veil, and pushed against the partition and were seen from within it. And concerning that, it is said, “My beloved is to me a bag of myrrh, that lies between my breasts” (Songs 1:13). This midrash in the Tosefta is based on the external resemblance between the appearance of the poles of the ark and that of a woman’s breasts, or The verse 39.(בדים/דדים) perhaps on the similarity in sound between them from Song of Songs is interpreted quite vividly: The poles of the ark pushing against the veil are like a woman’s breasts pushing against the cloth of her garment. Yet the ark is concealed from the eyes of all Israel, even from the priests. Its staves are not even visible in the holy place, but are only suggested by the veil, which they push outwards. The ark is the place where the Holy One blessed be He is revealed (see Exod 25:22). In the parallel to the above passage in Shir ha-Shirim Zuta 1:13, this appears in expanded form: “ ‘My beloved is to me a bag of myrrh that lies between my breasts’ [. . .] This is the Shekhinah that was between the two cherubim.”40 Implicit in this image is the notion of the divine revelation that is reserved for the eyes of the beloved alone. Not only are the staves revealed to His

a bag of myrrh,” refer to the incense mentioned in Tosefta“ ,צרור המור The words 39 Kippurim 2.14. 40 Aggadat Shir ha-Shirim, Schechter, 17. 200 tamar kadari eyes, but the ark is His fixed place of residence, as suggested by the verse, “My beloved lies between my breasts.” The following is another example concerning the other features of the body of the beloved, from Midrash Song of Songs Rabbah 7.3.2: “Your belly is a heap of wheat” (Songs 7:3)—this is the [book of] priestly teaching [i.e., Leviticus]. Just as the belly is located with the heart on one side and the innards on the other, it being in the middle, so is Leviticus located with two books on one side and two books on the other side, and it is in the middle [. . .]. “Encircled with lilies” (ibid.)—these are the words of Torah, which are soft as roses. How many commandments and details there are in Leviticus! How many a fortiori and unfit and leftover sacrifices there are in this priestly teaching! This verse is part of a blazon depicting the dancing maiden and the various parts of her body, from feet to head. The comparison of the belly to a heap of wheat suggests abundance and fruitfulness. Just as the belly is the center of the body, the book of Leviticus is the middle book of the Torah. The centrality of the book of Leviticus in rabbinic eyes may be inferred from a number of different sayings, and it is quite possible that the rabbis began the study of the Torah with this book.41 Shlomo Naeh has suggested that the Sifra, the tannaitic midrash on the book of Leviticus, was the earliest rabbinic work transmitted in writing.42 Just as all of Torah learning begins with the middle book, all life begins in the belly, the site of gestation. In all these examples, we have seen that the verses that consist of physical descriptions of the beloved, are interpreted in relation to the Torah. The Torah is described in these midrashim as a female entity, while the verses from Song of Songs used to describe her, lend her the status of the beloved of the Holy One blessed be He. The feminine personification of the Torah reflects an older idea found in the Bible concerning the feminine Sophia or Wisdom.43 Sophia is described as created before all other beings: “The Lord created me at the beginning of his course, as the first of his works of old” (Prov 8:22) and as a feminine enjoyment to God: “I was with Him as a confidant, a source

41 See Lev. Rab. 7.3 (ed. Margaliot, 156); Avoth de Rabbi Nathan, version A:6 (ed. Schech ter, 29); Gunter Stemberger, Introduction to the Talmud and Midrash, trans. and ed. Markus Bockmuehl (Edinburgh: T & T Clark, 19962), 260. 42 Shlomo Naeh, “The Structure and the Division of Torat Kohanim: (A) Scrolls,” Tarb. 66 (1997): 483–515, esp. at 512 [Hebrew]; idem., ibid., “(B) Parashot, Perakim, Halakhot,” Tarb. 69 (2000): 59–104 [Hebrew]. 43 Prov 1:20–32, 9:3, Job 28:12–28. rabbinic interpretations of the song of songs 201 of delight every day, rejoicing in his inhabited world, finding delight with mankind” (Prov 8:30–31). The identification of Torah as Ḥokhmah (Wisdom) became widespread in the classical rabbinic sources.44 For the rabbis, the Torah assumed a personality of its own, and they deciphered it as a pre-existent entity that served as an instrument with which God created the world. The usage of the feminine descriptions in the Song of Songs to define the Torah may reflect a consistent exegetical approach to the book, one that already existed in the tannaitic period, in which the Song of Songs is seen as God’s love song to his Torah—that same Torah which was created even before the supernal beings, which He praises by means of the book, which is this selfsame song.45 The interpretation of the Song of Songs as the song of the Holy One blessed be He to his Torah, or as the angels’ song of praise to God, represents the scroll as a heavenly song in which Israel does not participate. This exegetical direction may have led to the interpretation of the book, according to the approach of Jewish mystical teaching, as an esoteric book, which includes descriptions of the divine body revealed only to the eyes of heavenly live beings.46 To summarize this portion of our discussion: one finds that the rabbis made efforts to identify the speaker in the song and even suggested various different exegetical options. We have found some approaches which view Song of Songs as a dramatic poem comprised by the voices of a number of different figures. We also noted other approaches in which there is only one speaker, whether an earthly or a heavenly one, with varied identifications.

44 Ephraim E. Urbach, The Sages: Their Concepts and Beliefs (Vol I; Jerusalem: Magnes Press, 1975), 286–288; Hans Conzelmann, “The Mother of Wisdom,” in The Future of Our Reli gious Past: Essays in Honor of Rudolf Bultmann, ed. James M. Robinson (London: S.C.M. Press, 1971), 230–243; and see the bibliography cited by Wolfson, Circle in the Square, 123 n. 1. 45 For an extensive discussion of this view, see Tamar Kadari, “ ‘Within it Was Decked with Love’: The Torah as the Bride in Tannaitic Exegesis on Song of Songs,” Tarb. 71 (2002): 391–404 [Hebrew]. 46 David Stern, “Ancient Jewish Interpretation,” 100–102, has recently noted a series of midrashim in ShirR 5:11 that substitute the Torah (and its scholars) for the figure of God himself. Stern suggests that exchanging the physical description of the measurements of Gods body by a depiction of the Torah as a material artifact, is a way to avoid an esoteric interpretation. 202 tamar kadari

The Situation of Utterance

In addition to identifying the quantity and nature of the speakers, the reader of a poem must also decipher the poetic situation: when did the event to which the poem relates, occur and what is the accompanying atmosphere? This information is crucial for the proper understanding of the poem by the reader. If we take as our premise the integrity and independent status of the work, then the answers to these questions must be found within the poem itself.47 Culler emphasizes that the poetic situation is not (only) a biographical matter, but one created within the mind of the reader during the process of reading the poem: The poem is presented as the discourse of a speaker who, at the moment of speaking, stands before a particular scene, but even if this apparent claim was biographically true it is absorbed and transformed by poetic convention so as to permit a certain kind of thematic development. The drama will be one of mind itself when faced with external stimuli, and the reader must take account of the gap between object and feeling, if only in order that the fusion which the poem may enact be taken as an achievement.48

Where was It Said—Four Tannaitic Approaches Within the context of exegesis of Song of Songs, one might ask what, in the opinion of the rabbis, was the context in which the verses of the song were said, or what event it describes. This question was formulated by the rabbis themselves, in Midrash Song of Songs Rabbah 1.2.1, “O, that he would kiss me with the kisses of his mouth” (Songs 1:2). Where was it said? Rav Ḥanina bar Pappa said: At the Sea, as is said, “[I compared you, my love,] to a mare of Pharaoh’s chariots” (1:9). [. . .] Rabbi Joḥanan said: At Sinai it was said, as is said, “O, that he would kiss me with the kisses of his mouth” [. . .]. Rabbi Meir said: It was said in the Tent of Meeting, and is inferred from the following verse, “Awake, O north wind, and come, O south wind” (4:16) [. . .]. The rabbis said: It was said in the Eternal House, and the rabbis also brought proof from that same verse, “Awake, O North wind” [. . .].49 The rabbis ask a basic question: Where was it said? In other words: what is the situation of utterance, or what is the time in which the dramatic

47 Culler, Structuralist Poetics, 165. 48 Culler, Structuralist Poetics, 167. 49 See above, note 26. rabbinic interpretations of the song of songs 203

dialogue within the poem is taking place? This discussion is brought at the beginning of Midrash Song of Songs Rabbah, from which we may infer that the rabbis regarded the deciphering of the poetic situation as an important key to the understanding of the song. As a response to this question, the midrash cites a dispute that consists of two tannaitic positions (those of Rabbi Meir and the rabbis), and two amoraic positions (Rav Ḥanina bar Pappa and Rabbi Joḥanan).50 One should note that all four opinions view Song of Songs as an allegorical song describing a foundational historical event in the life of the nation. Yet, their positions are presented as a dispute, as they disagree regarding the specific event which is referred to: God’s revelation to Israel at the Sea of Reeds, the Epiphany at Sinai, the dedication of the Tabernacle in the wilderness, and that of Solomon’s Temple. Lieberman already noted the main significance of this controversy. He argues that the rabbis are not concerned with a specific, limited interpretation of a particular verse. The rabbis thought that all the verses of the song ought to be read in light of one single approach: “and each one took the trouble to interpret the scroll consistently, according to his own approach.”51 According to Lieberman, the four approaches cited represent consistent views that date back to the tannaitic period.52 To substanti- ate his claim, he cites midrashim, verifying that the approach stating that “it was said at the Sea” was that of the tanna Rabbi Eliezer,53 while that which held that “it was said at Sinai” was articulated by Rabbi Akiva.54 One of the more convincing proofs of his view is found in a series of five petirot55 brought in Midrash Song of Songs Rabbah 2.14.3–6 on the verse “O my dove, in the clefts of the rock”:

50 On the fact that the position of “the rabbis” reflects an earlier approach than that of the Amoraim, see the series of petirot below (note 56); the position of Rabbi Meir already appears in Seder Olam Rabbah 7 without attribution. 51 Lieberman, “Mishnat Shir ha-Shirim,” 119. 52 Lieberman, “Mishnat Shir ha-Shirim,” 118–126; Isaac B. Gottlieb, “The Jewish Allegory of Love: Change and Constancy,” JJTP 2 (1992): 1–17. 53 See, for example, the proofs brought by Lieberman, ibid., of the position of R. Eliezer: “It was said at the Sea”; ShirR 2.2.2; Mekhilta de-R. Ishmael, Bo, 7 (22); ShirR 1.12.1. 54 See, for example, the proofs brought by Lieberman, ibid., of the position of R. Akiva: “It was said at Sinai”; ShirR 1.12.1; Avot de-Rabbi Nathan, Version A, 27 (42). 55 A petira is a midrashic technique commonly used in ShirR. The interpreter assumes the verse contains a concealed meaning and in order to reveal it, he solves each passage by reference to an overall hermeneutic code. See Maren Niehoff, “A Dream Which Is Not Interpreted Is Like a Letter Which Is Not Read,” JJS 43 (1992): 58–84. 204 tamar kadari

Rabbi Lazar resolved this verse in relation to Israel at the time they stood at the Sea [. . .]. Rabbi Akiva resolved the verse as pertaining to Israel when they stood before Mount Sinai [. . .]. Rabbi Yossi the Galilean resolved the verse as pertaining to [subjugation to the] kingdoms [. . .]. Rav Huna and Rav Aḥa in the name of Rav Aḥa bar Ḥanina interpreted the verse following the opinion of Rabbi Meir, in the Tent of Meeting [. . .]. Rav Tanḥuma said: You resolved it according to Rabbi Meir as regarding the Tent of Meeting, I shall resolve it as regarding the opinion of the rabbis [pertaining to] the Eternal House [i.e., Solomon’s Temple] [. . .].56 The first two petirot cite the two familiar positions, those of “the Sea” and “at Sinai,” yet here they are attributed to the tannaim, Rabbi Eliezer and Rabbi Akiva; the two latter solutions, “in the Tent of Meeting” and “in the Eternal House” are cited in the names of amoraim, who interpret the verses of Song of Songs according to the approaches of the tannaim. The statement of Rav Tanḥuma: “You resolved it according to Rabbi Meir as regarding the Tent of Meeting, I shall resolve it as regarding the opinion of the rabbis [pertaining to] the Eternal House“ indicates that he knew the systematic approaches of the tannaim, and wanted to applying them also to verses which he had no specific exegetical tradition.

Other Systematic Approaches However, close examination of Midrash Song of Songs Rabbah 2.14.3–6 “O my dove, in the clefts of the rock”, indicates that the picture is more complex than suggested by Lieberman. The tannaim had further allegorical approaches beyond the four noted by Lieberman. For instance, according to Rabbi Yossi the Galilean, the Song of Songs must be understood as alluding to matters of exile and troubles in the present.57 This view appears in other tannaitic sayings, and may indicate another consistent interpretative approach to the Song of Songs.58

56 For a detailed discussion of this series of petirot, see Kadari, “On the Redaction,” 175–192. 57 Goshen-Gottstein, “Polemomania,” 155, opposes Lieberman, and thinks that the quest for consistent allegorical interpretations does not match the data before us. To my opinion, there were systematic readings, but they were more numerous than those which research has identified thus far. 58 For example, Mekhilta de-Rabbi Ishmael, Masekhta de Shirata, Beshalaḥ 3 (Horowitz- Rabin ed., 127); Mekhilta de-Rashbi 19.17; Sifrei Devarim, §304 (Finkelstein ed., 323); ibid., §305 (325); Midrash Tannaim to Deut 14:2 (Hoffman ed., 73). rabbinic interpretations of the song of songs 205

There are other subjects repeated in the tannaitic midrashim on Song of Songs, such as the Exodus from Egypt,59 the wandering in the desert,60 the ful- fillment of the Torah and its mitzvot in the present,61 the future redemption.62 In the amoraic sayings on Song of Songs, as well, one may identify the repetition of these and other subjects.63 Thus, the allegorical–historical interpretation is not fully exhausted by the four approaches pointed out by Lieberman, but is more multi-faceted and varied.

Atmosphere and Psychological State of the Speaker In addition to the insight that the historical allegory encompasses a wide variety of subjects, one may also identify substantive differences among the various approaches. Uncovering the poetic situation will help us understand the psychological state of the speaker and the atmosphere that prevailed in the song. Thus, for example, the statement of Rabbi Yossi the Galilean depicts the people of Israel in a situation of trouble, “That they were hidden in the shadow of the kingdoms” and, according to the parallel in Mekhilta de-Rashbi 19.17: “ ‘In the covert of the cliff’—this refers to Israel, who dwell alongside their enemies in their afflictions until their time shall come.” Rabbi Yossi depicts the feelings of alienation and the concealment of the divine face, with the Song of Songs as a song of mutual longing and of unrealized yearning for closeness. Other statements interpreting the verses of the Song as referring to subjugation of the kingdoms, depict an even harsher picture. Thus, for example, in the passage from Mekhilta de-Rabbi Ishmael, Masekhta de-Shirata,

59 Mekhilta de-Rabbi Ishmael, Pisḥa, Bo 5 (Horowitz-Rabin ed., 14–15); ibid., 7 (22–23); ibid (25); ibid., Va-yeḥi Beshalaḥ 1 (87–88). There is a lack of distinction in the research between this position and that of the splitting of the Sea of Reeds. However, there are homilies which bring these positions alongside one another, such as ShirR 2.1.1; 4.1.1; 1.5.1. 60 Sifrei Bamidbar, §139 (Horowitz ed., 186); Midrash Tannaim to Deut 1:4 (Hoffman ed., 4); ibid., 20.1 (119). 61 Sifrei Devarim, end of §36 (Finkelstein ed., 67–68); ibid, §41 (86–87); Midrash Tan naim to Deut 15:9 (Hoffman ed., 83); ShirR 6.8.1, in the name of Rabbi Yudan b. Ilai. 62 Mekhilta de-Rabbi Ishmael, Mas. Pisḥa, Bo, 14 (Horowitz-Rabin ed., 52); Mekhilta de- Rabbi Ishmael, Shirah, Beshalaḥ, 3 (Horowitz-Rabin ed., 127–128); Sifrei Devarim §10 (Fin kelstein ed., 18). 63 For a list of the verses and subjects expounded in ShirR, see Birke S. Rapp-de Lange, “Rabbinische Liebe: Untersuchungen zur Deutung der Liebe des Hohenliedes auf das Studium der Torah in Midrasch Shir haShirim Rabba” (PhD diss., Katholieke Theologi sche Universiteit Utrecht, 2003), Appendix 1, 439–462. The list is organized according to the verses in Song of Songs, divided into four categories: past, present, Torah study, and future. 206 tamar kadari

Beshalaḥ 3: “ ‘What is your beloved more than another beloved, that you thus adjure us?’—that thus you die for him and thus you are killed for him. ‘Therefore the maidens [‘alamot] love you’ (Songs 1:3)—they have loved you to death [‘al-mut].”64 The following story in Sifrei Devarim, §305 (Finkelstein ed., 325) likewise conveys a harsh atmosphere: Once Rabban Joḥanan Ben Zakkai was riding upon a donkey and his disciples were walking after him. They saw a young woman gathering barley from beneath the legs of the animals of the Arabs [. . .]. Rabban Joḥanan ben Zakkai said to his disciples: I signed the marriage contract of this [woman], and I read therein a thousand of thousand golden dinars [. . .]. And all my days I sought [the meaning] of this verse, and now I have found it: “If you do not know, O fairest among women, follow in the tracks of the flock, and pasture your kids beside the shepherds’ tents” (Songs 1:8). Do not read “your kids (gediyotayikh),” but “your corpses” (geviyotayikh). For so long as Israel do the will of the Omnipresent, no nation or kingdom can rule them. But when Israel do not do the will of the Omnipresent, they are given into the hands of a lowly nation; and not into the hands of a lowly nation, but beneath the legs of the animals of a lowly nation!65 Gathering barley from the dung heaps is an ancient topos, signifying the most extreme level of hunger.66 The personal situation of this young woman here reflects the lowly state of the nation as a whole. Song of Songs serves as a kind of elegy for the nation, which was once “the fairest among women,” but is now poor and miserable, trampled beneath the feet of the animals of the gentiles.67

64 Zvi Meir Rabinowitz, Ginzei Midrash (Tel Aviv: Tel Aviv University Press, 1977), 13–14 (Mekhilta de-Rabbi Ishmael, ed. Lauterbach, vol II.25–27); Stern, “Ancient Jewish Interpre tation,” 106. Alongside the historical allegory, this saying also sees Song of Songs as a dra matic song, above p. 188. 65 For the most recent discussion of the story, its history and variants, see Menahem Kister, “Studies in Avot de-Rabbi Nathan, Version A, Chapter 17: Editing and Convolution of Traditions,” in Meḥqerei Talmud 3: Talmudic Studies Dedicated to the Memory of Profes sor E. E. Urbach, ed. Yaakov Sussman & David Rosenthal (Jerusalem: Magnes Press, 2005), 703–739. 66 See Kister’s references and bibliography, ibid., 720–721, and n. 88. 67 Kister demonstrates that the negative characteristics of the young woman, which appear primarily in the Tosefta and in its parallels, are exchanged for empathy and a more general theological moral in the versions of the story in Sifrei, in Mekhilta de-Rabbi Ish mael, in the Bavli, and in Avot de-Rabbi Nathan. The concrete aspect in the version of the Sifrei is greatly emphasized in the analysis of Ofra Meir, The Poetics of Rabbinic Stories (Hebrew; Tel Aviv: Sifriat Poalim, 1993), 67–70. rabbinic interpretations of the song of songs 207

A Song of Praise Those who interpreted the Song of Songs as referring to the harsh reality of current times, were addressing the failing and weakness of the people of Israel. This was not accepted by all the rabbis, as demonstrated by another tannaitic dispute in Midrash Song of Songs Rabbah 2.4.1, “He brought me to the banqueting house [lit.: house of wine]” (Songs 2:4). Rabbi Meir and Rabbi Judah [discussed it]. Rabbi Meir said: The Congregation of Israel said: The Evil Urge dominates me like wine, so that I said to the calf, “These are your gods, O Israel” (Exod 32:4). Rabbi Judah said: Enough [You have gone too far,] Meir! One does not interpret Song of Songs in a disparaging way, but only for praise, for the Song of Songs was given only for the praise of Israel. What is meant by, “He brought me to the house of wine”? The Congregation of Israel said: The Holy One blessed be He has brought me to the great wine cellar, which is Sinai. And there He gave me the flags of Torah and mitzvot and good deeds, and I received them with great love.68 The two tannaim, Rabbi Meir and Rabbi Judah, both identify the situation of utterance as the revelation at Sinai, in accordance with the approach of their teacher, Rabbi Akiva.69 They also both identify the speaker with the Congregation of Israel. Nevertheless, Rabbi Judah’s sharp reaction to the interpretation of his colleague indicates that there is a dispute between them regarding a substantive matter. According to Rabbi Judah, Song of Songs is a song of praise, and therefore it is impossible for it to express contempt for the object of its description, the Congregation of Israel. And indeed several verses of Song of Songs focus on the beauty and merits of the lover or the beloved. Hence, they are easily and naturally identified as belonging to the genre of praise songs. One should note that the dispute between Rabbi Judah and Rabbi Meir is not an isolated exegetical disagreement on the interpretation of a particular verse, but rather expresses a principled position, formulated in a sweep- ing way: “One does not interpret Song of Songs in a disparaging way, but only as praise, for the Song of Songs was only given for the praise of Israel.” Here, we find the approach of Rabbi Judah, who in principle opposes the use of Song of Songs to present the failings and shortcomings of the people of Israel. This opposite stance is suggested by Rabbi Meir, and possibly also

68 This dispute reappears with some variations in ShirR 1.12.1. 69 For Rabbi Akiva’s position regarding Sinai, see above, pp. 203–204, while the position of Rabbi Meir was that it was said in the Tent of Meeting. 208 tamar kadari other rabbis. One might speculate that Rabbi Judah, who viewed Song of Songs as a hymn of praise to Israel, did not favor the sayings of Rabbi Yossi the Galilean and the story of Rabban Joḥanan ben Zakkai, both of whom interpreted the book as referring to the harsh reality which existed during the period of the Destruction and thereafter.70 The determination of genre, in the case of Song of Songs, has far-reaching implications regarding the contents and mood of the song, and also impacts questions about the work’s holiness, its status and its function. As I show elsewhere, the determination of the genre to which Song of Songs belongs, has become part of the discussion of its interpretation or exegesis. Returning to the five petirot, the atmosphere reflected in the saying by Rabbi Yossi the Galilean is completely different from the other four tannaitic approaches that appear in Midrash Song of Songs Rabbah 2.14.3–6, “O my dove, in the clefts of the rock”. Rabbi Eliezer, for example, speaks of “Israel, at the hour that they stood by the Sea.” According to Rabbi Eliezer, Song of Songs portrays a foundational historical event in which the entire nation of Israel witnessed the miracle of the splitting of the Sea of Reeds and experienced a direct revelation of the Almighty.71 According to this view, the Song of Songs describes a spiritual and religious high point. The physical descriptions of the lover and of the beloved express a situation of mutual knowledge, of closeness and intimacy, in which both sides behold one another.72 The motif of revelation is common to all four approaches cited above in the dispute “Where was it said?” and it would appear that there was a clear goal in presenting these positions in succession. The context of revelation lends to the verses of the book a dimension of holiness and presents them as words of prophecy, as an expression of the experience of seeing God face to face. This approach gives significance to the exegetical activity as a whole, and it is that which provides a rationale for exegetical engagement in the verses of the Song.

70 The saying of Rabbi Yossi the Galilean, above, p. 204, and the story about the fairest among women, p. 206. 71 Concerning the splitting of the Reed Sea, it says, “And Israel saw the great work [lit.: great hand], which the Lord did against the Egyptians” (Exod 14:31). 72 According to Lieberman, “Mishnat Shir ha-Shirim”, 119, Song of Songs in fact depicts that selfsame historical event. From this, Lieberman goes one additional step, to argue that the description of the concrete seeing of God is indicative of the existence of an esoteric interpretation of Song of Songs. As for us, we prefer to remain within the realm of historical allegory. rabbinic interpretations of the song of songs 209

As against these four tannaitic approaches, Rabbi Yossi the Galilean and Rabban Joḥanan ben Zakkai attempt to draw a contrast between the harsh everyday reality and the reality longed-for, as described in Song of Songs. Through the verses of the song, they seek to once again arouse God’s love for his people.

Conclusion

Examination of rabbinic midrashim on Song of Songs using literary parameters—the number of speakers, their identification, and the situation of utterance—reveals a wide range of exegetical approaches: from a heavenly song of praise of God by the angels to an elegy for the people of Israel, that feel concealment of the divine face while they are trampled beneath the feet of gentiles. A full understanding of the position of a given rabbi requires examination of all parameters. Each illuminates Song of Songs in a new and different light. Such a reading reveals the richness and multifaceted nature of the interpretation of Song of Songs during the rabbinic period.

Between Tradition and Innovation: Seder Eliyahu’s Literary Strategies in the Context of Late Midrash

Lennart Lehmhaus

Abstract

For a long time, late midrashic traditions were perceived as a rather eclectic afterglow of their prototypes in the Golden Age of classic midrash. According to this view, these texts were dominated by narration as a new stylistic paradigm, whereas they are characterized by stagnation regarding their exegetical interests and their religious ideas. These observations hold true for certain aspects of those midrashim. However, one also has to consider other, more subtle strategies of literary transmission, adaptation, and innovation in order to grasp their transformative function as a link between late antique and early medieval times. This contribution will address questions concerning Seder Eliyahu’s complex interaction with and modification of biblical and rabbinic traditions, hermeneutics, and rhetoric in order to convey particular messages or pursue specific goals. This discussion includes a comparative perspective towards other late traditions and earlier rabbinic discourses (e.g., Mishnah, Talmud, midrashim). These findings will raise our awareness to the similarities and differences regarding other rabbinic traditions and help to allocate Seder Eliyahu within the framework of early medieval Jewish literature and its broader cultural context.

Introduction

The majority of early scholars read rabbinic texts as historical evidence and took their reliability in this respect for granted. Later scholars, however, questioned profoundly those assumptions of their precursors and demonstrated the vagueness and naïveté underlying them. From their inquiries, it became evident that neither a pure historical nor a solely biographic reading (i.e., the reconstruction of the life of a certain rabbinic sage) of rabbinic traditions leads to a deeper understanding of the historical background of those texts and their producers.1

1 A profound critique of earlier attempts of rabbinic historiography and the background of this scholarship is given in Seth Schwartz, “The Political Geography of Rabbinic Texts,” in The Cambridge Companion to the Talmud, ed. Charlotte E. Fonrobert & Martin S. Jaffee 212 lennart lehmhaus

Through reception of and induced by the theoretical and methodological watershed in French and Anglo-Saxon literary studies, also the study of biblical and rabbinic texts experienced its “literary turn.” Beginning with some pioneering studies by Arnold Goldberg, Isaac Heinemann and Yonah Fraenkel, rabbinic traditions were now increasingly understood as fine pieces of literary art. Thus, the research preferred to survey rabbinic text, primarily focusing on their stylistic, composite or general literary features. In the following decades, narrative elements like exempla, proverbs, parables or riddles have been the subject of intensive form-critical and rhetoric examinations. Other studies chose a comparative approach or addressed the dimension of intertextual relations between singular teachings and/or whole documents or text-traditions. This literary approach contributed immensely to the understanding of the textual fabric and structure of rabbinic works. However, its self-imposed restraints sometimes limited its own range of significance, when scholars assumed that absolutely no insights about the extra-textual world or the context (Sitz im Leben) can be discerned from closed, fictional texts.2 This gap between text and context has been more recently studied by applying a careful and more balanced approach which navigates between ‘literary’ and ‘historical’ questions. Thus, rabbinic traditions are understood particularly as parts of a pedagogic discourse in which the authors applied different literary and rhetorical elements in order to convey their ideas. However, from these ideas and the way they are presented, one can learn about the self-image of the rabbis and their perception of the world (Weltanschauung). Furthermore, this self-reflexive reading is augmented by taking into account a text’s socio-religious and ideological dimension, for characteristic features like recurring ideas, specific argumentative patterns or literary strategies may help us to decipher possible functions of

(Cambridge: Cambridge University Press, 2007), 75–96. A very learned answer to the chal lenge by the methods and theories of literary studies and a reassessment of talmudic historiography provides Isaiah M. Gafni, “Rethinking Talmudic History: The Challenge of Literary and Redaction Criticism,” Jewish History 25 (2011): 355–375. 2 While the analysis of the literary structure and purpose constituted a big leap for ward, the above mentioned dilemma is most evident in Arnold Goldberg’s form-critical approach of segirut (‘clo (ז"ל) and solely descriptive studies and also in Yonah Fraenkel’s sure’) applied to rabbinic aggadah. For a discussion of those two important scholars, see Catherine Hezser, “Classical Rabbinic Literature,” in The Oxford Handbook of Jewish Studies, ed. Martin Goodman et al. (Oxford: Oxford University Press, 2002), 115–140; Hillel Newman, “Closing the Circle: Yonah Fraenkel, the Talmudic Story, and Rabbinic History,” in How Should Rabbinic Literature Be Read in the Modern World?, ed. Matthew Krauss (Atlanta, GA: Gorgias Press, 2006), 105–135. between tradition and innovation 213 those traditions in their socio-cultural context.3 This approach has been aptly summarized by Joshua Levinson who describes rabbinic literature neither as a separate and separable aesthetic realm nor as a mere product of culture, but as one realm among many for the negotiation and production of social meaning, of historical subjects, and of the systems of power and identity that at once enable and constrain those subjects.4 In contrast to the traditions of Mishnah, Tosefta, Talmud and so-called classical midrashim (tannaitic and amoraic), the post-talmudic texts have remained fairly understudied for a long time. It was only recently that these texts received more attention from scholars who analyzed those works, using pre-eminently literary theories and methodologies. Still, virtually all of these studies of early medieval, late midrash were mainly focusing on one aspect of those traditions: time and again, scholars emphasized the ‘narrativity’ of later midrashim. It has now become a common assumption that the later midrashim mark a considerable shift in rabbinic literature from the classical exegetical and dialectic focus to a distinct narrative orientation.5 Thus, some later midrashim are subsumed under the label of “Rewritten Bible,” although this is but one of many facets of those texts.6 Other traditions, in which the story seems to be the prevailing literary element, have been compared to earlier midrashic material as well as to narrative traditions from surrounding cultures. These studies focused on

3 See the concise discussion of a helpful combination of several approaches and meth odologies in Carol Bakhos, “Method(ological) Matters in the Study of Midrash,” in Current Trends in the Study of Midrash, ed. Carol Bakhos (Leiden: Brill, 2006), 161–187. Richard L. Kalmin provides a study of how extra-textual contexts can be discerned (“The Use of Midrash for Social History,” in Current Trends in the Study of Midrash, 133–160). Another approach concentrating on shared symbols and cultural semiotics is offered by Galit Hasan-Rokem, Tales of the Neighborhood: Jewish Narrative Dialogues in Late Antiquity (Lon don: University of California Press, 2003). 4 Joshua Levinson, “Literary Approaches to Midrash,” in Current Trends in the Study of Midrash, 189–226, especially 204. 5 See for example the chapter heading in Günter Stemberger, Einleitung in Talmud und Midrasch (München: Beck, 20119), 363–378. The first category of other aggadic works is delin eated here as “from Midrash to narrative literature” (“Vom Midrasch zur Erzählliteratur”). 6 In Stemberger, Einleitung, 365–369, we find only two works that are explicitly described as “Rewritten Bible” (i.e. PRE and Divre ha-yamim shel Moshe). Also Joseph Dan, Ha-sippur ha-ivri bi-yeme ha-benayim (Jerusalem: Keter Publishing House, 1974), 134, described PRE as belonging to this category. However, such an attribution with regard to PRE is highly questionable, as noticed by several scholars. See Dagmar Börner-Klein, Pirke de-Rabbi Elieser: Nach der Edition Venedig 1544 unter Berücksichtigung der Edition Warschau 1852 (Berlin/New York: de Gruyter, 2004); Steven Daniel Sacks, Midrash and Multiplicity: Pirke de-Rabbi Eliezer and the Renewal of Rabbinic Interpretive Culture (Berlin: de Gruyter, 2009), 17–22. 214 lennart lehmhaus the new narrative mode of structuring texts, which includes strategies of reworking motifs and expanding stories. Even for some smaller forms (like exempla, parables, etc.), such a shift has been described and scholars suggested a turn towards a new dominant ‘novelistic’ paradigm in medieval times.7 Other studies attest for a mere collector mentality on the side of the authors/editors of later midrashim who were driven by their anthologi- cal and eclectic interest in former traditions. Thus, the later works are conceived as merely editing a mosaic-like anthology, consisting of building blocks from already existent material. This entails a process of selection of certain passages or teachings from earlier rabbinic texts, which have been then integrated into a new corpus by using editorial markers and similar structural devices.8 According to some studies, this even holds true for pre-rabbinic ideas and motifs (like mythic material), which were re-introduced and expanded in later midrashic texts. Based on these assumptions, scholars asserted that the later midrashim exhibit little to no ideological or theological innovation. Thus, their contribution is seen not in their content but in their literary style and form, which changed profoundly into the direction of narrative.9

7 See especially Eli Yassif, The Tales of Ben Sira in the Middle-Ages: A Critical Text and Literary Studies (Jerusalem: Magnes, 1984) [Hebrew], who compares the literary art of this text with narrative traditions from the surrounding Indian and Arab cultures, and explores the relationship to earlier Jewish texts. Also Anat Shapira stresses the narrative expan sion of former rabbinic motifs and story-patterns. However, in her edition and analysis of Midrash Asseret ha-Dibrot she ascribes a two-faced character to the text, which oscillates between exposition in the style of classical midrash and a collection of stories illustrating the importance of the Ten Commandments. See Anat Shapira, Midrash asseret ha-dibrot: Midrash on the Ten Commandments Text, Sources and Interpretation (Jerusalem: Bialik Institute, 2005). Eli Yassif suggests that in the medieval period stories and collection or cycles of narratives became increasingly popular in Jewish literature (Sippur ha-am ha-ivri: Toldotaṿ, sugaṿ u-mashmaʻuto [Jerusalem: Bialik Institute, 1994], 232–399). The observa tion of a novelistic expansion is shared by David Stern, Parables in Midrash: Narrative and Exegesis in Rabbinic Literature (Cambridge, MA: Harvard University Press, 1991), with special regard to the nature of the rabbinic mashal. 8 Although Elbaum includes the more editorial approach also in the title of his impor tant study on late midrash, one has to acknowledge that in the body of the article he stresses the more innovative features that pertain to our discussion. Cf. Jacob Elbaum, “Between Redaction and Rewriting: On the Character of the Late Midrashic Literature,” in The Ninth World Congress of Jewish Studies, Vol. 3 (Jerusalem: Magnes Press, 1985), 57–62 [Hebrew]. 9 Jeffrey L. Rubenstein states regarding the non-innovative character of later traditions: “The Tanhuma and Pirqe de-Rabbi Eli’ezer do not reflect new ideas as much as a new style. At the same time, the narrative style and freedom from rabbinic midrashic form does allow for creativity and more vivid and daring mythic expressions” (“From Mythic Motifs between tradition and innovation 215

In conclusion, the late midrashic traditions, analyzed against the foil of their precursors, were all too frequent only perceived as a mere afterglow of the Golden Age of classic (i.e., amoraic) midrash. All these above mentioned developments are taken by some scholars as evidence for a process of decline in post-talmudic rabbinic literature. From a comparative perspective these studies attributed to them eclecticism as a governing principle which attests for the stagnation of rabbinic literature and culture.10 While the turn to the narrative model is a crucial aspect of nearly all later midrashim, other literary strategies, compositional dimensions, and thematic developments should not escape our notice.11 The insights of to Sustained Myth: The Revision of Rabbinic Traditions in Medieval Midrashim,” HTR 89 (1996): 131–159 [158]). described a decline of the literary art in later agga (ז"ל) Already Yonah Fraenkel 10 dic traditions (Sippur ha-Aggada: Achdut shel tokhen ve-tzurah [Tel Aviv: Hakibutz Hameuchad, 2001], 51–74). Rinat Drory, argues that in gaonic time the rabbinic (or in her diction “Rabbanite”) literature was the established and dominant factor with a long tradi tion in the Jewish literary system (Models and contacts: Arabic Literature and Its Impact on Medieval Jewish Culture [Leiden: Brill, 2000], 126–157). However, the reliance on tra dition together with a fixed and normative shape navigated the rabbinic literary modes of expression into a situation of “impasse.” This deadlock was resolved then by Karaite writers, who—coming from the margins—adopted and adapted Arabic literary models in order to create a fresh, non-Rabbanite literary system. Drory’s poly-system-model which compares all Jewish (Rabbanite and Karaite) and Arabic textual production, has been a promising, theoretical and methodological improvement. However, her discussion of the relation between Palestine and Babylonia, and the sharp dichotomy drawn between old Rabbanite and new Karaite literary models relied heavily on older research. Thus, while her overall findings are still very compelling, several points call for qualification in light of more recent studies of the gaonic era as well as the interaction between rabbinic and Karaite Jews. For a concise survey, see Meira Polliack, ed., Karaite Judaism: A Guide to Its History and Literary Sources (Leiden: Brill, 2003); Marina Rustow, Heresy and the Politics of Community: The Jews of the Fatimid Caliphate (Ithaca: Cornell University Press, 2008). A fresh approach to talmudic Judaism and its gaonic formative period is offered by Talya Fishman, Becoming the People of the Talmud: Oral Torah as Written Tradition in Medieval Jewish Cultures (Philadelphia: University of Pennsylvania Press, 2011). With regard to the parable, David Stern describes a gradual decline from the heyday of the classical, exegeti cal mashal in the amoraic period. According to his study, this decline is attested by a nov elization of the rabbinic parable which is unable to attain the perfection of its prototypes. See Stern, Parables, 211–233. 11 See Jacob Elbaum, “Between Redaction and Rewriting.” As the main differences and characteristics of the later midrashim (Tanhuma, PRE/ SE), he mentions: a) use of material from amoraic midrashim (GenR/ LevR/ PRK), b) a partly use of material from the Babylonian Talmud, c) the absence of mixed language (Hebrew-Aramaic and Greek loanwords) and a shift to a “clear Hebrew”, d) a blurring of the exegetical and homiletical approach, paired with a new topical and compositional order, e) an affin ity to rhetorical terminology and a poetic or flourish style (similar to gaonic melitzah), f) a simultaneous omission of rabbinic names and “false” or creative use of such attributions, g) a stylistic and contential convergence to medieval forms of exegesis and literature. 216 lennart lehmhaus philological, linguistic, intertextual, and discursive analysis may help to decipher and to define more subtle literary dimensions as displayed in those rabbinic traditions. Based on a thoughtful analysis of these aspects, one can try to reflect about the motivation for new stylistic and structural strategies and their function within the discourse of each respective work. Recently, a number of scholars started analyzing some later midrashim (especially SE and PRE) more thoroughly in order to grasp the complex interplay between the religious content of their discourse and the literary forms and functions applied. The later traditions are then not at all conceived as texts of decline. In fact, they are rather described as important forerunners that attest for the transition of rabbinic culture from collective works to individually authored texts.12 By analyzing them, one might be able to follow the trajectories from late antique to early and later medieval Jewish culture. In the remainder of this article, I would like to illustrate some of these new features in one specific tradition, while keeping a comparative eye on other later midrashim. I will explore stylistic, compositional, and intertextual dimensions as well as adaptive and innovative literary strategies. Our discussion will focus on several examples from Seder Eliyahu Zuta (SEZ- the minor order of Elijah). Together with its cognate text Seder Eli- yahu Rabba (SER), this work forms a puzzling and fascinating tradition of hybrid character. At first glance, it deals primarily with questions of ethical lifestyle and righteous conduct for all kinds of Jews. However, a second glance reveals that these teachings are intertextually interwoven with biblical and rabbinic traditions. These points of contact with other traditions, provided by and alluded to in the text, could be explored by a learned audience of religious experts that the rabbis certainly were. The text integrates ethical and religious concepts in theory and practice with a complex discourse on diversity and identity in Jewish and/or rabbinic culture of late antiquity and early medieval times. These topics are addressed skillfully through an artistic composition and combination of different genres, literary structures, and strategies.

12 Cf. Dina Stein, “Pirqe de-Rabbi Eliezer and Seder Eliyahu: Preliminary Remarks on Poetics and Fictional Space in late Midrash,” JSHL 24 (2011): 73–92 [Hebrew]; Moshe Lavee, “Seder Eliyahu,” in Encyclopedia of Jews in the Islamic World, ed. Norman A. Stillman (Leiden: Brill, 2012). between tradition and innovation 217

The question of date and origin was almost the only focus of scholarly interest and has been disputed for more than a century.13 Still, dozens of studies could not help to solve these questions due to an uncritical trust in the historic reliability of the text and a naïve usage of realia in order to pinpoint some exact dating.14 Yet, more recent studies about the literary character and the thematic concerns point to a post-talmudic date in early medieval time, which is also known in Jewish history as the gaonic period (seventh to eleventh century ce).15

13 The suggestions for place and date of origin range between third-century Palestine, Babylonia in the fifth or ninth century, and eleventh century Italy. The absence of attribu tions to rabbinic sages, confusion of two titles (Seder Eliyahu / Tanna devei Eliyahu) as well as a puzzling transmission history complicates the picture. A concise summary of the early Hebrew scholarship is given in William G. Braude and Israel J. Kapstein, Tanna debe Eliyyahu: The Lore of the School of Elijah (Philadelphia: Jewish Publication Society, 1981), 3–15, also cf. Leopold Zunz, Die gottesdienstlichen Vorträge der Juden historisch entwickelt (Frankfurt am Main: J. Kaufmann, 1892), 103–125. 14 Scholars of Jewish literature and history have contested premature conclusion about the historic reliability of rabbinic traditions that were first and foremost substantial ele ments of a religious and pedagogical discourse. For an in-depth discussion of the prob lem of historicity, see the contributions of Zeev Safrai, Jacob Neusner, Richard Kalmin, Günter Stemberger and others in Judaism in Late Antiquity, eds. Jacob Neusner and Alan J. Avery-Peck, (Part 3; Leiden: Brill, 1999), 123–232. I have to mention a bold exception from these historical approaches. In his work Max Kadushin tried to give a concise survey of the main religious ideas featured in this tradition (The Theology of Seder Eliyahu: A Study in Organic Thinking [New York: Bloch, 1932]). However, some of his assertions call for clarification and qualification. First, Kadushin does not differentiate between SER and SEZ. His thoughts about the theology of the text do neither pay sufficient attention to the text’s redactional history nor do they relate to the periodization of interrelated rab binic works. Second, Kadushin developed an appealing idea of a rabbinic organic thinking which includes four core concepts (God’s loving kindness, his justice, Torah and Israel). However, it is precisely this concept which tends to predetermine his close readings and constrains them to those passages that prove his analytical presuppositions. 15 Only a few studies addressed questions of language, literary style, composition and thematic concerns in a more detailed and critical way. Their findings as well as my own, with special regard to the text’s close relation to the Babylonian Talmud, Jewish ethical Wisdom traditions (Avoth / ARN / Derekh Eretz / Kallah) and the later midrashim (PRE / KohR / MidrPs / MidrPrv / PesR), point rather to a post-talmudic date, in early gaonic time (post-7th century ce). A possible terminus ante quem is given by the reference to structure and content of the two traditions in the writings of Natronai Gaon (ca. 860 ce). See Ephraim E. Urbach, “The Question of Language and Sources of Seder Eliyahu,” Leshonenu 21 (1956): 183–197 [Hebrew]; Jacob Elbaum, “Tanna devei Eliyahu,” EJ XV, 803f; idem, “Between Midrash and Mussar Literature: A Study of the Chapters 1–6 in Tanna debe Eliyahu,” Mehkarei Yerushalayim be-Sifrut Ivrit 1 (1981): 144–54 [Hebrew]; Günter Stemberger, Einleitung in Talmud und Midrasch (München: Beck, 19928), 332f. The literary and intertextual evidence in SEZ for a post-talmudic date is further elaborated in my dis sertation (“ ‘Derekh Eretz im Torah’—Seder Eliyahu Zuta as universal, religious ethics for rabbinic and non-rabbinic Jews.” [PhD diss., Martin-Luther-Universität Halle-Wittenberg, 2012/2013]) against the backdrop of the text’s discursive strategies as expressions of its socio-cultural and literary embedment. 218 lennart lehmhaus

Some General Outlines of SEZ’s Literary Character

Before delving into detailed analysis of some of the most innovative literary devices, I would like to discuss more briefly some innovative features and patterns shared by SE and other late midrashic traditions. The following brief survey intends not to be comprehensive: it can neither delineate the whole of SEZ’s literary structures nor is the discussion of each relevant feature exhaustive. The following section will rather concentrate on the most important characteristics and highlights their peculiar function for the discourse.

Topical Arrangement Already Jacob Elbaum in his preliminary study of late midrash pointed to a shift in the literary structure. The important difference in late midrash lies in a superordinate thematic arrangement of the text which diverges significantly from the two major principles of order in earlier midrash. The first guiding principle is realized in several midrashim that display a rather linear reading and exposition of a biblical text. The second principle pertains to a division of midrashic texts which is dependent on external factors, like following elements of the liturgical order, the year cycle of holidays, or the sequential reading of Torah. Often these texts are seen as homiletic in nature in contrast to the exegetical focus of the first group.16 By contrast, the compositional usage of thematic chapters, patterns, or clusters can be identified as the major innovative feature of some later traditions (like PRE and other minor midrashim).17 Similar structural

16 The first principle can be found in most of the halakhic and aggadic midrashim (like MekhY, Sifre and Sifra; GenR, LamR, MidrProv, MidrTeh). Other midrashic works tend to follow the second structural model (like PRK; LevR; PesR; NumR; ExR, etc.). Still, one should keep in mind that these labels cannot be more than heuristic tools which allow scholars for an approximate classification and characterization. In practice, this division is not strict at all. Midrashic traditions can contain elements of both major groups. This is all too evident when it comes to the label “exegetical Midrash,” since exegesis and exposition on a micro-level is one key pattern in almost every Midrash. The same holds true for the difference of halakhic and aggadic texts, for aggadah often deals with religious issues that are equally important with respect to halakhah. 17 Here we can mention texts like Midrash Asseret ha-Dibrot, Midrash Elle Ezkera, Midrash Temura, Midrash Ma’asse Torah; Pereq Shalom, Pereq ha-Shirah, and also some lay ers of the complex Tanhumah-traditions. These thematic approaches resemble partly the structure of sugiya-units in the Babylonian Talmud, which might be seen as an antecedent model for topological arrangement. However, the sugya is intertextually dependent, since it is structured around the topic of a certain Mishnah and its style is mostly dialectical in presenting a polyvocal discussion between several sages and/or traditions. We have also to between tradition and innovation 219

features figure also prominently in the so-called rabbinic ethical or Wisdom-traditions, which show in some parts a close proximity to SE.18 In SER and SEZ, the thematic approach of the texts is most obvious. Both texts consist of larger and smaller portions of teachings which deal with one specific issue or several interconnected subjects. These discussions follow at times the structure of the chapters. In other instances, the topical arrangement is realized in dense textual units. While such a thematic cluster can stand alone, in the context of SE they are artistically linked to each other. The text is structured by keyterms and trigger words that intertwine its main concepts and topics like a hidden thread. Thus, those connectors enable and prompt the audience to follow Seder Eliyahu’s religious and ethical discourse.

The Usage of Smaller Forms for New Thematic Purposes An additional effect of the increased orientation towards thematic patterns can be found also in the different application of several genres. Many literary forms exhibit a shifting focus of interest that is induced by new rhetorical goals and strategies. In Seder Eliyahu one can observe this development by looking at the parables (mashal, plur. meshalim) inter- spersed throughout the whole text. David Stern described the mashal in Seder Eliyahu—though his analysis is confined to SER alone—as already deteriorated due to its deviation from a classical model of exegetical parable in earlier midrashim. He suggests that the late parables restrict themselves to a lively narrative expansion and a quasi-novelistic retelling, while lacking the exegetical brilliance and complexity of their precursors.19 The argument that in later midrash the mashal serves merely to illustrate teachings in order to present them in a more appealing, novelistic outlook, leads us astray. For also in the classic, exegetical midrashim, the

mention the amoraic midrash of Leviticus Rabbah which, according to the view of some scholars, also tends more towards a thematic order than to a exegetical mosaic. Burton Visotzky even suggests that this work, because of its discursive unity, resembles a book- like type of literature which prevailed in its Hellenistic-Roman and Byzantine surround ings. See Burton Visotzky, “On Some of the Redactional Principles in Leviticus Rabbah,” in Higgayyon le-Yonah: New Aspects in the Study of Midrash, Aggadah and Piyut: In Honor of Professor Yona Fraenkel on the occasion of his 75th birthday, ed. Jehoshua Levinson et al. (Jerusalem: Magnes, 2006), 333–345 (Heb.). 18 Those thematic structures are evident in works like Pirke Avot, Avot de-Rabbi Nathan (ARN), Massekhet Derekh Eretz Rabba (DER) and Zuta (DEZ), Massekhet Kalla und Kalla Rabbati (KallaR). 19 Cf. Stern, Parables, 211–216. 220 lennart lehmhaus mashal illustrates and enacts the exegetical steps taken by the rabbinic interpreter through narration, as Stern himself has demonstrated. The illustrative function is clearly beyond dispute. However, we can find additional features acquired by the mashal in Seder Eliyahu Zuta. First of all, in this tradition a major shift from exegetical to a strong rhetorical function takes place. The mashal proper, in SE as in other traditions, is narrative by nature and as such, it is at times expanded into a comprehensive story including characterization, ornamentation, and dialogues. In some instances, it straightforwardly serves to reinforce a preceding teaching or to explain another narration. Yet, often we can distinctly observe how the parables (re-)construct meaning and convey more complex teachings in another way. The eschewal of a clear reference to a biblical pre-text is one major change in the outlook as well as in the function of the mashal. While the classic parables convey their own exegesis through participation of the audience, we can find in SEZ similar explanatory purposes, albeit on another level of discourse—not regarding biblical verses but rather with a focus on the text’s own teachings. This makes the mashal one of the most important discursive elements with a primary bridging function. Instead of merely serving as a novelistic illustration in the guise of a fairy-tale, in SE the mashal is utilized as a powerful rhetoric tool to impart its core ideas.

First-Person Narratives of Encounter and Dialogue A major narratological innovation is the invention of a first-person narrator character in several episodes that are part and parcel of SE’s literary structure.20 All such narratives consist of instructive dialogues between an anonymous, probably rabbinic, first-person narrator and a great variety of non-rabbinic questioners. Sometimes, we find very short question-and- answer episodes, serving as a trigger for the narrator to elaborate on a specific topic. Other dialogues are far more complex and comprise different literary forms (lists, midrashic explanations, parables, etc.) that are

20 Passages told from the first-person perspective, without attribution to a sage, are scarcely to be found elsewhere in rabbinic literature. Furthermore, about one-sixth of all pa‘am ahat passages (22 passages) and one half of all first-person narratives (18 of 40) in post-tannaitic texts occur in SER and SEZ. The anonymous narrator is in SEZ identified as rabbinic or learned due to his teachings and the way he is addressed by his opponents. For the reader, the narrator remains only slightly overt since the perspective (Genette’s “focalization”) of the narration is internal and not made explicit through commentary or retelling of his personal, biographic details. Cf. H. Porter Abbott, The Cambridge Introduc tion to Narrative (Cambridge: Cambridge University Press, 2002), esp. 62–76. between tradition and innovation 221 utilized to discuss broader theological and ethical issues. In SEZ, these narratives of encounter are framed within a setting of a non-urban, rural space of travel and passage.21 A typical introduction to such a narrative reads as follows: “Once22 I was travelling from place to place, when a man23 accosted me and greeted me. But he did not know me.”24 Other episodes depict encounters with a Persian priest (SER 1) or interaction with some students (SER 15 and 16), and even conversations with the Sages in the academy of Jerusalem (SER 10). The first person-perspective turns the story into a personal experience and lends the message a greater authenticity and thus authority. At the same time, the dialogues appear as vivid conversations and actions (resembling a dramatic performance on stage) with paradigmatic and exemplary features.25 Elsewhere I have argued that in SEZ those narratives as instructive dialogues promote core features of a ‘minimal Judaism’ open to all: basic

21 In this universal narrative-type (of wandering or journey), the encounter with the other provides an open but still framed sphere for contest and examination of one’s own cultural or ideological identity. In rabbinic literature, often a blurring of the boundaries of Jewish and surrounding cultures (typical areas are the bathhouse, the marketplace or the house of the matrona) can be observed. Sages encounter typological non-Jews (priest/ emperor/philosophers/officials) with whom they discuss complicated and sometimes even delicate religious issues. Those topics seem to be crucial for the rabbis’ identity and their cultural self-definition. 22 For the Hebrew text, see the edition by Meir Friedman (Ish Shalom), Seder Eliahu rabba und Seder Eliahu zuta (Tanna d’be Eliahu) (Wien: Verlag der Israelitisch-Theologischen Lehranstalt, 1902). The English version is based on Tanna debe Eliyyahu: The Lore of the School of Elijah, trans. William G. Braude and Israel J. Kapstein (Philadelphia: Jewish Publication Society, 1981), with slight changes of my own. Textual variants of other manuscripts of Seder Eliyahu Zuta (SEZ) are given in the footnotes. 23 While MS Parma 2785 refers simply to “a man”, the text of SEZ 1 in MS Vatican ebr. 31 ,which should be read as quaes(i)tor. Braude/ Kapstein, Tanna debe Eliyyahu ,כשדור has and occurring קסדור translate: “magistrate.” The term (usually written in Hebrew as ,360 12 times in rabbinic tradition, mainly to label a Roman police investigator) denotes the lowest office of the higher Roman administrative career, responsible for financial and juridical matters or police investigations. But it could be literally understood as “someone who seeks/asks for knowledge”, as the narrative presents him to us. 24 Other such narratives of wandering and encounter which include a dialogue with a non-rabbinic opponent can be found in SEZ 2, 171 and SEZ 14, 195–196. 25 These dramatic dialogues feature the mimetic narrative mode of showing (the char acters acting as if they were on a scene) instead of an indirect retelling of actions. This intensifies the authentic impression of the conversation. Simultaneously, it blurs the fic tional and composite aspects for the reader or the audience. Unfortunately, we know little if nothing about the performative dimensions (gesture, intonation, etc.) in effect if such a story was read, told or performed for a (wider) audience. However, one can assume that the effect of authenticity of a first-hand report of was amplified in a performance. See Steven D. Fraade, “Literary Composition and Oral Performance in Early Midrashim,” OT 14 (1999): 33–51; Gary G. Porton, “Rabbinic Midrash: Public or Private,” RRJ 5 (2002): 141–169. 222 lennart lehmhaus individual knowledge of Scripture, the most important prayers and benedictions as well as moral behavior and pious mind set. Most of these dialogues are non-confrontational in nature. The contact with non-rabbinic others is not used to demonstrate the supremacy of a learned class. Instead of behaving arrogantly or launching a sharp backlash against the opponents, the narrator demonstrates patience and indulgence in his interaction. He does not rebuke his challengers or gives them stern advice, but rather he argues, explains, and tries to convince them of the positive effect of the values (‘minimal Judaism’) promoted by him. The close and affirmative interaction with different non-rabbinic others in SEZ is framed within a setting of liminality and border-crossing. Thus, questions of Jewish and rabbinic identity could be imagined and examined through the safety screen of the narrative.

Author and/or Narrator Character An additional aspect of the above described narratives is the invention of a narrator/author character. A first-person perspective is not only apparent in these instructive dialogues. Indeed, throughout SE, one can find several instances in which an anonymous voice (“I”) is speaking from the text. Sometimes, this voice appears in form of an oath formula (me’id I call heaven“ / מעיד אני עלי שמים וארץ /ani alai shamayim wa-aretz and earth to witness”) that comes to illustrate and to reinforce certain teachings, which are discussed in a preceding paragraph. This formula has its most appearances all over rabbinic texts in SER and SEZ where it functions as a strong, characteristic rhetorical tool.26 In other instances, this narrator/ author voice seems to turn to God by uttering blessings, which are widely scattered throughout SE. They serve as an intercessory utterance, linking the praise of God to a depiction of his love and mercy towards Israel. Moreover, these benedictions function, in a way similar to the mashal, as connectors or bridging devices within the discourse of SEZ.

26 This function of the formula is most obvious in b. Jeb 16a, which has a parallel in SEZ 1, 169. Other instances can be found in SEZ 3, 175, and SEZ 15, 197. Cf. SER 5, 26; SER 18, 91, and PSEZ 18, 24f. While in SE this oath-formula is uttered by an anonymous voice; in most of the other rabbinic traditions, it occurs as part of the speech of a rabbinic sage or in dialogues between biblical characters. For examples of character speech, see MidrPs 137,2 (Jeremiah) or MidrProv 16,10 (Salomon). Throughout the rabbinic literature, includ ing medieval collections like Leqah Tov or Yalqut Shim’oni, we find 46 instances of this formula. In SER and SEZ we find exactly the same number of appearances as in both Talmudim and all midrashim together (13). This comparison underscores the rhetorical importance of this device in SE’s discourse. between tradition and innovation 223

Furthermore, the voice also addresses the reader or audience directly.27 However, there is no clear indicator to assume that the first-person narrator in the narratives and the voice(s) in different other parts of the text are always identical. The appearance of a first-person perspective is in itself, as mentioned above, a rare and extraordinary phenomenon. If a tradition provides such a voice speaking from the text, this always contributes to the way in which it is understood. By providing some vivid, personal speech, this form lends some authenticity (personal experience) and probably also greater authority to the teachings conveyed in the work. In some manuscripts of SEZ, we find a clear and recurring reference to a certain Eliyahu as the narra- (משום דבי אליהו) ”tor of the dialogues and to a “school/ house of Eliyahu as the source of certain teachings.28 This can probably be understood as a reference or an allusion to the biblical prophet Elijah himself. Such an understanding is underscored by the explicit reference of other manuscripts and the printed edition that refer in some passages to Eliyahu ha- navi (“the prophet Elijah”). This kind of attributed authorship as well as the perception of the prophet as its narrator-character in the course of its transmission would have given importance and a high degree of authority to the text. Thus, when text alludes to the prophet Elijah as its protagonist, those narratives could be conceived as more deeply entrenched in biblical and rabbinic tradition.29 For, in those traditions the prophet Elijah

27 For an example of the narrator’s voice addressing God, see SER (10) 11, 53; SER 17, 83. In SER 17, 86, this voice turns to the audience (“O my brothers and my people, listen to me!”). 28 This reference occurs three times in SEZ 1, while SEZ 15 (197) refers to a certain “Abba Eliyahu”. Regarding the latter, see bSanh113a-b; PesR 32; KallaR 4,31. A reference to a “school of Eliyahu” can also be found in the talmudic passages introduced with “tanna debe Eliyahu”, cf. bPes. 94b and 112a; bMeg. 28b; bQidd. 80b; bSanh. 92a and 97a; bAZ 5b and 9a; bTam. 32b; bNid. 73a. This introductory formula occurs also in some later traditions (MidrKoh Zuta; Pitron Torah; MidrPs). 29 Already in biblical traditions, Elijah, who never died but was taken away in a fiery whirlwind, is mentioned as a herald of messianic redemption. This role has been further elaborated in rabbinic traditions which also portray Elijah as revealing himself to excellent rabbinic scholars in order to solve exegetical problems or to offer other religious advices. In fact, the legend in bKet 106a draws a direct connection between Elijah and the origin of a “Seder Eliyahu” which, according to this tradition, the prophet taught to Rav Anan. Although it seems rather unlikely that the Talmud provides a reliable report of the emer gence of our text, the legend could have served as the source of origin in another way. I suggest that the author(s) of Seder Eliyahu were familiar with this legend in bKet 106a, which served them as a hook in rabbinic tradition to which the text could be attached and from which it gained basic authority. Still, Jewish tradition granted the biblical prophet also functions that made him more accessible for common and unlearned Jews, either. Thus, the prophet was given a seat of honor in the ceremony of circumcision (brit milah). Cf. PRE 29. Moreover, this attribution to Elijah might have had further socio-religious and 224 lennart lehmhaus figures prominently as a messianic herald as well as a mediator between the human and divine realm. This amplifies the extraordinary importance ,throughout the text. Moreover (תשובה) given to the idea of repentance the dialogues resemble also the popular rabbinic sub-genre of “revelations in which the prophet appears to rabbis and other (גילוי אליהו) ”of Elijah Jews in order to spur them on to study, or to reveal knowledge and to clarify halakhic problems and religious concepts.30 Finally, the allusion to Elijah, who is rabbinically depicted as a role model or requires moral behavior, corresponds also with the agenda of the text. The combination of practical ethics (Derekh Eretz) with the appeal to acquire basic religious knowledge occupies center stage of Seder Eliyahu Zuta’s discourse. A similar, also rather pseudepigraphic, author-figure character is constructed in PRE. The opening chapter presents a kind of initiation- or mat- uration-story (“coming of wisdom”) about Eliezer ben Hyrcanos. It recounts how he suffered hardships before becoming a student of Torah in Jerusalem. Finally, this career is crowned with success when R. Eliezer is vaunted an unsurpassed sage of his generation. His greatness is recognized even by his father who intentionally came to disinherit his son, but changes his mind given the latter’s supremacy regarding Torah.31 It seems rather question-

discursive functions. We know that in (proto-)Karaite circles Elijah the prophet played a major role in their messianic expectation, which included a king messiah and a priestly/ prophetic messiah. Elijah was also considered to be the “teacher of righteousness” (more tzedek), who will lead the people from the wilderness of nations to repentance and final redemption. This concept, which is partly mirrored in rabbinic tradition, could have served as a discursive connector between rabbinic and (proto-/ pre-)Karaite recipients of Seder Eliyahu’s teachings concerning repentance and moral behavior. Cf. Yoram Erder, “The Mourners of Zion: The Karaites in Jerusalem in the Tenth and Eleventh Centuries,” in Karaite Judaism: A Guide to Its History and Literary Sources, ed. Meira Polliack (Leiden: Brill, 2003), 213–235, esp. 219f. 30 Thus, we can assume a certain similarity between the didactic impetus and the posi tive attitude of the narrator-character in SEZ and the prophet Elijah in the gilui-type sto ries. Thus, SEZ’s narrator resembles the basic characteristics of the prophet as described by Kristen H. Lindbeck, Elijah and the Rabbis. Story and Theology, New York: Columbia Uni versity Press, 2010, XIX: “Elijah thus appears as a teacher, an immortal Sage who is friend to the sages. [. . .] Elijah’s words and actions are always positive.” And ibid., 67–69: “[. . .] Elijah is an ideal teacher for the Rabbis. [. . .] he often acts as an ideal teacher, displaying patience and dispensing encouragement [. . .]. “For Elijah’s role as a motivator (as in PRE 1/ bNed 50a / bTa’an 22a) or a moral teacher of respect and solidarity between the sages and the unlearned (as in bBB 7a/ bKet 61a and 105b/ bMak 11a) see ibid., 62–72, 96–105, 116–135. 31 Cf. Dina Stein, Maxims, Magic, Myth: A Folkloristic Perspective on Pirkei de-Rabbi Eliezer (Jerusalem: Magnes, 2004), 115–168 [Hebrew]. A similar construction of a narra tor/ author-character can be found in the medieval stories of Ben Sira (Alphabeta de-Ben Sira). The initial story is a “birth of a hero”-type narrative that recounts Ben Sira’s miracu lous pedigree and introduces him as an extraordinary kind of Wunderkind. The instructive dialogues with a teacher of young children reveals more facets of his superior knowledge, although due to his lack of practical experience, he seems often more witty than wise. between tradition and innovation 225 able if the title of the work should suggest that R. Eliezer has to be seen as its pseudepigraphic author/redactor. However, several scholars pointed to the fact that the majority of attributions to Rabbi Eliezer together with the two opening chapters about his initiation as a sage constitute a strong element of anchoring the work in the rabbinic tradition and thus providing for its authority.32

Framing Devices: Introduction and Conclusion Another innovative and interesting aspect in SER and SEZ is their use of framing devices. Both texts employ chapters, clusters or passages that function as expositional or concluding structures. In his preliminary study, Elbaum pointed to some of the governing structural strategies in SER. Of special interest for our discussion is his remark regarding the opening and concluding chapters which build a frame for the whole narrative that ranges from creation and the Gan Eden to the messianic age and Israel’s redemption.33 Thus, SEZ, chapter 1 introduces central topics, motifs, and keyterms that recur time and again throughout SEZ’s discourse. For example, by the text ,צער and תורה ,צדקה ,צדיק means of the keyterms or trigger words explores the interplay of charity, righteousness, Torah-study, solidarity, and the role of the righteous in society in times of distress. Most of these points are reprised as quasi-doublets in the last chapter, SEZ 15, where they are complemented by a hopeful, messianic prospect in the guise of clas- Similar tendencies can be observed for .(חתמה) sic midrashic conclusions example in PRE, where the first two chapters do not only report the scholarly rise of Eliezer ben Hyrcanos against all odds to become an ingenious sage of his days. In fact, as Dina Stein has pointed out, these passages, together with the third chapter, serve as a meta-poetical exposition in

Only in his increasingly confrontational dialogue with the arch-enemy of the Jews, namely Nebuchadnezzar, Ben Sira succeeds to turn his coming-of-age into a “coming of wisdom” as well. See Dagmar Börner-Klein, Das Alphabet des Ben Sira, Wiesbaden: marix, 2007; and Lennart Lehmhaus, “” 32 See ibidem. This point was emphasized by Katharina Keim in her paper “The Func tion of the Rabbinic Attributions in the Pirke deRabbi Eliezer,” which was presented at the International Meeting of the Society of Biblical Literature, held from July 22nd–26th 2012 in Amsterdam. Cf. also the critique by Sacks, Midrash and Multiplicity, 53–59. 33 Cf. Jacob Elbaum, “Between Midrash and Ethical Literature: Studies in the Chapter 1–6 of Tanna debe Eliyahu,” JSHL 1 (1981): 144–154 [Hebrew]. Elbaum refers to SER which frames its ethical-religious discourse and forge a bridge from creation and the introduction of the conditio humana as the loss of paradise (SER 1) to a future messianic redemption (SER 31), to which mankind is guided when they follow the path of prayer, study and especially moral conduct. 226 lennart lehmhaus which the hermeneutic agenda of the text and its author(s) is elaborated. She assumes that these chapters introduce the Torah as the principal tool for God’s creation as well as an ultimate device to understand through the world and its creator.34 (דרשה) exposition These framing structures can be understood as forerunners of program- matic agendas, introductory elaborations and systematic conclusions which were to become typical of treatises in gaonic and medieval Jewish literature. In ways structurally similar to these monographs, the two midrashim deploy a narrative exposition, which is less structured but highly pro- grammatic in elaborating central topics, key terms, and the hermeneutic or literary agenda. Likewise, we can see a chiastic closure of those texts in which they return to these issues and generally provide eschatological consolation.35

Strategies of Reference and Embedment

Of special interest is the innovative and adaptive use of classical rabbinic terminology in SE. From Urbach’s short treatment of SE, we learn that these traditions demonstrate an overt stylization of language. The texts tend to accumu- lation and classicistic combination of older terminological patterns.36 In the following paragraphs, the text’s interplay with earlier traditions from the Bible or from other rabbinic works shall be the subject of the discussion. We will ask how the author(s) relate to these traditions, how they are adopted or adapted, and finally which new discursive functions they serve in the context of SE.

The Rhetoric of Instructive Formulas This phenomenon can be observed by turning to an exceptional termino- /whence (do we know“ מניין ,logical (re)creation. In general, the question

34 Stein, “Pirqe de-Rabbi Eliezer,” 78–80. 35 Drory, Models and Contacts, 134–138 and 153f, stressed the great divide between all rabbinic midrashic traditions and medieval forms of the hibbur-type (‘composition’). While she described the former as a loosely connected mosaic of teachings and exegesis, the latter type is characterized by compositional and topical unity. Drory connected the hibbur-model with Karaite and medieval authors (i.e., Sa’adya Gaon). However, Dina Stein correctly calls attention to the structural similarities between some later midrashim (like PRE and SE), sharing elements like introductions and conclusions, and those medieval texts of the hibbur-type. Cf. Stein, “Pirqe de-Rabbi Eliezer,” 77. 36 See Urbach, “The Question.” between tradition and innovation 227 learn)?” is applied to support a certain teaching with a fitting biblical ,תדע לך שכן proof text. In a similar manner, rabbinic texts use the phrase “know/learn that it is like that,” followed either by a biblical verse, the repetition of older rabbinic teachings or an illustrating device, like a parable (mashal) or a retold biblical episode. SEZ combines the two and augments מניין תדע לך this phrase in some instances by using the unique formula Whence (do we learn/know)? Know that it is like that; go“ ,שכן צא ולמד out and learn (it) from . . .” This fresh and accumulative combination of exegetical and rhetorical terminology points to a shift in SEZ’s didactic and discursive interests. In contrast to classical rabbinic works, the formula serves in this text to introduce and to amplify own teachings by connecting them with proofs of different kinds. These are not simply biblical verses. Indeed, the text creates complex interconnections between its own central teachings and illustrative or discursive devices like parables, lists or aggadic stories unique to SEZ. This process is obvious in the opening part of SEZ, chapter 4 (178): ,toward every fellow man (עלוב) Be humble :(שנו חכמים) The sages taught and more so, toward the members of your household than toward all others.37 (מנין) ?[And whence [do we know/learn תדע לך שכן צא) You shall know that it is like this. Go out and learn toward his (עניו) from the Holy one, blessed be he, who was forbearing (ולמד people not [only on] one occasion but on two or three [other occasions as .(ולא הלך עמהם בדרכיהם) well]. He does not walk with them in their ways -but sim ,(עונותיהן) Thus, he did not judge them according to their iniquities .(עניו) ply was forbearing And whence [do we know/learn]? You shall know that it is like this :(מנין תדע לך שכן) [From the] Hundred and twenty days from [the day] the perfect Torah 38.(יום הכיפורים) was given to Israel until the Day of Atonement If [in] the first forty days, when Moses ascended to Mount Sinai to bring the ,(ענוה) the Torah for his people, there have not been [God’s] forbearance Torah would not have been given to Israel. They told a parable: [. . .]

37 The whole maxim or parts of it, which exhibit proximity to SEZ 4, can be found in KallaR 3,1; 4,19; DEZ 1,25, and DEZ 2,4. 38 In SEZ, this time-span of 120 days is divided into three parts describing what hap pened in these 40 days. The first period includes the first giving of the Torah, Moses’ absence and the sin of the Golden Calf. The second sequence addresses Moses’ separation from his people and God’s final consolation for him, which makes him return. The last period encompasses the second ascent of Moses’ and Israel’s praying, fasting and weeping, which induces God to promise them well-being and a final redemption. 228 lennart lehmhaus

This short passage illustrates the complex bridging function of the minay- in-formula in SEZ. At the beginning, the rabbinic maxim, marked in the text as a quotation of an older (tannaitic) tradition, stresses the positive effect and reward for humbleness and forbearance practiced towards fellow humans. The formula ties the validity of this moral maxim to the example of God himself who was forbearing to his people Israel when he did not punish them according to their transgressions. This proposi- tion about God’s mercy as dominating his strict justice is then connected via our technical term to the biblical episode of Israel at Mount Sinai. The remainder of the chapter develops a sophisticated aggadic retelling of the story of Torah-revelation, Israel’s sin, and God’s reconcilement. The text employs parables, micro-exegesis, and retold biblical passages as well as exegetical dialogue narratives on Moses, God, and Israel. This passage demonstrates and reinforces SEZ’s main focus on God’s endless mercy and eternal love for Israel who can resort to repentance through studying, praying, fasting, and mourning.39

Quotation Marker as Rhetorical Device and Intertextual Authentication Another process of adoption and adaptation of classical terminology is related to the character of rabbinic literature as a “literature of citation” (Zitatliteratur) by Arnold Goldberg.40 Thus, in a direct manner, teachings from one layer of tradition (e.g., Mishnah, Tosefta, tannaitic midrash) refig- ure and re-appear in later traditions (e.g., Talmudim, amoraic midrash) and are marked explicitly as references through named attributions to rabbinic sages or technical terms indicating the layer of tradition referred to. This implies a strong intra- and intertextual or, if you prefer, intertra- ditional dimension which could be invoked, intensified, or blurred.41 As discussed in the preceding paragraph, attributions to certain sages as well as anonymous quotation-formulae serve as direct markers for interaction between traditions in rabbinic texts. I would like to take a look

39 Cf. SEZ 6, 182–183; SEZ 8, 185–186; SEZ 9, 187–189. 40 Cf. Arnold Goldberg, “Zitat und Citem: Vorschläge für die descriptive Teminologie der Formanalyse rabbinischer Texte,” in Rabbinische Texte als Gegenstand der Auslegung, ed. Peter Schäfer and Margarete Schlüter (Tübingen: Mohr Siebeck, 1999), 96–98. 41 Furthermore, most of rabbinic literature is construed against a web of intertextual threats interwoven with Scripture. These relations are realized through different tech niques of direct and indirect commentary, exposition, gap filling or via parallel narratives or teachings as well as indirect allusions to specific images, settings or motifs. On the intertextual dimensions of rabbinic texts see Daniel Boyarin, Intertextuality and the Read ing of Midrash (Bloomington IN: Indiana University Press, 1990). between tradition and innovation 229

מיכן\מכאן the sages taught,” and“ ,שנו חכמים on the two introductions of From here/hence it is said.” In most rabbinic works, these formulas“ ,אמרו introduce early tannaitic traditions (Mishnah/Tosefta). However, in SEZ and SER, one observes an innovative usage of these terminologies which may be paralleled by changing attitudes to older traditions in later rabbinic and gaonic texts.42 Through a more flexible application, the formulas refer in SEZ to post-tannaitic texts like the Talmudim, the midrashim and ethical traditions (DE/ARN/Avot). In other instances, such formulas serve to camouflage the introduction of traditions that only have parallels in SEZ’s sister tradition SER. In neither case, we find reliable direct quotations or even a straightforward correspondence between the related traditions. This phenomenon is not only due to an unquestionable orality of rabbinic culture as the background of its transmission. In fact, it points also to strategies of skillful integration of other traditions, while transforming them in order to create new meanings. Most interestingly, we find also the usage of markers of reference in the case of traditions or teachings that are original features of SEZ.43 This kind of “forged” application may be conceived as pseudepigraphy. However, Marc Bregman’s study on “creative attribution” has demonstrated that reference is an important tool for a collective and flexible rabbinic authorship.44 Instead of being a deliberate fraud, this referential strategy assures a strong connection to the rabbinic chain of tradition and its authority (Torah le-moshe me-sinai). Against this backdrop, I suggest that the application and adaptation of classical terminologies are

42 Gaonic texts, as well as later midrashim like PRE, display a considerable expansion of possible pre-texts for this introduction formula which now alludes also to talmudic and midrashic material. For examples, see PRE 25 (reference to bEr 18b); PRE 40 (bSanh 40a); PRE 37 (bSota 41b); PRE 38 (bHul 4a); PRE 50 (bMeg 15a). which serves as a מיכן\מכאן אמרו Bacher stressed the rhetoric function of the formula connection between a rabbinic teaching and a biblical verse. Dina Stein observes the same function in PRE but describes it as a connecting device that links post-biblical customs (minhag) to certain biblical proof-texts. Cf. Wilhelm Bacher, Die exegetische Terminologie der jüdischen Traditionsliteratur (Leipzig: Hinrichs, 1905), 82; Stein, Maxims, Magic, Myth, 41. .as a characteristic terminology of the talmudic period מיכן אמרו Zunz treats the formula Cf. the discussion of the formula in PRE in Sacks, Midrash and Multiplicity, 135–140. 43 Such a move is also discernible in PRE (18/25), where original teachings about certain as if they מיכן\מכאן אמרו customs (Kiddush/ circumcision) are introduced by the formula were parts of earlier rabbinic traditions (like Mishnah, Toseftah or the two Talmudim). 44 See Marc Bregman, “Pseudepigraphy in Rabbinic Literature,” in Pseudepigraphic Per spectives: The Apocrypha and Pseudepigrapha in Light of the Dead Sea Scrolls, ed. Esther G. Chazon et al. (Leiden: Brill, 1997), 27–41. 230 lennart lehmhaus best understood as literary strategies of embedment. SE does not employ received terminology as a means of deliberate pseudepigraphy, but as a strong referential anchor for its own original teachings. Following Fou- cault, I would like to describe these explicit markers of intertexuality as a way of SEZ to inscribe itself into the bigger discourse of biblical and rabbinic tradition in order to participate in and benefit from its authority. This observation is shared by Steven Daniel Sacks, who argues that the use of attributions as ‘tradition markers’ is a “literary and expository feature of rabbinic literature which can provide insight into a work’s engagement with the tradition [. . .] which reflect[s] the expositor’s participation in classical rabbinic discourse.”45 On a second level, however, the deliberate eschewal or omission of attribution to named rabbinic sages is not a random phenomenon but might attest for an additional discursive strategy which simultaneously points into two directions. First, the anonymity of a teaching was regarded in rabbinic tradition since the earliest stages as the opinion of the majority. Thus, as an expression of the collective consensus, it was considered most trustworthy and was given ultimate authority.46 Second, this move away from the many-voiced midrashic or talmudic discourse might have been also a purposeful strategy of dissociation from a closed, and probably more elitist rabbinic-gaonic milieu. This would have rendered Seder Eliyahu’s messages neutral and thus more appealing throughout a broader diversity of non-rabbinic Jews.

A Linguistic Turn in Late Midrash?

In different passages of SEZ, one observes interesting and rather uncom- mon compositional techniques. Their focus is clearly directed to one keyword or word-stem and its semantic field which is central for the related teachings. In this, the passage resembles the structural feature described by Robert Alter with regard to biblical literature as a “word-motif.” This literary strategy utilizes the Hebrew system of trilateral roots which structurally emphasizes the etymological relation between word forms in order to develop a thematic argumentation. Alter based his observation on the so-called Leitwortstil applied by Martin Buber and Franz Rosenzweig. This

45 Sacks, Midrash and Multiplicity, 60f. 46 Cf. bYeb 42b; bBer 9a and bShab 46a (“The Halakhah is according to the anonymous teaching”). between tradition and innovation 231 model includes a repetition of a word or cognate word-forms in which the slight differences between words and their contexts of application constitute the dynamic element in the construction of meaning.47 A very instructive example can be found in SEZ, chapter 14, which is pervaded by this hermeneutic and literary strategy. Across different micro- expositions and thematic clusters, the text explores variants of meaning The analytical but skillful play with word forms is 48.ירש for the root enhanced by its hermeneutic and rhetorical-didactic purposes. SEZ, chapter 14 (196–97)

Among those reading [Scripture] and reciting/repeating [Mishnah],49 who -He who spends much time eat 50?(מווריש את עצמו) impoverishes himself ing, drinking [and sleeping]. As is said: For the drunkard and the glutton will come to poverty (and slumber will clothe them with rags. (Prov 23:21 ,(כי סובא וזולל יוורש) And as is said: Love not sleep, lest you come to poverty open your eyes, and you will have plenty of bread ;(אל תאהב שינה פן תיוורש) (Prov 20:13) .(פקח עיניך ושבע לחם) And bread does not refer to anything else than words of Torah, since it is said: .[and drink of the wine I have mixed] (לכו לחמו בלחמי) Come, eat of my bread (Prov 9:5) And [Scripture] says: The Sovereign LORD has given me an instructed tongue to know the word that sustains the weary ,(לשון למודים) (Isa 50:4) .(לעות את יעף דבר) are only absorbed by (דברי תורה) Haven’t you learned that words of Torah .(שהוא עיף להם) one who wearies himself for them

47 See Robert Alter, The Art of Biblical Narrative: New and Revised Edition (New York: Basic Books, 2011), 111–122, esp. 116f. 48 It is especially this root which exhibits a broad range of different meanings in biblical as well as in later rabbinic Hebrew. Basic meanings can be translated as follows: to take possession of; to subjugate; to conquer; to inherit; to give as heritage; to expel, to dispos sess; to impoverish. Cf. Wilhelm Gesenius, Hebräisches und Aramäisches Handwörterbuch über das Alte Testament (Berlin: Springer, 1962 [reprint]), 321; Georg Fohrer, Hebräisches und aramäisches Wörterbuch zum Alten Testament (Berlin: de Gruyter, 1997 [reprint]), 115; Marcus Jastrow, A Dictionary of the Tragumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature (Jerusalem: Horeb, s.d. [reprint]), 595 and 598. seldom (שונין) and the related verb (משנה) In Seder Elijahu (SE), the term mishnah 49 refer to the corpus of rabbinic law known to us as the Mishnah. In fact, the reference in SE (Mishnah & shonin) aims at the whole learned tradition of rabbinic lore or Oral Torah (Torah she-be-al-pe) which complements and builds upon the Written Law or Scripture .(קוראין /and reading מקרא /referred to in SE via the terms Miqra) One mean .מוריש :In accordance with the MS Parma 2785, the verb is emended into 50 ing of the Hiphil is: ‘to impoverish someone’. Cf. Jastrow, Dictionary, 598. 232 lennart lehmhaus

Thus the spirit of the Holy proclaimed good news to the disciples of the sages and said to them: My children, though I have given to you a good Torah in this world, I need not to say [anything concerning the Torah] for the world-to-come. Though there is a double reward for you in this world, I need not to say [anything concerning the reward] for the world-to-come. Do not belittle the Torah and do not increase your eating and drinking, since it is said: Return to your stronghold, O prisoners of hope; today I declare that I will Zech 9:12)51) .(משנה אשיב לך) restore to you double .(ושלמתי לכם את השנים) And [Scripture] says: And I will repay you twice (Joel 2:25–26)52 -dis 53,(בשתכם משנה) And [Scripture] says: Instead of their doubled shame grace and dishonor in their land they shall inherit a double portion (they shall have everlasting joy (Isa 61:7 ;(משנה יירשו) Haven’t you learned that you will savor a double portion in the days of the anointed?

Even if only a single individual of the people of Israel were dwelling at the end of the world, and a thousand rivers before him, the Holy One, blessed be he, would divide them and bring him to himself [. . .]

And whence [do we know/learn]? You shall know that it is like this :(מנין תדע לך שכן) For Israel will not be redeemed because of suffering, not because of the bondage, not because of [forced] migration, not because of expulsion, not because of need, not because of hunger, but because of ten men who are together; and one is reading to his fellow and his voice is not heard. Since it is said: But on Mount Zion will be deliverance; it will be holy and the House of Jacob, they will take possession of their dispossessors (Obad 1:17) .(וירשו בית יעקב את מורשיהם)

Thus, we found that our ancestors [. . .] And whence [do we know/learn]? You shall know that it is like this

51 Or: “I will restore you your learning of oral lore (mishnah).” This seems to be the understanding or the intended reading of the author(s) of SEZ. 52 Or: “I will repay you/ restore you the years [of wearing yourself in study].” For the close connection between a life of learning and certain hardship and self-deprivation, see the discussion of ascetic dimensions in Michael Satlow, “ ‘And on the Earth You Shall Sleep’: Talmud Torah and Rabbinic Asceticism,” JR 83 (2003): 204–225, and Eliezer Dia mond, Holy Men and Hunger Artists: Fasting and Asceticism in Rabbinic Culture (New York: Oxford University Press, 2004). 53 Or: “Instead of your shame [which you endured because of your devotion to] mish nah.” This promise is closely linked to the preceding one in emphasizing the physical and also social deprivations accepted by rabbinic sages. This teaching seems to point to a cer tain degree of external discrimination or even hatred directed at the learned circles. between tradition and innovation 233

that Israel will be redeemed only if they increase and (מנין תדע לך שכן) multiply and thus rise to plentifulness in the world. As it is said: For you will spread out to the right and to the left; your descendants will dispossess nations (and settle in their desolate cities. (Isa 54:3 (וזרעך גוים יירש) Throughout this passage, we find six instances in which differing verb are applied within a cluster of related teachings or ירש forms of the root in the embedded biblical quotations. The first appearance is in a short teaching about the moderate lifestyle of the students and opens with the The two ”?(מי מווריש את עצמו) question: “Who impoverishes himself answering biblical quotations provide other verb forms of the same root that tie excessive eating, drinking, and sleeping to self-impoverishment: ,(כי סובא וזולל יוורש) For the drunkard and the glutton will come to poverty“ and slumber will clothe them with rags” (Prov 23:21). “Love not sleep, lest open your eyes, and you ;(אל תאהב שינה פן תיוורש) you come to poverty will have plenty of bread” (Prov 20:13). ירש These three applications define the semantic word field of the root negatively in a context of admonition. The text conveys a sober and very explicit message. A student of Torah (and preferably everyone inclined to a moral lifestyle) should avoid any behavior that brings him close to the self-indulgence described in the biblical verses. For such a conduct disturbs him and leads him away from his study. The final effect will be that he “impoverishes himself,” not regarding his material properties but with respect to the merits of his study. Thus, in this passage, both question and proof texts help to narrow down the semantic range of the verb in order to build a link between material and spiritual poorness. Material neediness and destitution are valued throughout SEZ and also in the following paragraph as a midda tova (good measure/attribute). For conditions of travails and tribulation encourage Israel to Torah study, prayer, and good deeds. Hence, the language of impoverishment here aims rather at intellectual deprivation which comes from exorbitant consumption and an immoderate lifestyle. The connection to one’s study is further developed through the metaphor of bread as a symbol for Torah as “spiritual food.” The verse from Prov 20:13 clearly condemns sleeping as a reason for impoverishment, while the keen student is appreciated and motivated to receive his bread (i.e., Torah and its merit). The connection between bread and Torah is reinforced through a micro-exegesis that presents two apposite biblical verses. The first is rather common when it ascribes the invitation in Prov 9:5 (Come, eat of my bread) to God addressing the students of Torah. However, the second 234 lennart lehmhaus verse is more complex since it draws the line back from the metaphor of bread to the complex of learning and self-deprivation. This is achieved from Isa 50:4. While למודים through an ambivalent reading of the word this word and its root are usually understood as referring to learning and instruction, we might assume that the author(s) also alluded to its second meaning: a common bread with the shape of a shingle.54 Thus, the usage of this verse teaches about value of learned words for those who trouble themselves (“to know the word that sustains the weary”), while the metaphorical dimension of bread and Torah reverberates in the background. The following teaching serves as a summary of the preceding exegesis and explicitly emphasizes that God only grants a portion of bread (respective Torah) to the student ready to weary himself for his study. Based on this bridging motif of bread, the next paragraph is devoted to God’s proclamation of a double reward for those who concentrate on Torah and refrain from excessive consumption and sleeping. This passage exploits the semantic range of two word roots that exhibit a homographic and homophonic closeness. In the first biblical proof text from Zech 9:12, one finds the divine promise: Today I declare that I will restore to you Through only a slight change, reading mishnah .(משנה אשיב לך) double instead of mishne, the verse matches exactly the preceding teaching. God gives a double reward to those who are committed to study the tradition (Mishnah). Also the following verse from Joel 2:25 points into the same as either שנים direction. Here both possible meanings of the written word shanim, ‘years’ or shenayim, ‘two/twice’ can be understood to refer to two interconnected promises for the students: I will restore/repay you the years [of wearing yourself in study] and I will repay/restore you twice. The last bib- Qal) and gives it a positive) ירש lical proof from Isa 61:7 picks up the root connotation within an eschatological context. The reward of the students is described as follows: Therefore in their land they shall inherit a double they shall have everlasting joy.55 In ;(חלקם לכן בארצם משנה יירשו) portion a reappraisal, this verse links the topic of abstention and austerity back to with the meaning ‘inheritance’. Simultaneously, this keyword ירש the root

54 Cf. Jastrow, Dictionary, 708, and its use in bBer 38a. 55 Cf. Isa 61:7: “Instead of your shame there shall be a double portion; instead of dis honor they shall rejoice in their lot; therefore in their land, they shall possess a double תחת בשתכם משנה וכלמה ירנו :portion; they shall have everlasting joy.” The BHS reads For an alternative translation which .חלקם לכן בארצם משנה יירשו שמחת עולם תהיה להם renders the double-entendre of mishne and Mishnah, see above n. 48. between tradition and innovation 235 is here enmeshed in a complex wordplay based on the homography and / שנים shnayim / משנה homophony between the double reward (mishne .(משנה and the study of rabbinic lore (Mishnah (כפול kaphul within ירש In the remaining chapter, two more applications of the root a setting of redemption can be observed. The first is introduced by a teaching that highlights individual and collective study as the main pre- condition for divine salvation. The quoted verse from Obad 1:17 describes וירשו בית יעקב את :Israel’s eschatological return to Zion concluding with ,we-yarshu beit yaaqov et morasheihem). The usual translation) מורשיהם and the House of Jacob, they will take possession of their inheritance, would be merely a variation of the theme of inheritance, already alluded to in the verse from Isaiah. However, a different reading recommended in the BHS and by Gesenius’ dictionary discloses additional dimensions of the we-yarshu) וירשו בית יעקב את מורישיהם biblical statement. In the reading beit yaaqov et morisheihem), the meaning shifts to and the House of Jacob, they will take possession of their dispossessors. Thus, the verse adds further meanings as “taking possession” and “dispossessing/ expelling” to .which are mirrored in this statement ,ירש the semantic range of the root Moreover, this reading bestows upon the statement a circular understanding of eschatological events. The common Jewish and rabbinic motif of return to Zion and messianic reward is reinforced by the reversal of present balance of power between Israel and their oppressors. The final application of the key-word in a Qal follows this line of argumentation. For the last passage of chapter 14 presents the observance of as a requirement sine qua non (פרו ורבו) the commandment of procreation for Israel’s future salvation. The biblical proof text from Isa 54:3 combines the promise of Israel growing to a powerful people with the motif of re- conquest: For you will spread out to the right and to the left; your descendants .(וזרעך גוים יירש) will dispossess nations and settle in their desolate cities The foregoing analysis demonstrated that more than two third of the chapter’s (SEZ 14) compositional structure evolves from a dense cluster As .ירש of teachings and verses connected to the key element of the root connecting devices, the different forms of the verb explore the semantic field of the root, while tying together ethical, intellectual-educational and eschatological topics, using a wide range of rhetorical devices, like maxims, prophecy-like speech and micro-exegesis. The observed techniques of semantic and lexical survey and their implied linguistic interest can be found also in other thematic sets, enumerations and passages of keyword or trigger word compositions throughout SEZ 236 lennart lehmhaus and SER.56 The literary strategy is twofold and mutual reinforcing. On the plain level, the text uses different morphological forms of a key term in quotations of biblical verses. They are merged into a strong rhetorical cluster of interconnected teachings corresponding to one or two lager topical frameworks. As such, the word forms are of crucial importance for the text’s construction and conveyance of meaning. At the same time, by means of application within the moral-religious context, the semantic range of the root is explored and outlined to all intents and purposes. Seder Eliyahu is not the only tradition in which an increasing interest in different aspects of language can be discovered. Several scholars pointed to similar features in the tradition of Pirqe de-Rabbi Eliezer (PRE). Dagmar Börner-Klein and Ute Bohmeier even labeled this complex text a “philological midrash” owing to the main literary strategies they observed in the work. According to their studies, PRE is particularly concerned with a clarification of biblical verses containing a morphological, lexical, grammatical, or semantic problem. They suggest that the text meets this challenge by adopting only interpretations pertaining to those problems and arranging them into a narrative structure that loosely follows biblical history.57 How the shared features of those texts relate to their wider cultural context shall now be discussed in the concluding part of this article.

Conclusion: Transmission, Transformation and Innovation in a Context of Change

The main part of this contribution presented some of the literary developments, differences, and innovations in SEZ, which are partly paralleled in other post-talmudic texts. Due to limited space, this discussion was confined to a more detailed analysis of only a few literary strategies: the

56 In SEZ 11 (191), one finds another passage which discusses the semantic field of and reproof (tokhahah (מוסר discipline (mussar ,(יסורין suffering/chastisement ( yissurin In contrast to the already analyzed example does this discussion not concentrate .(תוכחה on different manifestations of one root (or lexeme). Instead, one finds several word-forms of different roots that constitute a semantic field due to their similar or closely related meaning. 57 See Börner-Klein, Pirke, XXX–XXXIX; Bohmeier, Exegetische Methodik, 6–18. Stem berger, Einleitung, 366, expresses some reservations regarding Bohmeier’s description of the work’s literary character. Primarily, he disagrees with her dating of PRE to the time of Sa’adya Gaon. However, also Sacks, Midrash and Multiplicity, 88–108, argues for PRE’s fresh approach to the Bible and its language which transcends the dichotomy between narrative and exegesis. between tradition and innovation 237 adaptation of rabbinic pedagogical terminology and reference markers as well as the exploration of a semantic field. These examples as well as other features described at the beginning yield some interconnected results. In all analyzed strategies, one notices a strong tendency towards application and adaptation of former terminology and traditions for the text’s own innovative purposes. This includes a simultaneous process of transmission and transformation. Rabbinic genres and terminologies were not just adopted but got adapted and appropriated according to the new rhetorical-didactic orientation of a thematic discourse. Seder Eliyahu, as well as other post-talmudic works, exhibits a shift in their dealing with language. Semantic and lexical dimensions of keywords are explored and utilized to create topical units. Simultaneously, this strategy is smoothly intertwined with a complex discussion of theological or ethical issues. For instance, the exploration of a semantic field in SEZ 14 deploys the camouflage of the traditional form of proof texts and micro- exegesis in order to address lexical issues and to build a rhetorical cluster on ethics, study, and reward. Since this is only one of many principles for organizing the text, I am still reluctant to call the approach ‘philological midrash’, as Ute Bohmeier did regarding PRE. However, the phenomenon itself definitely accounts for an increased attentiveness to and a growing interest in language. This pertains especially to the language of Scripture and certain rabbinic traditions, with all their linguistic and semantic dimensions that shape and are shaped by an extra-textual reality. This interest might have been triggered by the texts’ broader cultural context of origin. The early gaonic period witnessed a ‘return to Scripture’ in two segments of Jewish society. First, in various places, learned circles engaged in thorough study and commentary of the biblical text. Drory suggests that this masoretic enterprise constituted an ambivalent or neutral literary mode which was accessible to and accepted by rabbinic and non-rabbinic Jews.58 Second, we find similar scripturally oriented groups among the different branches who later merged in the (proto-)Karaite movement. The focus of their study was the Bible upon which they commented. Scripture became their authoritative source of individual halakhic inquiry and theological reason- -lit.: ‘search’). Both developments were probably encour חיפוס ing (hippus aged and inspired by the Muslim literary and textual ideal of the Quran. Moreover, Arabic literature provided also models for commentaries and

58 See Drory, Models and contacts, 138–142. 238 lennart lehmhaus treatises which were directed by linguistic and (pre-) grammarian interest in the Quran and its language. As shown above, classic rabbinic terminology and quotation markers— instead of functioning merely as an editorial device—serve in Seder Eli- yahu the mediation and legitimation of its own discourse. SEZ prefers the open and anonymous general reference to rabbinic tradition which includes both moves—distancing and bringing near. On the one hand, the anonymity of quotations as well as the anonymous first-person narrator lends to the text universal authority, as pointed out above. On the other hand, simultaneously, this universal outlook makes these traditions function quite independent from a specific socio-cultural and regional milieu (e.g., the yeshivot in Babylonia and Palestine) and distances them from a certain intellectual or literary ideal (like the dialectic discussions of the Talmudim). Thus, the absence of polyphonic discourse and attributions as well as the departure from midrashic exegesis and talmudic dialectics mark a major difference in the text’s overall character. This structural changes dovetail with the textual program or agenda to be found in SEZ.59 While rabbinic texts generally can be described as ahistoric in their interest, SEZ provides an example par excellence for a timeless and ubiquitous discourse, which probably came not from the mainstream of rabbinic culture. The extensive use of universal teachings, maxims or parables instead of exegesis, dialectics, and talmudic reasoning, may have appealed to a variety of Jewish addressees of different educational and communal backgrounds. Elsewhere, I have argued that SEZ promotes a condensed version of minimal Judaism, which is appealing and accessible to a wider Jewry.60 The omission of attributions to named sages detaches those traditions from a distinct scholarly milieu of the yeshivot or batei midrash. It rather points to a basic and more individual and decentralized form of learning which was available in all Jewish communities throughout the Muslim dominion and beyond.61 The universal approach and the idea of

59 It is striking that Seder Eliyahu Zuta—despite some deviations—correspond to Drory’s characterization of the innovative Karaite modes of writing. Drory, Models and contacts, 154: “a thematic piece following the logic of reading, systematically divided into chapters and topics and headed by a methodological introduction, written in first person and attributed to its real author.” 60 See 8–9. 61 On this point, I fully agree with Dina Stein, who suggests such a universal approach in her comparison of PRE and SE in a study that I quote throughout this article (Stein, “Pirke de-Rabbi Eliezer”). Unfortunately, until shortly before attending the conference between tradition and innovation 239 a ‘minimal Judaism’, as promoted in SEZ and other traditions, might have served as connecting elements between rabbinic and non-rabbinic Jews. As such, it was likely aimed at bringing down rabbinic concepts to a common denominator for a broader, more diverse Jewry between Meso- potamia and the Iberian Peninsula, which got (re-)united under the new Muslim dominion.62 This period was once termed the “Dark Age” of Jewish history by Rob- ert Brody due to the paucity of Jewish sources.63 However, from external evidence, we know that the emergence of late midrash, precisely in early Muslim and gaonic time, coincides with radical transformations and shifts occurring in the socio-cultural, economic, and religious spheres of the Mediterranean. The rise of Islam brought in its wake political, cultural, and religious innovations and uncertainties with a deep impact on the popula- tion of the Middle East. In the formative period of Islam, we find different philosophical schools and religious branches who stood in competition and whose ideas were highly influential. As a response to the changes, several ascetic, messianic, and millenarian movements as well as hybrid groups sprung up among Muslims, Christians, and Jews.64 More recent scholarship argues that rabbinic culture was but one of many facets of late antique and early medieval Jewry. Its consolidation and rise to hegemony took rather a long and winding road and rabbinic culture possibly became

“Ancient Jewish Texts and the ‘Literary’ ” in Antwerp (March 2012), I was not aware of this paper and had no possibility to obtain a copy. When Dina Stein and I met at the confer ence for the first time, we were both astonished that we had come to similar conclusions, although our points of departure and methodological approaches differed. I am indebted to Dina Stein who was so kind to send me her article. On alternative milieus of learning, see also Adiel Kadari, “Talmud Torah in Seder Eliyahu: The Ideological Doctrine in its Socio-historical Context,” Daat 50–52 (2003): 35–59 [Hebrew]. 62 Moreover, this strategy of dissociation might have been intentionally chosen as a safeguard against Karaite (and Muslim) critique that the rabbinic Oral Torah is merely a product of human reasoning and invented by individual sages, who differ from one another often. Simultaneously, this eschewal of certain traditional forms and norms likely facilitated its reception among non-rabbinic Jews. We have to be aware that our descrip tion of these segments of society and culture as ‘peripheral’ or ‘marginal’ is due to the triumph of rabbinic-talmudic Judaism which prevailed as the dominant manifestation of Jewish culture only since the Middle Ages. Cf. Talya Fishman, Becoming the People of the Talmud: Oral Torah as Written Tradition in Medieval Jewish Cultures, Philadelphia: Univer sity of Pennsylvania Press, 2011. 63 See Robert Brody, The Geonim of Babylonia and the Shaping of Medieval Jewish Cul ture (New Haven: Yale University Press, 1998), 4. 64 See Steven Wasserstrom, Between Muslim and Jew: The Problem of Symbiosis Under Early Islam (Princeton: Princeton University Press, 1995). 240 lennart lehmhaus more dominant in gaonic times.65 The rabbinic formative process and consolidation as well as its struggle for influence triggered the emergence of non-rabbinic, (proto-) Karaite trends. These groups preferred alternative modes of learning, developed different religious traditions in dialogue with their Jewish and rabbinic counterparts as well as Islamic cultural models.66 However, recent research questions a clear schism between Rabbanite and Karaite Jews—labels that were most likely useful to create a discourse of heresy for some intellectual or political purposes. Scholars rather suggest a closer interaction between those groups within a perme- able society where affiliation and belonging was affected rather by socio- cultural and familial constellations than by religious deed and creed.67 This setting was complicated by the internal competition between old rabbinic centers of learning themselves (Babylonia/Palestine) and newly arising outposts around the Mediterranean. Simultaneously, these centers had to accommodate differing customs and traditions from the geographical and cultural margins.68

65 Seth Schwartz suggests a rather late process of ‘rejudaization’ or ‘rabbinization’ in late-antique Jewry (Imperialism and Jewish Society, 200 B.C.E. to 640 C.E. [Princeton/Oxford: Princeton University Press, 2002], 240–74). This observation is shared by Hayim Lapin, “Aspects of the Rabbinic Movement in Palestine 500–800 C.E.,” in Shaping the Middle East: Jews, Christians, and Muslims in an Age of Transition 400–800 C.E., ed. Kenneth G. Holum & Hayim Lapin (Bethesda MD: University Press of Maryland, 2011), 181–194. Catherine Hezser and Stuart S. Miller resp. argue more cautiously for a slow broadening of influence via the backbone-network of rabbinic households and circles in close interaction with and the support from non-rabbinic Jews (The Social Structure of the Rabbinic Movement in Roman Palestine [Tübingen: Mohr Siebeck, 1997], 155–239; 307–27; Sages and Commoners in Late Antique ’Erez Israel: A Philological Inquiry into Local Traditions in Talmud Yerushalmi [Tübingen: Mohr Siebeck, 2006], 446–466). 66 Haggai Ben Shammai, “Between Ananites and Karaites: Observations on Early Medi eval Jewish Secterianism,” in Studies in Muslim-Jewish Relations Vol. 1, ed. R.L. Nettler (Chur: Harwood Academic Publishers, 1995), 19–29; Meira Polliack, “Medieval Karaism,” in The Oxford Handbook of Jewish Studies, 295–326. Fred Astren describes six disparate elements within the proto-Karaite milieu: 1) Annanites; 2) peripheral Jewish traditions; 3) Judeo-Muslim or other hybrid identities; 4) anti-rabbinic scripturalists (the bene miqra in Pirqoi ben Baboi’s polemic?); 5) remnants of messianic movements (Isawiyyah/al-Ishafani); 6) (semi-) ascetic trends among Palestino-centric Jews (e.g. Mourners of Zion) (“Islamic Context of Medieval Karaism,” in Karaite Judaism: A Guide to Its History and Literary Sources, ed. Meira Pollack [Leiden: Brill, 2003], 145–177). 67 Cf. Rustow, Heresy and the Politics of Community; idem, “Laity versus Leadership in Eleventh-Century Jerusalem: Karaites, Rabbanites, and the Affair of the Ban on the Mount of Olives,” in Rabbinic Culture and Its Critics, ed. Daniel Frank and Matt Goldish (Detroit: Wayne State University Press, 2008), 195–248; idem, “Karaites Real and Imagined: Three Cases of Jewish Heresy,” P & P 197 (2007): 35–74; Frederick E. Greenspahn, “Sadducees and Karaites: The Rhetoric of Jewish Sectarianism,” JSQ 18, 1 (2011): 91–105. 68 Moshe Gil, A History of Palestine, 634–1099 (Cambridge: Cambridge University Press, 1992). This gaonic claim to power was consolidated through religious and juridical expertise between tradition and innovation 241

Therefore, one can imagine a socio-historical context of challenge and change for Seder Eliyahu and similar late traditions that had a deep impact on their authors and called for transformations in form and content. This stands in contrast to the analysis of the Jewish literary system of the gaonic period given by Rinat Drory. She assumed that the hegemonic rabbinic or “Rabbanite literature [. . .] was progressively stagnating, its creative models showing signs of ossification” due to their reproduction of classical paradigms and a strict adherence to their literary norms and to an ideology of oral transmission. Moreover, “no new literary material could be admitted in a Rabbanite text unless it would be molded after the classical models.”69 This description, which serves Drory’s overall dichot- omist argument concerning ‘Rabbanite stagnation’ and ‘Karaite innovation,’ fails to notice the more subtle changes taking place in late midrashic literature and thought. With respect to PRE, scholars described these developments as a “renewal of rabbinic interpretive culture” or “the return of the repressed.”70 In the case of Seder Eliyahu, one can focus on its hybrid character, which oscillates between exegesis, narrative, and moral treatise and so anticipat- ing literary trends known from the Middle Ages onwards. It seems obvious that these new elements did not enter into Jewish, rabbinic culture through the well-guarded portal of gaonic and talmudic halakhic literature. It might have been rather through the backdoor of aggadah that gradual shifts found their way from the ‘margins’ into the ‘center.’ In the course of this development, the later midrashim were receptive to and probably participating in other, overlapping ‘peripheral’ discourses shaped in masoretic, (proto-) Karaite or esoteric circles and their Muslim-Christian surroundings.71 Further attentive analyses of those traditions, their literary disseminated in responsa and a tight inter-communal network of money-collectors, mes sengers and representatives. See also Brody, The Geonim, 35–53, 83–99, and 123–134. As a parallel, one can read the claim to leadership of religious specialists (the Ulama) in Islam. See Astren, “Islamic Contexts,” 156–158. The interplay and struggle between center and periphery is discussed in Menahem Ben-Sasson, “Inter Communal Relations in the Gaonic Period,” in The Jews of Medieval Islam: Community, Society and Identity, ed. Daniel Frank (Leiden: Brill, 1995), 17–31. 69 Drory, Models and Contacts, 150–152. 70 Cf. Sacks, Midrash and Multiplicity; Rachel Adelman, The Return of the Repressed: Pirqe de-Rabbi Eliezer and the Pseudepigrapha (Leiden: Brill, 2009). 71 This observation can be confirmed and supported by looking at the inventor of medi eval rabbinic culture: Sa’adya Gaon. Scholarship mostly agrees on his outstanding role as a cultural and religious pioneer, who appropriated several modes of writing (treatises, Bible-commentaries, prose texts and lexical works in Hebrew, Arabic and Judeo-Arabic) and introduced new philosophical and theological ideas. Cf. Drory, Models and Contacts, 242 lennart lehmhaus and religious innovations as well as their transformative functions might allow us to gain more important insights into the formative, but also more diverse period of what later became known as medieval, talmudic Jewish culture.

140–146 and 178–190; Brody, The Geonim, 235–332. However, Talya Fishman argues that Sa’adya succeeded precisely because he made his way as a maverick from the Palestino- Egyptian periphery into the heart of Babylonian gaonic culture, being highly influenced by his contacts with Masoretes, Arab writing culture and Karaite literature and thought. See Fishman, Becoming, 57. It seems unlikely that Sa’adya—his designation as a “revolution- ary champion of tradition” (Brody) notwithstanding—was the first and only one who all at once turned upside down Jewish gaonic culture out of the blue. Even though this idea of an all-encompassing genius is tempting, one may assume that others before him have been traveling this road. These precursors and contemporaries remained anonymous and did not gain the top of our historical consciousness. Medieval Rationalist Exegesis and Intertextuality in Samuel Ibn Tibbon’s Ma’amar Yiqqawu ha-Mayim

Rebecca Kneller-Rowe

Abstract

Samuel Ibn Tibbon, famous for his translation of Maimonides’ writings, was also a philosopher, an exegete and an original thinker in his own right. His exegesis was guided by a basic intertextual view of Scripture and of the role of the exegete. Despite the difference of style and approach related to the different cultural setting, most of the premises underlying Ibn Tibbon’s hermeneutics served as guidelines for midrashic exegesis as well as for the writings of his rationalist predecessors. Samuel Ibn Tibbon’s contribution resides in the framing of these understandings within a fully-fledged theory of interpretation and in the systematic implemented of the theory in his exegesis. According to Ibn Tibbon, biblical texts are interconnected to one another in a way that posterior texts contain explanations, elaborations and revelations of enigmatic passages and secrets concealed in earlier parts of Scripture. Other texts connect by virtue of their being varying expressions of paradigmatic topoi. Exegesis, accordingly, is an inter- generational project of progressive unraveling of the esoteric subtext, beginning with the prophetic books and the hagiographa and continuing with post-biblical interpretations by the Sages, by Maimonides and by Tibbon himself. Under the guise of continuous interpretation, Ibn Tibbon introduces a radical or perhaps a subversive reading of Scripture in an attempt to bridge the gap between Aristo- telian science and the biblical texts.

The lens of intertextuality has been turned towards biblical studies in recent years.1 A number of studies have focused on the interconnectedness of the Scriptural texts and several studies have been devoted to midrashic literature.2

1 This expression is borrowed from the title of Beth LaNeel Tanner, The Book of Psalms Through the Lens of Intertextuality (New York: Peter Lang, 2001). 2 See for example James T. Hibbard, Intertextuality in Isaiah 24–27: The Reuse and Evo cation of Earlier Texts and Traditions (FAT 2; Reihe, 16; Tübingen: Mohr Siebeck, 2006); LaNeel Tanner, The Book of Psalms; Gerald Bruns, “Midrash and Allegory: The Beginnings of Scriptural Interpretation,” in The Literary Guide to the Bible, ed. Robert Alter & Frank Kermode (Cambridge, MA: Belknap Press, 1987), 625–646; Daniel Boyarin, Intertextuality and the Reading of Midrash (Jerusalem: Hartman Institute & Alma, 2011); idem, “Old Wine in New Bottles: Intertextuality and Midrash,” PT 8 (1987), 539–556; Donald Polaski, Autho rising and End: The Isaiah Apocalypse and Intertextuality (Leiden: Brill, 2001); James D. Nogalski, “Intertextuality and the Twelve,” in Forming Prophetic Literature: Essays in Honor of John D. W. Watts, ed. James W. Watts & Paul R. House (JSOT.SS 235; Sheffield: Sheffield 244 rebecca kneller-rowe

Intertextual aspects of medieval Bible exegesis have attracted little or no attention as yet. However, no matter how far ancient and medieval exegesis is from modern literary theory, some of the concepts linked to intertextuality provide useful criteria and insights for the understanding of the workings of traditional Bible scholarship. This article will try to trace Samuel Ibn Tibbon’s exegetical tra- jectory on the one hand and make a contribution, however modest, to the study of medieval rationalistic Bible interpretation from an intertextual viewpoint on the other hand. Taking intertextuality as a central paradigm for the investigation of exegesis of a medieval rationalist will allow us to view the subject from a somewhat different perspective.3

Samuel Ibn Tibbon (Provence 1160–1232), famous mainly for his translation of Maimonides’ Guide of the Perplexed, was also an original philosopher and Bible exegete in his own right. Many aspects of his work have been investigated since Aviezer Ravitzky’s illuminating research on Samuel Ibn Tibbon and the Tibbonide School that developed after him.4

Academic Press, 1996), 102–124; Michael Stead, The Intertextuality of Zechariah 1–8 (New York: T&T Clark, 2009); Danna Nolan Fewell & Gary A. Phillips, eds., Bible and the Ethics of Reading (Semeia 77, Atlanta: Scholars Press 1977); Sipke Draisma, ed., Intertextuality in Biblical Writings: Essays in Honor of Bas van Iersel (Kampen: Kok, 1989); Robert P. Caroll, “Intertextuality and the Book of Jeremiah, Animadversions on Text and Theory,” in Lit erary Criticism and the Hebrew Bible, ed. Cheryl Exum & David Clines (Sheffield: JSOT Press, 1993), 55–78; David M. Gunn & Danna N. Fewell, eds., Narrative in the Hebrew Bible (Oxford/New York: Oxford University Press, 1993), especially chapter 7. See also relevant discussions in Michael Fishbane, Biblical Interpretation in Ancient Israel (Oxford: Claren don Press, 1985) and Meir Sternberg, The Poetics of Biblical Narratives: Ideological Literature and the Drama of Reading (Bloomington, IN: Indiana University Press, 1987). 3 For general introductions to intertextuality and related questions in modern liter ary theory, see Graham Allen, Intertextuality (London/New York: Routledge, 2006); Bar bara Goddard, “Intertextuality, ” in The Encyclopedia of Contemporary Literary Theory: Approaches, Scholars, Texts, ed. Irena I. Makaryk (Toronto: University of Toronto Press, 1993), 568–571; Heinrich F. Plett, ed., Intertextuality (Berlin: W. de Gruyter, 1991); Polaski, Authorising and End, 32–49; Michael Worton & Judith Still, eds., Intertextuality, Theories and Practice (Manchester: Manchester University Press, 1991); Zeev Levy, Hermeneutics (Tel Aviv: Sifriat Hapoalim, 1986) [Hebrew]; Ziva Ben-Porat, “The Poetics of Allusion” (PhD diss., University of California, Berkeley, 1973). 4 George Vajda was the first to turn his attention to Samuel Ibn Tibbon’s works. Cf. George Vajda, “An Analysis of the Ma’amar Yiqqawu ha-Mayim by Samuel b. Judah Ibn Tibbon,” JJS 10 (1959): 137–149; idem, Recherches sur la philosophie et la kabbale dans la pensee juive du Moyen Age (Paris: Mouton, 1962). The research on Ibn Tibbon and his impact was greatly extended by Aviezer Ravitzky. See Aviezer Ravitzky, “The Thought of Rabbi Zerahyah b. Isaac b. Shaltiel and Maimonidean-Tibbonian Philosophy in the Thir teenth Century” (PhD diss., Hebrew University, Jerusalem, 1978) [Hebrew]; idem, “The Possibility of Existence and its Accidentality in Thirteenth-Century Maimonidean Inter pretation,” Daat 2–3 (1978–9), 67–97; idem, “Samuel Ibn Tibbon and the Esoteric Charac ter of the Guide of the Perplexed,” AJSR 6 (1981), 87–123; idem, “The Hypostasis of Divine Wisdom,” Italia 3 (1981), 7–38; idem, “Aristotle’s Meteorology and Maimonedean Exegesis of the Account of Creation,” Jerusalem Studies in Jewish Thought 9 (1990), 225–250; idem, medieval rationalist exegesis and intertextuality 245

In this article, I would like to focus on an aspect that has not drawn much attention so far, namely his approach to Bible exegesis and some original hermeneutical assumptions that guided his work.5 The originality of these hermeneutic guidelines can, in fact, be contested. Some of these assumptions and presuppositions can be found already in the writings of Chazal, namely the sages of the Talmud and the authors of the homiletic literature, and in the writings of his rationalistic predecessors, mainly in Maimonides’ writings and in the Bible exegesis of Rabbi Abraham Ibn Ezra. Ibn Tibbon’s original contribution, as I hope to show, lies mainly in the attempt to incorporate these ‘not so new’ hermeneutic premises into a systematic theory of interpretation, and in the systematic implementation of these guidelines in his exegesis. For want of better, I shall describe Ibn Tibbon’s theory as an intertextual approach to Bible exegesis, firstly because of the anachronism in applying this modern term to the approach of a medieval scholar, and secondly because of a fundamental difference in a basic point de départ that separates Ibn Tibbon’s and Chazal’s intertextual approach from the general understanding of the concept in modern literary theory.6 Whereas intertextuality in the broad sense views the text as a platform for a variety of voices in conversation with each other, into which other texts are imbedded or quoted7 (or in Julia Kristeva’s words: “any text is a mosaic of quotations, any text is the absorption and transformation of another”), the traditional Pharisaic approach regards Scripture, mainly the Pentateuch, as divine or of divine inspiration.8

“The Secrets of the Guide of the Perplexed Between the Thirteenth and the Twentieth Cen tury,” in Studies in Maimonides, ed. Isadore Twersky (Cambridge, MA: Harvard University Press, 1990), 159–207. 5 Samuel Ibn Tibbon’s original works include a long commentary on Ecclesiastes, pub lished by James T. Robinson. A scientific edition of the Hebrew version is included in his “Samuel Ibn Tibbon’s Commentary on Ecclesiastes” (PhD diss., Harvard University, 2002) and an English translation was published by him under the same title, Samuel Ibn Tibbon’s Commentary on Ecclesiastes: The Book of the Soul of the Man (TSMJ 20; Tübingen: Mohr Siebeck, 2007). Ibn Tibbon’s major philosophical work is Ma’amar Yiqqawu ha-Mayin, first published by Mordechai Bislichis (Pressburg: Anton Edlen von Schmid, 1837). A new scien tific edition is included in Rebecca Kneller-Rowe, “Samuel Ibn Tibbon’s Ma’amar Yiqqawu ha-Mayim: A Philosophical and Exegetical Treatise” (PhD diss., Tel Aviv University, 2011). 6 For the complexity in using the term intertextuality, see Hibbard, Intertextuality in Isaiah 24–27, 1–21. 7 This definition is based on LaNeel Tanner, The Book of Psalms, 1. 8 Tori Moi & Julia Kristeva, “Word Dialogue and Novel,” in Tori Moi & Julia Kristeva, The Kristeva Reader (New York: Columbia University Press, 2001), 37. See also Jonathan 246 rebecca kneller-rowe

Yet, dialogical interactive aspects were ascribed to the Biblical texts and integrated into traditional exegesis. The Sages of the Talmud and midrash as well as medieval Bible scholars, as we shall see later, adopted certain premises that have much in common with aspects of modern theories of intertextuality, such as, for example, viewing texts as bearing multiple meanings or allowing for a reader-centered approach to the texts. The similarity in the approaches to interpretation between ancient and medieval Bible scholars and modern interpretative theory rests, however, on different grounds. The plurality of meaning and the freedom of the commentator in Jewish traditional exegesis are based, perhaps paradoxi- cally, on the perception of the divine origin of the texts. Precisely because these texts were regarded as God given or divinely inspired, they must contain, according to the traditionalists, all true teachings for its own time and for generations to come. Hence any true thought or theory can be read into it, and any verse or word can contain a key to the understanding of another passage. Another basic assumption, related to the previous one, underlying the multiple interpretations, is the notion of the multi- facetedness of revelation. This notion implies that Scripture, being divine in origin, contains a priori numerous facets. Exegesis, thus, is not about discovering the sole ‘true’ intention of the author, but about revealing more facets or selecting one interpretative option over others.9 In this essay, I shall also claim that as different as the social, political and cultural environment was for Jews living in medieval Spain and Southern Europe from the conditions in the time of Chazal, these contextual differences expressed themselves mainly in their approach to exegesis, in the style and methodology, rather than in the basic premises regarding the intertextual character of Scripture. At the start of this contribution, I shall devote a few paragraphs to intertextuality in Chazal exegesis, in order to better evaluate Ibn Tibbon’s contribution and originality on the one hand, as well as his indebtedness to his predecessors on the other hand. I shall continue with a short description of the medieval intellectual climate and the challenges it posed to Bible scholars of that period. Finally, I shall try to outline Ibn Tibbon’s response to these

Culler’s definition: “a work exists between and among other texts, through its relations to them” (Literary Theory [Oxford/New York: Oxford University Press, 1997], 34). 9 The multiplicity of interpretations, says Moshe Halberthal, stems from the multiplic ity of meanings and not from the lack of it or from the de-constructive undermining of meaning. For an extensive discussion of basic guidelines in Midrash exegesis, see Moshe Halberthal, Interpretative Revolutions in the Making: Ethic Values as Considerations in the Interpretation of Midrash Halacha (Jerusalem: Magness Press, 1997), 168–203 [Hebrew]. medieval rationalist exegesis and intertextuality 247 challenges, as expressed in his theory of interpretation, in the hermeneutic principles that guided his exegesis and in the methodical implementation of these guidelines in his reading of biblical passages.

1. If “any text is the absorption and transformation of another”, as Kristeva defines it, then the exegetical text will be intertextual par excellence, absorbing and transforming the authoritative source text, adapting it to a different context or to different readers.10 If any text is a grid where prior communications interconnect to form new readings, an interpretation is a revitalized version, allowing for creativity and unraveling conscious or hidden intertexts. In this sense, Scripture and all its interpretations are in dialogue with each other, with those of their own times as well as across the generations. Despite the Sages’ perception of the divine authorship of Scripture, their exegesis reflects an acute awareness of the freedom of the scholar to interpret texts according to his understanding and/or according to the cultural context. Their approach to texts echoes the reader- centered hermeneutic approach where the production of meaning is the work of the reader.11 A famous passage relating to this issue is the ‘story of the stove of Ach- nai’ in B. (= Babylonian Talmud) Baba Mezia 59:1–2. In this story, the Sages argued over the halakhic purity of a stove. Rabbi Eliezer called it pure and the Sages called it impure. R. Eliezer supported his arguments with conclusive proofs but his position was not accepted. He said, ‘If the law is as I say, this carob will prove it’ and the carob was uprooted from its roots 100 feet [. . .]. But the sages said to him, ‘One does not quote a carob as proof’. He then said, ‘If the law is as I say, the water pipe will prove it’, and the water began to flow backwards. Again the sages said, ‘One does not quote the water pipe as proof’. He then said, ‘If the law is as I say, the walls of the house of studies will prove it’, and the wall began to lean over [. . .]. Again they said, ‘One does not quote a wall as proof.’ He then said, ‘If the law is as I say, let it be proven from heaven.’ A voice went out and said, ‘What are you next to Rabbi Eliezer, according to whom the law is in every place.’ Upon that, Rabbi Yehoshua stood on his feet and said, ‘It is not in Heaven’ (Deut 30:11–12). And Rabbi Yirmiya explained, ‘Since the Torah had already been given from Mount Sinai, we do not pay attention to heavenly voices, for it is written there ‘Incline after the majority’ (Ex 23:2).12

10 See Fishbane, Biblical Interpretation in Ancient Israel, 3, 231. 11 This perception echoes Roland Barthes, “The Death of the Author,” in Roland Bar thes, Image, Music, Text, transl. Stephen Heath (New York: Hill & Wang, 1977), 142–148. 12 B. Baba Mezia 59:1–2. 248 rebecca kneller-rowe

After that, the Talmud continues, Rabbi Nathan met the prophet Elijah and asked him, “What was the Holy One, God, doing, while the Rabbis were arguing?” and Elijah answered, The Lord was “laughing and saying, ‘My children have defeated me, my children have defeated me’.”13 Among the messages contained in this story, first and foremost is the clear understanding of the Sages of the independence of the interpreter to read his understanding into the texts rather than submit to authoritative signs and omens believed to confirm the Author’s intent. According to their understanding, God has abdicated his right over the texts of the Torah after Sinai, and conceded it to the humans.14 From then on, the Torah is the interpretation of it15 by the qualified.16 This story also contains an example of the liberty of interpretation assumed by the Rabbis. The verse from Deuteronomy saying the Torah ‘is not in heaven’ was originally pronounced in the context of God implor- ing his people to guard his commandments, telling them that they are not in heaven—metaphorically, in the sense of being beyond their reach or capacity. In the story quoted here above, Rabbi Yehoshua is citing the same verse in its literal sense, to legitimize the appropriation by the sages of the right to determine the law even in face of heavenly signs to the contrary.17 Free re-contextualization of verses and quotations to support ideas clearly different to the meaning in the original context, or for that matter to what was taken to be divine words, was a legitimate exegetical tool in

אשכחיה רבי נתן לאליהו א"ל מאי עביד קוב"ה בההיא שעתא א"ל קא חייך ואמר נצחוני 13 בני נצחוני בני. 14 B. Eruvin 13:2 brings a similar story where freedom is granted to the scholars to deter mine the law, on the basis of contextual considerations, even where powerful reasoning pointed to another solution: “R. Aha b. Hanina said: It is revealed and known before Him Who spoke and the world came into existence, that in the generation of R. Meir there was none equal to him; then why was not the halachah fixed in agreement with his views? Because his colleagues could not fathom the depths of his mind.” 15 Various passages in the Talmud and Midrash depict God as studying the words of the Sages and determining the law according to their conclusions. Cf. Halberthal, Interpreta tive Revolutions in the Making, 200–201 and note 38. 16 Another point implied in this story is expressed by R. Yirmiya, who says that the determination of the law depends on the majority opinion. He is referring to the fact that alongside the recognition of the freedom of interpretation—Jewish tradition puts limits to this freedom. It is circumscribed and delineated by rules and regulation such as the Rule of the Majority, or the famous 13 Midot—the 13 hermeneutical principles of Rabbi Yishmael. Cf. Introduction to Midrash Sifra on Leviticus. 17 Cf. Isaac Heineman, Darchei Ha-Aggadah (Jerusalem: Magnes Press, 1974), 11–12 [Hebrew]. medieval rationalist exegesis and intertextuality 249 the hermeneutics of Chazal. In fact, this free use of sources, the association of words or verses based on phonetic or other similarities, multiple interpretations, at times, even contradictory—characteristic of midrash18 exegesis—were considered by some as a lack of interpretative capability.19 It was Daniel Boyarin20 who reframed it within modern literary theory as the intertextual character of midrash exegesis. According to Boyarin, the midrash’s seemingly random association of words and verses from all different parts of Scripture actually represents the understanding that all parts of the Bible are interconnected and that therefore any part can contain keys to the understanding of, or shed light and elaborate on passages from other parts. Precisely because the Bible was seen by the sages as a divine revelation, they saw the texts as an inexhaustible source of truth and wisdom. Every possible enlightening thought read into one verse could, therefore, be re-contextualized to elaborate or interpret another Biblical passage. An eloquent metaphor to the numerous possibilities of interpretation due to the unlimitedness of the source can be found in B. Sanhedrin 34:1, that tells, “In R. Ishmael’s school, it was taught: ‘like a hammer that brea- keth the rock in pieces’ (Jer 23:29), i.e., just as [the rock] is split into many splinters, so also may one biblical verse convey many teachings.” Rabbi Ishmael’s rock, splitting potentially into an infinite number of splinters, represents his perception that it is the specificity of the divine Word to contain multiple facets.

18 Homiletic literature, most of it written by the end of the 7th century c.e. 19 See a discussion of the attitude to Midrash and an enumeration of critical quotations by various central Rabbinic figures in Abraham Heschel, Torah from Heaven in Perspective of the Generations (Vol 1; London/New York: Soncino Press, 1962), xxv–xxviii; Abraham Geiger, “Das Verhältniß des natürlichen Schriftsinnes zur thalmudischen Schriftdeutung,” WZJT 5 (1844), 53–81 (81). See also Maimonides’ reference to denigrating attitudes to Midrash in Guide of the Perplexed III:43 (transl. Shlomo Pines; Chicago: University of Chi cago Press, 1963, 572–573; all references in this article to the Guide of the Perplexed are to this edition). In this chapter, Maimonides qualifies the Midrash as poetical conceit. How ever, in the introduction to the Guide (20), he includes the Midrash among the texts that contain contradictions of the seventh type, using this device for the purpose of concealing the inner most secrets from the ordinary reader. For a general discussion of attitude to Midrash literature, see Julius Guttman, Philosophies of Judaism (New York: Holt, Rinehart & Winston, 1964), 388–389; Heineman, Darchei Ha-Aggadah, 1–14, 186–195; and Boyarin, Intertextuality and the Reading of Midrash, 13–34. 20 Cf. Boyarin, Intertextuality and the Reading of Midrash; idem, “Old Wine in New Bottles,” 539–556. 250 rebecca kneller-rowe

This understanding appears already in the Mishnah Avot 5:25: “Turn it [the Torah] and turn it again, for everything can be found therein,” as well as in the relatively late Midrash Pesikta Rabati Ecclesiastes 12:11,21 Why are the words of the wise Kedorbanot, like goads that direct the cattle to plough in the farrow. [. . .] Rabbi Berechiah answered Kadur shel banot, like the ball girls play with. One tosses it here and another tosses it there, so are the Sages who come to the house of study, one stating his view and another stating his view, still another stating his view and another stating a different view. Yet the words of these and of the other sages, all of them were given by Moses, the shepherd, from what he received from the Unique One of the universe, ‘They are given from one shepherd’ (Eccl 12:11). This passage, like the story of the stove of Achnai, contains a theoretical statement as well as a sample of the midrash exegetical technique. The theoretical statement giving legitimacy to the multiple readings of the sages, all of them, according to R. Berechia, unfold in some way God’s words transmitted to Moses in Sinai. At the same time, we have here an example of word play, based on phonetics, on the similar sound of Kedor banot (‘like goads’) and Kadur shel banot (‘a ball girls play with’).22 Related to the intertextual character of Scripture that underlies midrash exegesis, there is another presupposition, more limited in scope, according to which later sections of the Bible, namely the prophetic and scribal parts, contain keys to the interpretations of passages in the earlier books. As Gerald Bruns put it,23 different parts of the Bible “are made to relate to one another reflexively, with later texts, for example, throwing light on the earlier, even as they themselves always stand in the light of what precedes and follows them.” Thus, for example, Song of Songs was taken to be an interpretation and elaboration on the story of Exodus,24 and the book of Chronicles a homiletical reworking of the book of Kings.25

21 Pesikta Rabati (transl. William G. Braude; Yale Judaica Series 18; New Haven/London: Yale University Press, 1968), 3:2. The translation is partially based on this edition and modi fied for additional clarity). 22 See also B. Eruvin 13:2 relating to the prolonged dispute between Beth Shammai and Beth Hillel, till a heavenly voice announced ‘[The utterances of] both are the words of the living God’. 23 Bruns, “Midrash and Allegory,” 626–627. 24 See for example Midrash Shir HaShirim Rabbah 2:15–16, 19, 30–31; Mechilta DeRabbi Yishmael on Exod 14:10 or on Exod 19:16 (ed. Saul Horowitz & Abraham Rabin; Jerusalem: Whorman Books, 1970), 91, 94 and 215. Cf. also Boyarin, Intertextuality and the Reading of Midrash, 169–184. 25 Fishbane, Biblical Interpretation in Ancient Israel, 5 and note 13. medieval rationalist exegesis and intertextuality 251

Midrash Canticles Rabbah 1:1 frames Solomon’s exegetical project within this hermeneutic context, comparing it metaphorically to the provision of water/livelihood/Torah to people incapable of attaining it themselves. Until the advent of Solomon, people had no access to the words of Torah. And since Solomon interpreted them, everyone started studying the Torah. [. . .] It can be compared to a deep well full of fresh sweet water no one could drink from, until someone came and attached cord to cord, pulling to pulling and drew the water from the well. Since then, all started drawing and drinking. Thus, from word to word, from fable to parable, Solomon decoded the secrets of the Torah.26 Bruns quotes another midrash on Cant 1:10, where Rabbi Akiva asks Ben Azai why fire is flashing around him. Is it because he is studying the subject of the Chariots? Ben Azai answers, “I was sitting and linking the words of the Torah, and from the Torah to the Prophets, and from the Prophets to the Scribes, and the words were as happy as on the day they were given at Sinai.”27 The same idea is also expressed in the Jerusalem Talmud Rosh Hashana 58:4, where it says, “Words of Torah are rich (of meaning) in one place and poor in another,” as well as in the Mechilta de-Rabbi Yishmael: “This Scripture is enriched from many places.”28 These quotations are but a small selection of examples giving an insight into the rich and colorful intertextual character of midrash exegesis,29 brought here as a background against which to compare the approach of Samuel Ibn Tibbon to exegesis.

2. Samuel Ibn Tibbon’s approach to Bible interpretation is representative, in many ways, of the rationalistic exegesis that flourished in Spain and southern France during the 11th–15th century. As against the narrative character, the multiple interpretations and the loose contact between the interpreting and the interpreted texts in Chazal exegesis, medieval culture was rigorous and logo-centric. The rationalist commentator sought to de-robe the biblical stories and the anthropomorphic expressions of their literal meaning, trying

26 Shir HaShirim Rabbah 1:8, (ed: Sh. Donski; Tel Aviv: x, 1980), 15–17. 27 Shir HaShirim Rabbah 1:52, 125. 28 Mechilta d’Rabbi Ismael, ed. Saul Chaim Horowitz and Israel Abraham Rabin (Jeru salem: Wahrman Books, 19702), chap. 4. 29 There is no systematic discussion of hermeneutics in Chazal, says Halberthal. Their writings, however, contain numerous reflective utterances on their exegesis (Halberthal, Interpretative Revolutions in the Making, 197). 252 rebecca kneller-rowe to uncover the abstract, universal message, believed to be underlying the text. Whereas Chazal exegesis sought to fill in details, to add associations, enrich and illuminate the Biblical texts, medieval rationalists regarded the colorful, concrete language of the Scriptures as symbols and metaphors to be decoded. Their scholarship was diachronic rather than synchronic, scrutinizing texts in search of the ‘true’ meaning30 and the exact authorship of the investigated topos.31 The leading representatives of the rationalist current, from Solomon Ibn Gabirol (11th century c.e.) to Isaac Abrabanel (died 1521), were deeply rooted in the study of Torah and equally immersed in the Greco-Arabic culture of their time. Aristotelianism in its Neo-Platonic garb was accepted as demonstrative truths as far as the physical sciences were concerned, and to a lower degree of certitude as to metaphysics. The existence, in this bi-polar cultural environment, where the double truth theory attributed to Latin Averroists (Sieger of Brabant, John of Jandum) was most marginal, bred tensions that demanded solutions.32 Reconciling the anthropomorphic picturesque language of the Scriptures with the rational abstract thinking of philosophy was mandatory.

30 In the alternate introduction to his Commentary on Genesis (28), Ibn Ezra claims that the Sages actually differentiated between the straight forward evident meaning of the texts and their own far-fetched interpretations. Their ‘fanciful homiletics’, as Morde chai Cohen describes them, were never meant as serious exegesis. They were intended as an asmachta, as a hint, a reminder or as a supportive argument (Mordechai Cohen, “A Talmudist’s Halakhic Hermeneutics: A New Understanding of Maimonides’ Principle of Peshat Primacy,” JSIJ 10 [2012]: 257–359 [272]). See also infra, note 45. As for Medieval Rationalistic exegesis, although it was not a characteristic feature, the search for the ‘true’ meaning did not exclude plurality of interpretations. It was justified either by the difference of approach, as for example in Ibn Ezra’s four approaches to Bible exegesis (Commentary on Genesis, 28–29), as inherent to the multi layered textual approach in Maimonides’ esoteric equivocality (Guide of the Perplexed, 3–20), or as com plementary interpretations (Maimonides’ Kovetz Teshuvot ha-Rambam ve-Igrotav, vol. 2; Leipzig: L. Scheneweiss, 1859), 24:1. On multiple interpretations in Maimonidean herme neutics, see Shalom Rosenberg, “On Bible Exegesis in the Guide of the Perplexed,” Mekha rei Yerushalayin be-Mahshevet Israel 1 (1981–2), 85–157 (135, note 73) [Hebrew], and Levy, Hermeneutics, 45–46. Multi-vocality, says Levy, was an inherent quality of Biblical texts for Medieval Rationalists (Hermeneutics, 44). 31 Ibn Tibbon evokes the legitimacy of multiple interpretations to conceal his criticism of Maimonides’ position (Commentary on Ecclesiastes, Hebrew version in Robinson, Sam uel Ibn Tibbon’s Commentary on Ecclesiastes, 2002, § 396 and English version in Robinson, Samuel Ibn Tibbon’s Commentary on Ecclesiastes, 2007, § 369 [henceforth H/E]). 32 The most noted text representing the Averroistic double truth theory in Jewish Medieval philosophy is Isaac Albalag’s Tikkun ha-De’ot, ed. George Vajda (Jerusalem: Israel Academy of Sciences, 1973). Cf. also George Vajda, Isaac Albalag: Averroïste juif, traducteur et annotateur d’al-Ghazali (Paris: J. Vrin, 1960). medieval rationalist exegesis and intertextuality 253

The way out of this conundrum was through resorting to the double layered conception of the Scriptures, known in Judaism and in the Greco- Arabic tradition since antiquity. According to this concept, the literal, straight forward meaning of the biblical texts addresses the uneducated masses, while the hidden, esoteric layer coincides with demonstrative knowledge, and addresses the intellectual few only. Mishnah Hagiga 2:1 already mentions esoteric realms to be concealed from the masses, namely, the Account of Creation and the secrets of the Merkava (the Chariot). According to the Sages, the obligation of concealment was prescribed already by the Scriptures, as for example in the verse: “It is the glory of God to conceal a thing” (Prov 25:2), or in Ps 25:14: “The secret of the Lord is with them that fear Him.” Esotericism, as mentioned, was known already to the Greeks and the Arabs, according to whom, philosophy, mainly metaphysics, should be divulged to the philosophically trained only. This concept is ascribed to Pythagoras in the 6th century bce. It is mentioned in Plato’s writings and is further elaborated in Alfarabi’s philosophy.33 Mystical currents, Jewish and others, also contributed to the wide-spread of the understanding that Scriptures contain a secret layer. The fact that neither the Mishnah nor the Talmud give a clear indication as to the contents of the ‘secrets’, allowed medieval rationalist, Maimonides being the most prominent amongst them, to identify the esoteric level with Aristotelian science and philosophy.34 The Account of Creation was thus identified with physics, and the secrets of the Merkava with metaphysics.35

33 For an extensive analysis of esotericism and its background, see Sara Klein-Braslavy, King Solomon and Philosophical Esotericism in the Thought of Maimonides (Jerusalem: Magnes Press, 1997) [Hebrew]. 34 Cf. Klein-Braslavy, King Solomon and Philosophical Esotericism, 38–48. 35 Aristotle defines metaphysics in three different ways, as a science concerned with the unmoved eternal intellects or theology (Met. VI,1.1026a10–22; XI,7.1064a31–1064b5; Phys. II,2.194b14–15), as a science of the being qua being (Met. IV,1.1003a22–25; IV,2.1003b21–24; VI,1.1025b3–5; XI,7.1063b) and as a science of the principles of the other sciences (Met. IV,7.1064b6–13; XI,1.1059a18–5.1062b11). See Alexander Altman, “Maimonides on the Intel lect and the Scope of Metaphysics,” in Alexander Altman, Von der mittelalterlichen zur modernen Aufklärung: Studien zur jüdischen Geistesgeschichte (Tübingen: Mohr Siebeck, 1987), 92–106. In his early treatise on logic, Millot HaHigayon (transl. Moses Ibn Tibbon, ed. Leon Roth & David Z. Bennet [Jerusalem: Magnes Press, 1965], 73), Maimonides defines metaphysics as the science that studies the eternal beings and as the study of the prin ciples of the sciences. In his Guide of the Perplexed (I:34, 77, and I:62, 152), Maimonides restricts the definition to the theological aspect, namely to the Chariot or the Divine Sci ence of the Lord and the Separate Intellects (cf. Altman, idem, 107–109). 254 rebecca kneller-rowe

Regarding Scripture as an equivocal text was the literary device rationalist hermeneutics adopted in order to reconcile biblical language and philosophy.36 The double layered concept of Scripture allowed medieval rationalists to anchor their innovative ideas in the source texts, proving their loyalty to ancient tradition on the one hand, and curbing accusations of subversiveness, on the other hand. In their biblical exegesis, medieval rationalists performed a double intertextual demarche; they read Aristotle’s Physics and Metaphysics into the biblical subtexts and they ascribed this understanding to the (presumed) authors of the Scriptures: Moses, David, Solomon, and the succeeding prophets.

3. If indeed this projection was one of the basic hermeneutic guidelines of all medieval rational exegetes, Samuel Ibn Tibbon was the one who formulated it in the most systematic and explicit words. In the last chapter of his book Ma’amar Yiqqawu ha-Mayim, Samuel Ibn Tibbon analyzes the dynamics of biblical interpretation. In the end note to the book, he apolo- gizes for revealing secrets his predecessors had concealed and presents a generalized theory describing the work of the commentator as a gradual, prudent process of unveiling the esoteric layer of the biblical texts, by the leading spiritual authorities, in accordance with the intellectual capacity of every generation. And we shall say that there were not in those times [of the tower of Babel] people admitting the existence of another deity but for some of the luminaries. Also in the days of Moses, of blessed memory, they were so. As Pharaoh said, ‘Who is the Lord . . . I know not the Lord’ (Exod 5:2), therefore Moses, of blessed memory, had to conceal many elements of the faith and present them differently to what they are. Until the veiling and the misrepresenta- tion will bring people over to the belief in the existence of God. The concealment was not meant for any other purpose. There is no doubt that he [Moses] transmitted [the secrets] orally to Joshua and to the seventy Elders together with the rest of the oral tradition. And he [Moses] also incorporated hints in many places, to alert the Sages in a veiled manner. [. . .] Until David and Solomon, of blessed memory, came along and added clues to those mysteries, seeing that the need for concealment had dimin- ished, for the belief in the existence of God had spread, while those who understood the original veiled clues, are less numerous. [. . .]

36 For a parallel in Islam, where esotericism served to bridge between the dominant Aristotelian culture and Koran, see Louis Gardet & George Anawati, Introduction à la théologie musulmane: Essai de théologie comparée (Paris: J. Vrin, 1970), 321–324. medieval rationalist exegesis and intertextuality 255

And so [did], the Rabbi, the Master, the great Sage, the Philosopher, the Torah scholar, the Divine, our Rabbi Moses, son of the great Rabbi Maimon. [. . .] And myself, young, coming after him, seeing that those understanding his clues are fewer, all the more so, those who understand the clues in the Scriptures, and seeing that the true sciences have spread greatly among the Christians,37 under whom we are living, even more than their diffusion in the Moslem lands. And I saw the great need to illuminate the intellectu- ally able with what the Lord had granted me to know and understand His words.38 What Ibn Tibbon is describing here, is a ‘chain of tradition’ as James Rob- inson has named the theory.39 Moses wrote the Torah in an anthropomorphic language, in metaphors, riddles and allegories, in order to spread the belief in God amongst the masses incapable of absorbing the abstract notion of a transcendent, invisible Deity.40 While concealing the esoteric truths from the masses, he inserted into the texts hints and clues in a veiled manner for the select few in future generations.41 Those select few were supposed to decode the secrets and reformulate them in a new garb

37 Although there are no direct documents attesting to Ibn Tibbon’s contacts with Scholasticism, we know (from the introduction to his translation of Aristotle’s Meteorol ogy (Cf. Resianne Fontaine, Otot ha-Shamaim: Samuel Ibn Tibbon’s Hebrew Version of Aris totle’s Meteorology: A Critical Edition, Translation and Index [Leiden/New York/Köln: Brill, 1995], 4) that he spent time in Toledo, one of the important cultural centers at the time, where translators and scholars gathered from all over Europe. Another indication pointing to probable contacts with the cultural environment is the shared interest he manifested in specific, sometimes esoteric, texts and issues, with scholars of his time, mainly with the English ones. For further details on Ibn Tibbon and his contacts with Scholasticism, see Kneller-Rowe, Samuel Ibn Tibbon’s Ma’amar Yiqqawu ha-Mayim, 16–23. 38 Samuel Ibn Tibbon, Ma’amar Yiqqawu ha-Mayim, chap. 22, § 1300–1307 (henceforth MYH). All quotations from MYH are from my edition of the text. The translations are mine. 39 Cf. Samuel Ibn Tibbon’s Commentary on Ecclesiastes, 30–31 (E). This theory is a cre ative elaboration on the tractate in Mishnah Avot 1:1. 40 This claim was articulated already in the Commentary on Ecclesiastes, H § 335–336/E § 309–310. See also Maimonides’ Guide of the Perplexed, 1:63, 164. 41 See also Alfarabi’s view on the gradual uncovering of the truths to the masses through riddles and allegories ever more transparent (Alfãrãbî on the Perfect State: Abu Nasr al-Farabi’s Mabadi Ara’Ahl al Madina al-Fadila [ed. & trans. Richard Walzer; Oxford: Clarendon Press, 1985], 276–285; idem, The Political Regime: Kitãb al-Siyãsã al-madaniyya [ed. Fauzi Najjar; Beirut: Imprimerie Catholique, 1964]). A Hebrew translation of the latter, attributed to Samuel Ibn Tibbon entitled Sefer ha-Hatkhalot can be found in Sefer ha-Asif (ed. Zvi Philipovsky; Leipzig: K.F. Köhler, 1849), 47–49 [Hebrew]. A Modern Hebrew trans lation The Political Society named also Ekronot ha-Nimza’im was made by Shukry Abed (Tel Aviv: Tel Aviv University Press, 1992). On Alfarabi’s impact on medieval Jewish esoterics, see also Klein-Braslavy, King Solomon and Philosophical Esotericism, 19–23. 256 rebecca kneller-rowe that reveals and conceals according to the abstracting capacity of their generation. Thus King David, Solomon and other prophets re-dressed Mosaic laconic passages in riddles and allegories that were more transparent. So did the Sages, so did Maimonides and so does he, Samuel Ibn Tibbon, who saw himself as a link in this chain of the select few rewriters of the secret traditions,42 or as Ibn Tibbon puts it, of those who widen the silver setting covering the apples of gold.43 Ibn Tibbon’s vision of biblical exegesis as a dynamic process of rewriting the texts, albeit his claim that he is only bringing to light ancient hidden truths in danger of disappearing, seems strikingly modern on the one hand, and corresponds, in fact, to a well-established notion of Chazal that “any idea a veteran scholar will ever innovate, was transmitted already to Moses in Sinai.”44 This idea is an extension of the perception of the Torah

42 See a more rudimentary version of this theory in Ibn Tibbon’s Commentary on Eccle siastes, H § 17–23. 43 This metaphor is a word play on Maimonides’ interpretation of Prov 25:11: “A word fitly spoken is like apples of gold in settings of silver.” This verse has been used by Mai monides metaphorically to represent the esoteric exegetical method. The apples of gold represent the inner meaning, the secrets that are precious as gold. The silver settings rep resent the outer layer, that contains the necessary teachings for the masses. These settings mask the secrets on the one hand and allow for some peering through them into the mys teries of the Torah on the other hand. Cf. Maimonides, The Guide of the Perplexed, 11–12. With slight variations, this word play is almost a Leitmotif in the writings of Ibn Tibbon. Ibn Tibbon saw his entire literary work as a project aimed at widening the holes of the settings on the surface of the golden apples, in the sense of revealing more of the secrets while not uncovering them totally. Cf. MYH § 232, 255,710, 762, 766, 781, 787, 1000, 1045, 1086, 1127, 1307, and in his Commentary on Ecclesiastes, E § 19, 20, 48, 279, 308, 338, 359, 471, 753. 756. For additional literature on Ibn Tibbon’s understanding of Scriptures, see Robinson, Samuel Ibn Tibbon’s Commentary on Ecclesiastes, 48 (E); Carlos Fraenkel, From Maimonides to Samuel Ibn Tibbon: From the Dalãlat al-Hã’irîn to the Moreh ha-Nevukhim (Jerusalem: Magnes Press, 2008), 160–162 [Hebrew]; idem, “From Maimonides to Samuel Ibn Tibbon: Interpreting Judaism as a Philosophical Religion,” in Traditions of Maimonid eanism, ed. Carlos Fraenkel (IJS 7; Leiden/Boston: Brill, 2009), 182–190; Aviezer Ravitzky, “Samuel Ibn Tibbon and the Esoteric Character of the Guide of the Perplexed,” AJSR 6 (1981), 87–123. On Samuel Ibn Tibbon and esotericism in medieval rationalistic literature in general and on Ibn Tibbon in particular, see also Moshe Halberthal, Concealment and Revelation: The Secret and its Boundaries in Medieval Jewish Tradition (Essays and Papers in Jewish Studies; Jerusalem: Yeri’ot, 2001), 30–52, 78–85 [Hebrew]. ואפילו מה שהתלמיד ותיק עתיד לומר לפני רבו כלן ,Midrash Rabbah Leviticus 22:1 44 .נאמרו למשה בסיני See also B. Menahot 29:2; Midrash Rabbah Numbers 19:4; Midrash Tanhuma, Hukat 8. The question of the possibility of innovating as against merely revealing the truths con cealed in the Torah was at the heart of an argument, with far reaching implications (for example on Halacha and on the status of the Hebrew language), between Rabbi Akiva and Rabbi Ishmael. Rabbi Akiva was a protagonist of the former opinion, Rabbi Ishmael of the latter. This issue divides halakhic authorities throughout the generations. For additional sources and a discussion on the topic, see Heschel, Torah from Heaven, 5–23. medieval rationalist exegesis and intertextuality 257 as a divine revelation, mentioned above. It legitimized the far-fetched textual support (asmachta)45 frequently used by Chazal and by the Andalu- sian exegesis on the one hand, and the claim that every demonstrated theory can be read into the Scriptures by the medieval rationalists, on the other.46 The view of Bible exegesis as a whole as a trans-generational project of interpretation entailed also a more restricted concept of intertextuality, mentioned earlier in connection with Chazal, and that is far more elaborate in the exegesis of Samuel Ibn Tibbon. Beyond the general conscious and subconscious interconnectedness between the texts described above, there is the assumption that the Bible is not a mere collection of the most sacred writings to Jewish tradition. There is, according to this understanding, an intentional interpretative relationship between the different parts of the Scriptures, where later books contain systematic elaborations and interpretations of obscure passages in the earlier parts. Solomon’s books, Proverbs, Ecclesiastes, and the Song of Songs, for example, says Ibn Tib- bon, were written with the aim of unveiling the esoteric subtext contained in Genesis 1–9, relating to the Original Sin of Adam and Eve, to the Garden of Eden and to the Tree of Life. Psalm 104, continues Ibn Tibbon, was written as a commentary on the Account of Creation in Genesis 1. And when I (Ibn Tibbon) saw that most of the covert indications of David, peace be on him, concerning what preceded the generation of the world and concerning the six days of creation [. . .] one of [the related issues being] the appearance of the earth [. . .] is [included in] the thirteen verses; four—at the end of chapter 103 and nine at the beginning of chapter 104—I ventured to interpret them for you in this monograph.47 The perception of the interpretative connection between Psalm 104 and the Story of Creation was noted already by Chazal in Midrash Exodus Rab bah, 15:22: Many deeds were written by Moses in a concealed way, until David came and interpreted them. We find in the story of Genesis, after He created Heaven and Earth, He created light, as it says “In the beginning the Lord

45 On the status of asmachta, whether an arbitrary aide mémoire or a sign/hint pointing to a content imbedded in the text, see Judah Hallevi, Kuzari (transl. Hartwig Hirschfeld; New York: Schocken Books, 1946), 3:73 [Hebrew]; Heschel, Torah from Heaven, 20–23, and Cohen, “A Talmudist’s Halakhic Hermeneutics,” in particular 282–302. 46 Cf. Maimonides, Guide of the Perplexed 2:25, 330, and Rosenberg, “On Bible Exege sis in the Guide of the Perplexed,” 95, 149–152. Fishbane describes the ascription of any remote exegetical idea to Scriptures as divinization of exegesis (Biblical Interpretation in Ancient Israel, 4). 47 MYH § 65. 258 rebecca kneller-rowe

created [Heaven and Earth]’ and later [it says,] ‘and the Lord said let there be Light’ (Gen 1:3). And David interpreted, after He created Light, He created the Heavens, as it says “Who coverest thyself with Light as with a garment: who stretchest out the Heavens like a curtain’ (Ps 104:2).48 The midrash is telling here that in David’s change of order between the creation of the Heavens and the Light in Psalm 104, as against the narrative in Genesis 1, lies an interpretation or a revised version of the story, revealing facts that Moses saw right to conceal. It hints to the fact that the light was an intermediary in the formation of the sub-lunar world and that the four elements existed prior to creation, as the midrash continues: “From here we learn that after He created Light, He created the Heavens. Three entities preceded the [creation of the] world, the Waters, the Wind and the Fire.” Surprisingly enough, Ibn Tibbon does not mention this midrash, nor does he quote Judah Hallevi (1075–1141), who also interprets Psalm 104 as a commentary on the story of Creation in light of Aristotelian physics.49 This silence leaves a question mark as to whether or not he was familiar with these particular passages50 or that he was simply silent about his sources, as was common in his time. Whether he knew these earlier texts or not, Ibn Tibbon acknowledges his indebtedness to Rabbi Abraham Ibn Ezra (1089–1164) and his exegesis.51 The interpretive connection between Psalm 104 and Genesis 1 is mentioned already, albeit succinctly, in Ibn Ezra’s commentary: “For in each day only one thing was created; in the first day the light, in the second the heavens, in the third the plants [. . .] and the above mentioned Psalm [104] is witness to it.”52 There are a considerable number of parallels between their cosmogonies as well as quotations from Psalm 104, in

48 Shemot Rabba 15:22. 49 Hallevi, Kuzari, 5:10. 50 That Samuel Ibn Tibbon knew some of Judah Hallevi’s work is apparent from Ibn Tibbon’s allusion to one of Hallevi’s poems in MYH § 263. 51 Aside from Maimonides, Ibn Ezra is the most frequently quoted authority in Ibn Tib bon’s writings, see MYH § 61, 66, 81, 111, 117, 135, 161, 225, 264, 394, 449, 467, 619, 686. 919, 920, 926, 947, 1178, 1221–1222. In § 111, 117, 225. Commentary on Ecclesiastes H § 243, 252,262, 266, 287. 294, 332, 347, 358, 385, 387. E § 162, 220, 229, 239, 243, 262, 320, 331, 358, 741. 52 Abraham Ibn Ezra, Commentary on the Pentateuch, Genesis (Mikra’ot Gedolot ha- Keter, Gen Vol. 1; Ramat Gan: Bar-Ilan University Press, 2001), 4. See also his Commentary on Psalms (Mikra’ot Gedolot ha-Keter, Psalms Vol. 2, Ramat Gan: Bar-Ilan University Press, 2004), 104 on Ps 104:1: “For in this Psalm he will tell the Account of Creation, beginning from the light, for it was created first, followed by the heavens and the earth and the grass and the luminaries, the fish, the animals and man.” medieval rationalist exegesis and intertextuality 259 support of their theory of creation (in Ibn Ezra’s writings less systematically than in those of Ibn Tibbon).53 The concept of interpretive interconnectedness was also shared by Maimonides. In the introduction to his Guide of the Perplexed, he quotes the midrash from Song of Songs Rabbah 1:8, mentioned above, about Solomon’s writings containing keys to the secrets of the Pentateuch.54 Throughout his work, Maimonides quotes verses from Solomon’s writings and from other prophetic books to decode messages or support interpretations believed to be concealed in the biblical narratives.55 However, he does not mention Psalm 104 in this connection. In other words, the notion that prophetic chapters contain keys to the understanding of passages from the Pentateuch or an elaboration on biblical narratives was accepted already by Chazal, as well as by early medieval scholars. Ibn Tibbon’s contribution resides in the extensive and systematic way he implemented this understanding in his exegesis. The most extensive example of this understanding is his verse by verse interpretation of Ps 103:19–104:35 in Ma’amar Yiqqawu ha-Mayim (chapter 20), stretching over almost a quarter of the work, reading into it a most original and daring explanation of the first chapter of Genesis. Samuel Ibn Tibbon tells us he spent some twenty years trying to reconcile the biblical narrative of creation with the precepts of science.56 According to scientific

53 As for example Ps 104:2, “who coverest thyself with light.” See MYH § 922, 1002–1003, and Abraham Ibn Ezra, Commentary on Genesis, 2. See also Ibn Ezra’s commentary to Ps 104:9 (Commentary to Psalms, 106). 54 Cf. Maimonides, The Guide of the Perplexed, 10–14. 55 For example, Maimonides’ interpretation of Adam and Eve’s sin, in Guide of the Per plexed I:2 (23–26), as a turning towards desires of the imagination and corporeal pleasures rather than to the pursuit of the intellectual perfection they were endowed with at their creation. Their punishment was that “the eyes of them both were opened and they knew that they were naked” (Gen 3:7). The opening of the eyes, says Maimonides, refers to a new mental vision, not to an acquisition of a new sense organ that could see things previously invisible, nor to an upgrading of their level of cognition. Adam and Eve’s sin, says Maimo nides, opened them to understanding social norms and ethics, the realm of good and evil, rather than to pure intellectual perception, recognizing truth and falsehood only. This is why it says “and they knew that they were naked” rather than “and they saw.” To support his claim, he quotes the verse from Isa 35:5, “Then the eyes of the blind shall be opened” or the verse from Ezek 12:2, “That have eyes to see and see not.” Another example can be found in Maimonides’ interpretation of Jacob’s dream, where he sees a ladder with angels ascending and descending on it and “the Lord stood above it” (Gen 28:13). Standing, says Maimonides, is an equivocal term that sometimes means being in an erect position, and sometimes, as in Jacob’s dream and in all references to the Deity, means stable, perma nent. In support of this understanding, he quotes the verse from Ps 119:89 “Forever, O Lord, Thy word stands erect in heaven” (Guide of the Perplexed I:15, 40–41). 56 MYH chap. 2, § 11. 260 rebecca kneller-rowe knowledge of his time, the world, in its natural state, was constituted of four concentric elements, where the heaviest, namely earth, was in the center, the waters around it, the air around the water and the fire on the outermost circle. For twenty years, he had been searching for a scientific theory, i.e., for a natural mechanism that could explain how the waters gathered at a particular point in time to allow the earth and the life on it to emerge and produce the universe we live in. Aristotelian physics could explain only partial, temporary transformations between humid and dry areas through the influence of the spheres causing evaporation and inun- dations. Partial exchanges of land and see could be explained also by the transformation of the elements, one to another. Heat and light emanating from the spheres were the central players in the process of generation, but all that could not account for the initial appearance of land after having been totally immersed under water.57 Ibn Tibbon finally found the answer to his query in the theory of congelation and solidification of wet soil into stones and rocks by Avicenna (980–1037).58 This theory allowed for the formation of mountains and valleys, which made the gathering of the waters possible and consequently the formation of the world without resorting to supernatural powers. In Avicenna’s system, he even found a natural explanation based on the influence of the heavenly spheres and the humidity of the earth for the spontaneous generation of life, animal and human.59 With time, according to this theory, the natural process of

57 For an explanation why Aristotelian Physics could not account for the emergence of mountains, see Gad Freudenthal, “(Al)Chemical Foundations for Cosmological Ideas: Ibn Sina on the Geology of an Eternal World,” in Physics, Cosmology and Astronomy: 1300–1700; Tension and Accommodation, ed. Sabetai Unguru (BSPS 126; Dordrecht/Boston/London: Kluwer Academic Publishers, 1991), 47–73. 58 Ibn Tibbon translates from Avicenna’s Kitâb al-Shifâ, at-Tabiyyât 5: Al-ma’âdin wa-l- âtâr al-’ulwiyya, ed. Abd al-Halim Muntasir et al. (Cairo: Organisation Générale Egyptienne, 1964), 24/7–25/9. An English translation of this passage exists in Paul Lettink, Aristotle’s Meteorology and its Reception in the Arab World (Leiden: Brill, 1999), 195–196. Samuel Ibn Tibbon probably knew Avicenna’s geological theory of the solidification of fatty earth into stones and rocks also from his treatise known as De Mineralibus or De congelatione et con glutinatione lapidum: Being Sections of the Kitâb al-shifâ, ed. & transl. Eric J. Holmyard & Desmond C. Mandeville (Paris: Librairie Orientaliste P. Guethner, 1927). This treatise was translated to Latin and attached to Aristotle’s Meteorology by Alfred of Sarashel (12th–13th century c.e., translator and commentator of Aristotle’s works, author of De Motu Cordis). Its authorship, Aristotle’s or Avicenna’s, was long disputed. On the history of this treatise and its identification, see Holmyard & Mandeville’s introduction to De Mineralibus, and James Otte’s introduction to Alfred of Sarashel, Commentary on the Metheora of Aristotle (Leiden: Brill, 1988), 8–29. 59 Cf. Avicenna, Kitâb al-Shifâ, at-Tabiyyât 5, 76–79. I am grateful to Prof. Ahmad Hasnawi for indicating to me this source. See also idem, Kitâb al-Shifâ, at-Tabiyyât 8: medieval rationalist exegesis and intertextuality 261 erosion eventually levels out the mountains and brings the world back to the concentric state covered by water, to re-emerge again in a cyclic pattern ad infinitum. This was Ibn Tibbon’s solution. As far as he was concerned, for the first time science could prove the historical truth of the biblical story of creation. Natural laws could account for the entire process of creation, doing away with any intervention of supernatural forces in the making of the universe, how surprising for an avowed Jew writing in the 13th century.60 All that, says Ibn Tibbon, is inscribed in King David’s Psalms, chapters 103–104: It seems to me that the appearance of dry land mentioned by the philosopher [Avicenna] through the birth of mountains and the depths, is what David alludes to in Psalm 104, which is the Psalm “Bless the Lord, O my soul”, since this Psalm was made by its author solely for the purpose of alluding in it and in the verses preceding it in the Psalm above, to amazing allusions relating to great secrets of creation.61 According to Ibn Tibbon’s interpretation, the last four verses of chapter 103, invoking the Lord’s Angels, his Hosts and his Works to bless their Mas- ter, allude to the role played by the Separate Intellects, represented by the Angels in medieval rationalist vocabulary, and the heavenly spheres and stars, represented by the Hosts, in the bringing about of the universe. Chapter 104, according to Samuel Ibn Tibbon, recounts in fair detail the natural process of creation. The verses “Bless the Lord, O my soul [. . .] Who coverest thyself with light as with a garment” (Ps 104:1–2) refer to the light/heat that lifted the vapours from the water. “Who stretchest out the heavens like a curtain” describes the appearance of the sky, while the next verse “Who layeth the beams of his chambers in the waters: who maketh the clouds his chariot” (vs. 3), tells about the formation of the clouds. In verses 5–8: “Who laid the foundations of the earth [. . .] The waters stood above the mountains. [. . .] At thy rebuke they fled; at the voice of thy thunder they hasted away”, David is referring to the process of the gathering of the waters that were driven away from their natural place by the elemental tendency to transform to one another and by meteorological and geological phenomena that allow for the formation of rocks,

Al-Hâyyâwan, eds. Abd Al-Halim Muntasir et al. (Cairo: Organisation Générale Egyptienne, 1970), 84–85. 60 For a detailed analysis of Ibn Tibbon’s theory of creation, see Freudenthal, “Samuel Ibn Tibbon’s Avicennian Theory of an Eternal World,” Aleph 8 (2008): 41–129, and Kneller- Rowe, Samuel Ibn Tibbon’s Ma’amar Yiqqawu ha-Mayim, 112–160. 61 MYH § 32. 262 rebecca kneller-rowe mountains and valleys.62 As it says, “They go up by the mountains, they go down by the valleys, unto the place that thou has founded for them”. Far- fetched as it may be, Ibn Tibbon continues his interpretation along these lines throughout the chapter and adds various other passages, mainly from Psalms and II Samuel, in support of his understanding of Genesis 1.63 As to the question why the last verses of Psalm 103, referring to the Luminaries’ and the Intellects’ role in creation, were separated from Psalm 104, describing the coming to be of the universe, Ibn Tibbon says, For his [David’s] intention was to hint to their connectedness and point to their separation [. . .] to add allusions to that which the Master of Prophets [Moses] concealed so totally, so that only the just, fearing God, will be able to comprehend them [. . .] while the literal wording will remain close to the way the Master of Prophets left it, as is preferable for the multitudes.64 Moses, says Ibn Tibbon, did not want to reveal the dominant role of the Intellects and the spheres in the process of creation, for fear that people of his time will consider the Luminaries as divine and worship them. For that reason, he did not mention their existence (in Genesis 1) until the fourth day, when the sub-lunar world was almost completed and there was no more danger of the masses ascribing creation to the Luminaries. David, according to Ibn Tibbon, felt time was ripe for some gradual revelation of the secrets, allowing him to hint to the spheres and their role, yet, not overtly. A certain degree of obscurity needed to be maintained, which is why he separated the two Psalms, alluding to the spheres and

62 In MYH § 1083, Ibn Tibbon explains why thunder and wind are mentioned and not the transformation of the elements into one another, nor the solidification of fatty waters into rocks and stones: “It seems that David considered the thunder and the blow ing winds of great assistance in pushing away the waters from above the mountains and for the emergence of the mountains, although he knew that there were other, more powerful causes for these phenomenon. Perhaps he did not want to mention them, and mentioned those [the thunder and the winds] for the purpose of concealment, since the masses recognize the thunder and the lightning as being caused by the Lord, but they do not recognize as such, the other, more essential causes for the appearance of dry land,” ונראה שדעת דוד ע״ה היא שלרעמים ולרוחות הנושבות עזר גדול להסיר המים מעל ההרים או להתילד ההרים, אף על פי שידע בדבר סבות אחרות חזקות מאלו, אפשר שלא רצה להזכירם, וזכר אילו להסתיר לצורך ההמון שמכירים ברעמים ובברקים שהם מאת השם, ואינם מכירם זה .בסבות אחרות שהם עיקר הראות היבשה 63 Ibn Tibbon interprets the following passages as confirming his theory of creation: Ps 18:8–16 (§ 1089, 1096, 1116, 1124–1128. 1192), 90:2 and 95:5 (§ 997), 102:26 (§ 995), 144:2 (§ 1193), 2 Sam 18:8–16 (1089. 1096. 1125–8), Isa 6:4 (§ 1096), 41:6 (§ 1167), Ezek 1:26 (§ 1085). 64 MYH § 900. medieval rationalist exegesis and intertextuality 263 their movers in Psalm 103, and describing the process of generation of the universe in Psalm 104.65 The example brought here above reflects the intertextual view of Scrip- tures and of exegesis that guided Ibn Tibbon in his theory and practice of Bible interpretation. It is but one example of radical re-contextualization of ancient texts, adapting them to the intellectual Zeitgeist of medieval rationalism, in an attempt to reconcile Aristotelian science with traditional Jewish scholarship. In his Commentary on Ecclesiastes, we have another example where he repeats the double intertextual demarche encountered in his interpretation of Psalms 103–104 and the story of Creation. He first translates Solo- mon’s writings into medieval Aristotelian philosophy, following which he proceeds with the claim that this reading was Solomon’s interpretation of Genesis 1–9. In paragraphs 45–56, Ibn Tibbon contends that the first nine chapters of Genesis contain veiled mysteries.66 While their literal message is beneficial for the masses, their inner philosophical meaning conceals three noble issues, and the reference is to the sin of the woman, lured by the serpent, to Adam’s rejection from the Garden of Eden and to the eating from the Tree of Life. Seeing that there hardly were any scholars left capable of deciphering the secretive philosophic subtexts, King Solomon, says Ibn Tibbon, wrote his three books, the book of Proverbs, Ecclesiastes and the Song of Songs, revealing and concealing the imbedded subtexts in the appropriate balance, preventing them from being lost.67 The examples mentioned here above reflect the hermeneutic principle, guiding Ibn Tibbon’s exegesis as well as the approach of other rationalistic

65 In MYH § 1084, Samuel Ibn Tibbon spells out explicitly the reasons for maintaining obscurity, as in the case of the causes of the formation of dry land saying: “Due to the weakness of their cognition, had they [the masses] known that the cause for the appear ance of dry land is the transformation of the other elements into earth, and their drop ping elsewhere and remaining there, and the transformation of parts of the earth to the lighter elements and rising to their new natural place, leaving behind a crevasse, as we mentioned in the name of Avicenna, and all that being caused by the luminaries, i.e. the Hosts, carrying out the will of God, [. . .] they would not have regarded that as God’s work,” אך לחולשת הכרתם, אילו היו יודעים שסבת הראות היבשה הוא השתנות שאר היסודות ונפלם במקום אחר מכדור העפר והשארם במקום ההוא, וכן השתנות חלק מכדור העפר אל אחד מן היסודות הקלים ועלותו אל מקומו לפי טבעו החדש, והניחו במקומו חפירה, על הסדר שזכרנו בשם בן ציני, וכל זה בסבת המאירים, ר״ל הצבאות עושי רצון השם על פי עושי .דברו, לא היו חושבים זה למעשה השם, כי לא ידעו כוונת אמרו 'עושי דברו' ולא 'עושי רצונו' 66 Commentary on Ecclesiastes, H§ 45–46/E§ 41–51. 67 For more examples see MYH § 67, and Commentary on Ecclesiastes H§ 166/E§ 149, 180–187/163–169, 194–206/176–187, 215–217/195–196, 324–347/298–320, 348–349/321–322, 375–380/348–352, 446–450/424–428, 491–492/472–473, 640–654/620–634. 264 rebecca kneller-rowe exegetes, that different passages of Scripture are in dialogue with each other. Clues to the mysteries and fully fledged interpretations of obscure passages of the Torah can be found in the writings of the later prophets and scribes. At the same time, it evokes the question, or rather the suspicion, that these hermeneutic devices served (consciously or sub- consciously) as a cover for the introduction of subversive ideas far and remote from traditional Judaism, under the guise of legitimate interpretations of the sacred texts.68 Ibn Tibbon’s perception of Scripture as interconnected texts, described in the preceding paragraphs, reflects a diachronic approach. The same perception underlies another original hermeneutic principle that guided his exegesis. It expresses itself in a synchronic, integrative approach to biblical narratives according to which, various narratives, considered traditionally as distinct historical events, are in fact a paradigm, representing one and the same topos garbed in different language by different narrators. Thus, the ladder dream of patriarch Jacob (Gen 28:10–22),69 God’s appearance to Moses in the rock (Exod 33:12–23), the Lord sitting on the throne surrounded by the Seraphim in the vision of Isaiah (chapter 6) and the Chariot of Ezekiel (chapter 1) represent varying visual expressions of the paradigm of revelation. Into that same paradigmatic topos, Ibn Tibbon intertwines the story of the Tower of Babel (Gen 11:1–9) and Moses seeing the burning bush (Exod 3:2–6). “There is no distinction between the Tower and the Ladder,” Ibn Tibbon says in Ma’amar Yiqqawu ha-Mayim § 2192–2193. The various visionary chapters represent, according to Ibn Tibbon, an enlightened revelation of the structure of the universe. The almost mystical vision of God is translated by him into an intellectual apprehension of God’s impact on the world, i.e. to the knowledge of the physics of the sub- lunar world, to the astrophysics and to whatever metaphysical knowledge man can attain.70 The rationale for this interpretation, according to Ibn Tibbon, resides in the impossibility of actually seeing the Lord:

68 On the problematic of ‘creative exegesis,’ see Yair Hoffman, “The Technique of Quotation and Citation as an Interpretive Device,” in Creative Biblical Exegesis: Christian and Jewish Hermeneutics Through the Centuries, ed. Benjamin Uffenheimer & Henning G. Reventlow (JSOT.SS 59; Sheffield: JSOT Press, 1988), 71–79. 69 In his article “The Ladder of Ascension,” Altman notes that Ibn Tibbon was the first to count Jacob’s dream among the narratives of revelation (“The Ladder of Ascension,” in Studies in Mysticism and Religion Presented to Gershom Scholem on his Seventieth Birthday by Pupils, Colleagues and Friends, ed. Raphael J.Z. Werblowsky et al. [Jerusalem: Magnes Press, 1967], 1–32 [19–22]). 70 The debate concerning the possibility of attaining knowledge of metaphysics was opened by Shlomo Pines, who argued for the impossibility of attaining metaphysical medieval rationalist exegesis and intertextuality 265

It should not even occur to you that his [Isaiah’s] saying “for mine eyes have seen the King the Lord of hosts” (Isa 6:5), or his saying “I saw also the Lord” (Isa 6:1) contradicts what is said “For there shall no man see Me and live” (Exod 33:20), for it is already explained in the Torah that the vision that is unattainable, is the vision named the vision of the face, whereas the vision named the vision of the back is attainable to man, as exposed by the Master of Prophets [Moses] “and thou shalt see my back” (Exod 33:23).71 The ‘vision of the face’ refers to Maimonides’ Guide of the Perplexed I:37: “But my face shall not be seen, meaning the true reality of my existence as it veritably is, cannot be grasped.”72 In the same way, the ‘vision of the back’ is a reference to Guide of the Perplexed I:38: “And thou shalt see My back, which means that thou shalt apprehend what follows Me, has come to be like Me, and follows necessarily from my will, that is, all the things created by Me.”73 In other words, since no apprehension of the essence of the Deity (vision of face) is possible, the vision of God has to be understood as referring to the apprehension of his deeds (vision of the back). Thus what Ezekiel saw,74 according to Ibn Tibbon, were the four elements constituting the sub-lunar world (the wheels and their work), the surrounding spheres (the living creatures) and the Separate Intellects (the divided man figure). “For the vision intended in this Revelation is the vision of the existents and their relationships to one another [. . .] as did Isaiah, peace be on him, and Jacob, peace be on him, and the other prophets speaking through the Holy Spirit, when speaking on this matter.”75 These visions, according to Ibn Tibbon, overlap yet in another way. Apart from their partaking in the same paradigm of revelation, they contain a common epistemological implication, indicating to man his way to

knowledge. Cf. “On the Limitations of Human Knowledge according to Al-Farabi, Ibn Bajja and Maimonides,” in Studies in Medieval Jewish History and Literature, ed. Isadore Twerksy (Cambridge: Harvard University Press, 1979), 82–109, and idem, “Les limites de la métaphysique selon Al-Fârâbi, Ibn Bajja et Maimonide,” Miscellanea Mediaevalia 13 (1981): 211–225. His position was criticized by Altman, “Maimonides on the Intellect and the Scope of Metaphysics,” 90–91, 107–129, and by Herbert Davidson in his article “Maimonides on Metaphysical Knowledge,” Maimonidean Studies 3 (1995): 49–103. On this topic, see also Warren Harvey, “Maimonides First Commandment, Physics and Doubt,” in Hazon Nahum, ed. Yaakov Elman & Jeffrey Gurock (New York: Yeshiva University Press, 1998), 149–162; idem, “Maimonides’ Critical Epistemology and Guide 2:24,” Aleph 8 (2008): 213–235: Car los Fraenkel, “Maimonides, Averroes and Samuel Ibn Tibbon on a Skandalon of Medieval Science,” in Aleph 8 (2008): 195–212. For an extensive discussion of the debate and more sources, see Kneller-Rowe, “Samuel Ibn Tibbon’s Ma’amar Yiqqawu ha-Mayim,” 273–304. 71 MYH § 211. 72 Guide of the Perplexed, 86. 73 Guide of the Perplexed, 87. 74 Ezekiel 1. 75 MYH § 317. 266 rebecca kneller-rowe intellectual perfection. “After we have spoken of the two visions, meaning the vision of Isaiah and that of Ezekiel, and the Dream of Jacob, as all the three [visions] allude to man’s way to intellectual accomplishment and to achieving final perfection.”76 Contrary to his silence on the earlier sources interpreting Psalms 104 as an exegesis on Genesis 1, Ibn Tibbon acknowledges the influence of Chazal on his adoption of this integrative hermeneutic guideline synthe- sizing the different chapters of revelation. Quoting B. Talmud Hagiga 13:2, where it is said that what Ezekiel saw in the vision of the Chariot, was identical to Isaiah’s revelation, Ibn Tibbon says, “And Ezekiel, peace be on him, withheld [his words] on the topic of the Man, for he relied on Isaiah, who elaborated on it [. . .] but concerning the wheels (the elements) and the living creatures (the spheres), it is to the contrary. Isaiah abridged his reference to them greatly, and Ezekiel elaborated on them extensively.”77 A similar comment appears already in the Commentary on Ecclesiastes:78 “We also find that Isaiah, peace be on him, wrote about the ‘work of the Chariot’ using brief chapter headings which, according to the testimony of the Sages, included everything Ezekiel would subsequently discuss at length.” In the same vein, Ibn Tibbon claims concerning Psalms 49 that “Its author used the method of brevity, alluding to all that was intended in the book of Ecclesiastes. But I do not know who was prior to whom.”79 Samuel Ibn Tibbon seems to extend the identity of the passages to imply some degree of textual coordination between the authors. Where Isaiah elaborated Ezekiel condensed his narrative and vice versa. Where the author of Psalms abridged, the author of Ecclesiastes elaborated. Ibn Tibbon’s identification of Isaiah 6 and Ezekiel 1 allowed him to decipher problematic passages in one prophetic chapter on the basis of

76 MYH § 340–341. In the same way, Ibn Tibbon says: “The Angel (the Active Intellect, referring to the Seraph in Isa 6:6) perfects man’s intellect and brings it from potentiality to actuality” (MYH § 229), is the speaking voice that Ezekiel heard (Ezek 1:28) (MYH § 300) and these are the Angels descending the Ladder in order to perfect the minds of those ascending it [the Ladder]” (Gen 28:12) (MYH § 349. 356). Likewise, see MYH § 234: “And the putting off the shoes there (in the story of the burning bush, Exod 3:5) is identical to the purging of the sin here (Isa 6:7), and thus we found with Jacob who feared when he ונשילת הנעלים שם, הוא ענין הסרת העון, כלומר טומאת השפתים הנה. וכן .(awoke” (Gen 28:17 .מצאנו ביעקב ע״ה, שירא אחר שהקיץ 77 MYH § 325. כן מצאנו ישעיה ע״ה שכתב במעשה מרכבה :Commentary on Ecclesiastes, H § 19/E § 17 78 .ראשי פרקים קצרים כללו כל מה שזכר יחזקאל בארוכה 79 Commentary on Ecclesiastes, E § 465 (with slight emendations for additional clarity)/ כי רמז בו מחברו בדרך קצרה אל המכוון בכל ספר קהלת, ואני איני יודע מי הוא הקודם :H § 465 .מהם medieval rationalist exegesis and intertextuality 267 clearer statements in the other, as for example: “And it brought me to understand the word ‘smoke’ [used] here, as the two vapours, their [the Sages’] saying [. . .] ‘all that Ezekiel saw, Isaiah saw’ and Ezekiel saw the fire scattered over the city from between the Cherubim by the man clothed with linen [. . .] therefore I said that Isaiah took the smoke indicating the fire in lieu of the two vapours.”80 Ibn Tibbon’s identification of the smoke in Isaiah’s vision as an allusion to the two initial vapours on the basis of Ezekiel’s explicit reference to the fire cast on earth (the city) by the sun (the man clothed in linen) reflects the two main presuppositions guiding his exegesis, namely, the view that later texts interpret earlier ones, and the assumption that the different narratives of the metaphysical visions are in fact parallel visuals of one and the same paradigm of revelation.

Conclusion

The preceding pages have provided an insight, however minimal, to medieval philosophic exegesis in general, and a better understanding of Sam- uel Ibn Tibbon’s contribution to Bible interpretation, in particular. The intertextual perspective adopted in this paper imposed itself, in view of its dominant role in Ibn Tibbon’s exegetical theory and practice. It provided, at the same time, a parameter for evaluating Ibn Tibbon’s innovation as against his indebtedness to his predecessors. It supported the conclusion that although medieval rationalism was far apart from the intellectual environment and makeup of Chazal, and in spite of the differences of style and approach, there are similarities in the exegesis, similarities in many of the basic assumptions guiding the Sages on the one hand, and Ibn Tibbon on the other. Whereas midrash endeavors to enrich Biblical narrative, to fill gaps, to illuminate, elaborate or respond to a textual problem, Ibn Tibbon’s rational exegesis addresses the general malaise generated by the clash of cultures. While midrash exegesis does not do away with the literal meaning of the texts, medieval rationalism seeks to replace the material anthropomorphic language of the Scriptures with a demonstrative rational narrative concurring with their cultural discourse. Whereas midrash exegesis uses free associations, anecdotes, parables and interplay of words

80 MYH § 187. 268 rebecca kneller-rowe and verses, medieval rationalism relies on Aristotelian rhetoric devices, especially metaphor and allegory, in order to ‘spiritualize’ and do away with the literal meaning of the texts. Whereas midrash admits multiple interpretations to one narrative or verse, Ibn Tibbon often merges and integrates several narratives into one story. However different the style of midrashic exegesis is from that of Ibn Tibbon, and for that matter of many other rationalists of his time, they share many of the hermeneutic presuppositions, presuppositions that have much in common with the intertextual approach to texts.

– Alongside the premise that Scripture is divinely inspired, and as such, contains all ultimate truths, no true opinion can be in collision with it and every demonstrated theory can be read into it, at the same time—in order to maintain the relevance and its veridical status, continual reinterpretation and re-contextualisation need to be carried on. – Midrashic and Tibbonide exegesis convey a reader-centred approach, assigning to the lector or to the exegete the power to determine the meaning of the text. – Another hermeneutic guideline common to the midrash and to Ibn Tibbon’s exegesis is that all Biblical texts are interconnected in general, and the more restricted assumption that later texts contain keys and interpretations of earlier biblical writings, in particular of the Torah. – And finally, Ibn Tibbon’s integration of different narratives into one somewhat amorphous paradigm is reminiscent of the midrashic a-historic associations of verses and narratives.

Alongside the description of Chazal’s and Ibn Tibbon’s literary oeuvre, the intertextual character of their exegesis points to an implicit awareness on the part of the Sages and to a more sophisticated understanding on the part of Samuel Ibn Tibbon of what is involved in reading, writing and interpreting authoritative texts. The image of God sitting and studying his Torah may be tradition’s most powerful metaphor for the understanding that the viability of Scripture depends on its constant reinterpretation.81

81 See B. Talmud Berachot 63:2, describing God saying to Moses, “you and I shall learn and unveil the Law (of the Torah)”, or the words of B. Talmud Avoda Zara 3:2: “In the first three hours of the day, the Lord sits and studies the Torah.” Index of Primary Sources

Hebrew Bible

Genesis 9:14b 159, 163 1 6, 257–258, 262, 9:14–15 147 266 9:14–16 144 1:3 25, 258 9:14–17 163 1:26 165 9:15 146, 159,161, 163 1:26–28 165 9:16 143, 147, 158–159, 1–9 257, 263 161, 163 2:7 32 n. 37 9:16a 159 3 171 9:16b 159 3:7 259 n. 55 9:17 147, 159, 161, 174 3:12 160 9:17a 159 3:13 160 11:1–9 35, 264 3:14a 160 11:3 35 3:15 160 11:7 36 3:16 160 11:9 36 3:17 160 12:10–20 129, 129 n. 52 7:12 146 20:1–18 129 n. 52 8:2 146–147 20:12 129 n. 52 9 13, 142 n. 7, 148, 21:16 148 164–165 21:20 148 9:1–7 164–165 26:1–11 129 n. 52 9:3 164 n. 55 27 13, 168, 170, 179 9:6 4 n. 11, 164 27:1–39 167 9:8–17 13, 136–137, 142, 142 27:3 148 n. 7, 143, 27:3–4 169 147–148, 156, 27:5b 169 158–160, 165–166 27:7 169 9:9 159, 162 27:7–10 169 9:9a 159 27:11–13 169 9:9–11 161–162, 174 27:14–17 169 9:9–13a 159 27:18 169 9:10 159, 162 27:18–29 169 9:11 146, 159, 162 27:20 169 9:11a 160 27:21 169 9:12 147, 159, 161–162 27:22 169 9:12–17 143 n. 8, 155, 156 n. 27:24 169 50 27:26 169 9:12a 159 27:30 169 9:13 143–144, 147, 158, 27:30–31 169 158 n. 54, 159, 161 27:32 169 9:13–16 146 28:10–22 264 9:13a 160, 162 28:12 266 n. 76 9:13b 159, 163 28:13 259 n. 55 9:14 143, 158, 159, 28:17 266 n. 76 161–163 28:18 142 9:14a 159, 163 29 13, 168–170, 179 270 index of primary sources

Genesis Cont.

29:15–30 167 8:24–25 67 29:21 170, 170 n. 10 8:28 67 29:23 170, 170 n. 10 9:1 59 29:25 169 9:1–2 67 29:25 170 9:7 67 31 171, 179 9:12 67 31:19 175 9:13 59, 67 31:19–55 167, 172, 175 9:14–16 67 31:32 175 9:17 67 31:33 175 9:22–26 67 31:34 176 9:28 67 31:35 175–176 9:30 67 31:37 176 9:35 67 34 127 n. 42, 171 10:2 67 37 171 10:3 59 41:6 63 n. 33 10:3–4 67 42:23 171 10:7 59, 67 48:22 148, 157 n. 53 10:10 67 49:23 157 n. 53 10:12–13 67 49:23–24 149 10:20 67 49:24 148 10:27 67 11:1 67 Exodus 11:4 68 2:23 58 11:10 67 3:2–6 264 12:29–31 68 3:5 266 n. 76 12:33 67 3:7–10 58 12:37 64–65, 67 4:1–5 67 12:42 68 4:15 110 13:15 67 4:21 67 13:17 67 4:23 59, 67 13:17–19 69 n. 57 5:1–2 67 13:20 64–65, 67, 67 5:2 67, 254 n. 48, 69 6:1 67 13:20–22 69, 69 n. 57 6:11 67 13:21–22 69, 71 n. 63 7:2 67 14 11, 53–56, 63–64, 7:5 67 64 n. 36, 67, 67 7:8–13 67 n. 50, 68–69, 73 7:13 67 14:1–31 53 7:14 67 14:1 69 n. 57 7:16 59, 67 14:2 64–66, 66 n. 44, 7:17 67 67, 67 n. 48, 69 7:19–25 67 14:4 63, 67 7:22 67 14:5 62, 67 7:26 59 14:8 67, 69 n. 58 7:26–27 67 14:9 66 8:1–3 67 14:10 56, 58–60, 60 8:4 67 n. 24 8:6 67 14:10–11 60 n. 24 8:15 67 14:10–12 56–59, 60 n. 24 8:16 59 14:11 59, 62 8:16–17 67 14:11–12 59, 59 n. 21, 60, 61 8:18 67 n. 25 index of primary sources 271

Exodus Cont.

14:12 58 Leviticus 14:13 56–57, 73–74 24:16 197 n. 35 14:13–14 56, 59–60, 60 n. 24, 70 Numeri 14:14 56 5:23–24 25 14:15 61 n. 26, 66, 70, 70 9:15–23 66 n. 60 11 25 14:16 60, 67, 71, 75 11:1–15 60 14:16–17 61 11:2 70 14:17 61 n. 26, 67 11:26–27 25–26 14:18 67 11:46 58 14:19 66 14:2 58 n. 16 14:19–20 69, 71 n. 63 14:2–3 58 14:20 71 n. 63 14:5 60 14:20–21 68 14:14 66 14:21 60–61, 63, 70–71 14:25 69 n. 57 4:21–22 71 16:3 58 n. 16 14:22 72 20:2 58 n. 16 14:24 57, 68–69, 71 n. 63 20:3 58 n. 16 14:25 72 n. 66 20:6 60 14:26–27 61 20:7–11 60 14:27–28 72 n. 66 20:8–11 68 14:28 57 21:2 58 n. 16 14:29 72 21:4 69 n. 57 14:30 57, 74 21:5 58, 58 n. 16 14:30–31 58, 61 n. 25, 62 21:7 70 n. 31, 70, 74 23:16 110 14:31 56, 60, 75, 208 31:16 117 n. 71 15 67 n. 50 Deuteronomium 15:4 69 n. 57 1:32–33 66 15:11 189 1:40 69 n. 57 15:19 69 n. 58 2:1 69 n. 57 15:22 67 n. 48, 69 n. 57 2:15 57 15:25 70 5:17 135 n. 70 16:2 58 n. 16 5:19 199 16:3 58 6:4 189 17:2 58 n. 16 20:3 56 n. 10 17:4 60, 70 30:11–12 247 17:5 67 33:29 189 17:6 60 34:10–12 75, 75 n. 71 20:13 135 n. 70 22:8 117 Joshua 23:2 247 9 13, 167, 172, 179 23:31 69 n. 57 9:1–27 171 23:37 57 9:4–5 172 25:22 199 9:5–6 173 31:18 25 9:6 173 32:4 207 9:9 173 32:32–33 25 9:9–13 173 33:12–23 264 9:13 173 33:20 265 9:14 172 33:23 265 9:18 173 40:36–38 66 9:21 172–173 272 index of primary sources

Joshua Cont.

9:23 173 5:12 78 9:27 173 6:1–3bα 78 10:1 172 n. 14 6:4 78 10:10 57 6:10–14 78 10:25 56 n. 10 6:16 78 6:19–7:1 78 Judges 7:7–11 56 n. 10 1:1–2 105 n. 19 7:10 57 1:2 105 n. 20 12:16 74 3 117–118, 123 n. 26, 12:16–17 74 127 12:18 74–75 3:15 119, 121 n. 18 14:22–23 74 3:15–30 171 14:23 74 3:16 118 n. 8, 119, 121 14:37 105 n. 19 15–1 Kings 2 88 3:17 123 16–2 Samuel 5 79 3:19 115–117, 119, 120 19 171 n. 14, 121 n. 17, 122, 19–21 78 130, 133, 133 n. 60, 22:10 105 134–135 23:2 105, 105 n. 19 3:19–20 119 25:20 118, 118 n. 11 3:20 118 n. 10, 130, 135, 28:6 105 135 n. 68, 136 30:8 105 n. 19 3:20–21 119 31:3 149 n. 28, 157 n. 53 3:21 122 n. 22 3:21–22 118 n. 9 2 Samuel 4 171 1:18–27 149 n. 28, 157 n. 53 4:15 57 2–7 84 4:16 57 3 13 4–5 127 n. 42 3:24–25 174 7:15–23 171 3:26 174 8 171 3:26–27 167, 172, 174, 179 9:54 171 3:27 174 n. 16 16 127 n. 42 5:19 105 n. 19 16:4–30 171 6 12, 83–85, 87–88 18:10 117 6:1–15 78, 84 20:15–16 122, 122 n. 23 6:14 86–87 20:23 105 n. 19 6:16 78, 84, 86 20:28 105 n. 19 6:17–19 86 21 171 6:17–20a 84 6:20 87 1 Samuel 6:20–23 84 3:20 73 7:11b 78 4 94–96 7:16 78 4:1b–18a 78 7:23 189 4:11 95 n. 59 9 83 4:17 95 n. 59 9:1–10:5 78 4:19 95 n. 59 9–20 79, 83 4:19–22 95 n. 60 9–24 84 4:21 95 n. 59 10:6–11:1 78 4:21–22 95 n. 58 10–12 89 n. 42 5:1–11bα 78 11 89 n. 49 index of primary sources 273

2 Samuel Cont. 11:2 111 17:11b 105 11:2–12:7a 78 17:14 81, 101–102, 108, 112 11:3 111 17:14b 108–109 12:9 114 17:15 101, 106–108 12:11–12 113 17:15–16 107 12:12 111 17:16 106 12:13–25 78 17:21 101, 107 12:14 114 17:23 107–108 12:14–15 92 n. 50 17:24 107 12:20 82 17–20a 78 12:26–31 78 18:18 82 13:1–14 172 18:8–16 262 n. 63 13:1–14:24 78 18/19–20:22 78 13:8 82 19:9 [8] 100 14:3 110 19:15 [14] 100 14:7 103 2 Sam 6:20ff 78 14:14 103 22 149 n. 28 14:19 110 22:8–21 62 14:28–18:17 78 22:15 57 15:4 112 22:35 149 n. 28, 157 n. 53 15:6 100 23:34 111 15:9 91 23:8–39 90 15:12 101 15:13 100 1 Kings 15:19–22 106 n. 26 1:1–21 78 15:30–37 108 1:20 79 15:31 99, 108 1:25 100 15:32 100, 108 1:27 79 15:34 113 1:34 100 16:15 107 1:39 100 16:16 100–101 1–2 83, 88 16:16–19 105 2:5–10 78 16:20 99, 102, 105, 112 2:5–46 81 16:20–17:14 101 2:12–27a 78 16:21 111 2:28–46 78 16:22 111 2–4 83 16:23 99, 101, 105, 113 3:16–28 172 16–17 12, 99, 100 8:8 199 16–21 105 9:26 69 n. 57 17:1 108 12 102, 109 17:1–3 104–105, 109–110, 12:7 103 112 12:10–11 103 17:1–14 101, 105 12:13–14 102 17:3 103–104, 111 12:15 109 17:4 101, 108–109 12:24 109 17:4b 110 15:5 117 17:5 102, 109 21:1–16 172 17:6 102 22 102–103 17:7 101 22:5 114 17:7–13 108 22:6 103 17:8 104 22:7 102 17:11 101, 107, 112 22:12 103 274 index of primary sources

1 Kings Cont.

22:15 103 51:36 36–37 22:17 103 51:37 37 22:20–21 111 51:41 37 22:23 109 51:44 37 22:34 150 n. 28, 157 n. 53 Ezekiel 2 Kings 1 264, 265 n. 74, 266 1:16 114 1:26 147 n. 22, 262 2 72, 75 n. 63 2:8 72 1:28 143, 144, 148, 266 n. 2:14 72 76 9 171 8–11 95, 95 n. 60 9:24 150 n. 28, 157 n. 53 12:2 259 n. 55 13:14–18 150 n. 28, 157 n. 53 17:10 63 n. 33 13:17 150 n. 28 19:6 94 18:20 117 19:12 63 n. 33 27:26 63 n. 33 Isaiah 29 63 6 264, 266 30 63 6:1 265 30:8 63 6:4 262 n. 63 30:12 63 6:5 265 30:19 63 6:6 266 n. 76 30:26 63 6:7 266 n. 76 6:8 111 Joel 7:4–9 57 n. 15 2:25 234 11:15–16 64 n. 36 2:25–26 232 35:5 259 n. 55 36:6 94 Hosea 41:2–3 150, 157 n. 53 3:1 87 n. 37 41:6 262 n. 63 13:15 63 n. 33 43:1b–2 62 43:16–19 62, 64 n. 36 Obadiah 44:27 62 1:17 232, 235 50:2 62 50:4 231, 234 Jonah 51:9–11 62 4:8 63 n. 33 54:3 233, 235 61:7 232, 234, 234 n. 55 Nahum 1:3 146 n. 18 Jeremiah 1:3–4 63 n. 32 7 96 7:4 96 Zechariah 7:8 96 9:12 232, 234 7:12 96 7:14–15 96 Psalms 18:17 63 n. 33 18:8–16 262 n. 63 23:29 249 18:8–20 62 36:18 25 18:35–36 150 n. 32 36:27–28 25 21:13 150 n. 30 49:21 69 n. 57 25:14 253 51:34 37 46:10–11 150 51:34–37 36, 63 n. 32 48:8 63 n. 33 index of primary sources 275

Psalms Cont.

49 266 6:17 47 68:26 192 6:18 47, 48 n. 25, 49 74:15 62 n. 25 78 64 6:18–20 46, 47 n. 20 78: 12–13 64 n. 36 6:19 47 78:14 66 6:19–20 48 90:2 262 n. 63 6:20 47 95:5 262 n. 63 6:21 46–47, 47 n. 20 102:26 262 n. 63 7:11 51 103 263 10:1 51 103–104 261 24:15 118, 118 n. 11 103:19–104:35 259 26:8 146 n. 18 104 257–259, 262–263, 27:21 63 n. 33 266 28:12–28 200 n. 43 104:1–2 261 104:2 258, 259 n. 53 Song of Songs 104:3 261 1:1 192–193, 195 n. 30, 104:5–8 261 196 105:39 66 1:2 202 119:89 259 n. 55 1:3 206 114 64 n. 36 1:8 206 144:2 262 n. 63 1:9 202 1:13 199 Proverbs 1:16 189 1:20–32 200 n. 43 2:4 207 2:7 192 2:5 87 n. 37 8:22 200 2:9 194 8:30–31 201 2:14 197 9:3 200 n. 43 3:7 196 9:5 231, 233 4:1 189 11:14 102 4:3 198 13:10 101 4:5 199 15:22 102 4:9 198 20:13 231, 233 4:16 202 21:14 118 5:2 189 23:21 231, 233 5:16 189 24:6 102 7:3 200 25:2 19, 253 8:5 188 25:11 256 n. 43 8:6 188 25:23 118 8:11 196 n. 34 8:12 196 Job 3:20 51 Qohelet 3:23–48 48 n. 25 1:10 117 4:9 49 n. 25 12:11 250 4–5 46 5:15–20 48 Esther 6 46 1:13–21 102 n. 6 6:14 46–47, 47 n. 20, 49 6:14–21 11, 39, 40, 46, 52 Daniel 6:15 47 Dan 1:15ff 123 n. 25 6:15–17 46, 47 n. 20 6:15–20 49 Ezra 6:16 47, 49 4:5 108 276 index of primary sources

Nehemia 2 Chronicles 4:9 108 18:4 114 4:15 108 20:17 74 9:12 66 9:19 66

1 Chronicles 12:2 122 n. 23

Extracanonical Works

Ben Sira 47:25 196 n. 33 43:11 148 50:7 143, 148 43:12–13 143

Targumim

Tg. Judg 3:15 122 n. 22 Tg. Judg 3:19 134

Other Versions

LXX 2 Sam 17:8 105–106 Judg 3:15 122 n. 22 Prov 9:17 134 Judg 3:19 134 Psalm 9 134

Dead Sea Scrolls

4Q51 2 Sam 12:14 114, 114 n. 50 2 Sam 12:9 114

Ancient Near Eastern Texts

Ugaritic Enuma Elish KTU 1.2 iv 11–13, VI 80–94 155 19–20 24 Epic of Gilg. KTU 1.6 vi 49–50 24 XI 43–47 34 KTU 1.6 54–58 29 K. 2582 rev ii, KTU 1.16 v.10–27 111 x + 21 33 Myth of Anzû Akkadian II 53–60 154 5R 46 23 156 Myth of Zu 9 111 ARM 26 197 13–14 105 n. 18 ARM 199 44 105 n. 18 Egyptian ARM 202 10–11 105 n. 18 Ramesside Dream Astrolabe B—KAV Manual 4:16 32 218 B I 14–16 156 Setne I col. 4/1–4 23 Enuma Elish I 1–2 21 Spell 1130 § 465a 32 Enuma Elish IV 35–50 154 index of primary sources 277

Classical Texts

Aristotle, Met., I,1.1059a18– Aristotle, Phys. 5.1062b11 253 n. 35 II,2.194b14–15 253 n. 35 Aristotle, Met. Quintilian, Inst. IV,7.1064b6–13 253 n. 35 6.3.47ff 130 n. 55 Aristotle, Met. Rhet. Her. II.XXVI.40 130 n. 55 VI,1.1026a10–22 253 n. 35 Rhet. Her. IV.LIV 130 n. 55 Aristotle, Met. XI,7.1064a31– 1064b5 253 n. 35

Josephus

Ant. 5.185–197 130 n. 54 Ant. 7.202 103 n. 10

Rabbinic Literature

Mishnah b. ‘Erub. 13:2 248 n. 14 m. ’Abot 1:1 255n. 39 b. ‘Erub. 13:2 250 n. 22 m. ’Abot 5:25 250 b. Ḥag. 13:2 266 m. Ḥag. 2:1 253 b. Ketub. 105b 224 n. 30 m. Yad. 3:5 183 n. 1 b. Ketub. 106a 223 n. 29 b. Ketub. 61a 224 n. 30 Tosefta b. Mak. 11a 224 n. 30 t. Yoma (= Kippurim) b. Meg. 7a 183 n. 1 2.14 199n. 39 b. Meg. 28b 223 n. 28 t. Yoma (= Kippurim) b. Menaḥ. 29:2 256 n.45 2.15 199 b. Ned. 50a 224 n. 30 t. Sanh. 12:10 184 n. 2 b. Nid. 73a 223 n. 28 t. Soṭah 9.8 188 b. Pesaḥ. 94b 223 n. 28 b. Pesaḥ. 112a 223 n. 28 Sifra b. Qidd. 80b 223 n. 28 Sifra Shemini 1.15–16 196 n. 33 b. Sanh. 34:1 249 b. Sanh. 92a 223 n. 28 Sifrei b. Sanh. 113a–b 223 n. 28 Sifrei Bamidbar §139 205 n. 60 b. Šabb. 46a 230 n. 46 Sifrei Devarim §10 205 n. 62 b. Šebu. 35b 196 Sifrei Devarim §36 205 n. 61 b. Ta‘an 22a 224 n. 30 Sifrei Devarim §41 205 n. 61 b. Tamid 32b 223 n. 28 Sifrei Devarim §304 204 n. 58 b. Yebam. 16a 222 n. 26 Sifrei Devarim §305 204 n. 58, 206 b. Yebam. 42b 230 n. 46

Talmud Yerushalmi Midrash Rabbah y. Roš Haš. 58:4 251 Exod. Rab. 15:22 257, 258 n. 48 Gen. Rab. 1.10 3 n. 9 Talmud Bavli Gen. Rab. 18:6 31 b. ‘Abod. Zar. 3:2 268 n. 81 Gen. Rab. 31:8 31 b. ‘Abod. Zar. 5b 223 n. 28 Gen. Rab. 34:14 4 n. 11 b. ‘Abod. Zar. 9a 223 n. 28 Gen. Rab. 99:3 125 n. 32, 128 b. B. Bat. 7a 224 n. 30 n. 46, n. 47 b. B. Meṣ. 59:1–2 247 Lev. Rab. 7.3 200 n. 41 b. Ber. 9a 230 n. 46 Lev. Rab. 22:1 256 n. 44 b. Ber. 63:2 268 n. 81 Num. Rab. 19:4 256 n. 45 278 index of primary sources

Midrash Rabbah Cont.

Qoh. Rab. 217 n. 15 Mek. de-Rabbi Ishmael, Ruth Rab. 2:9 125 n. 32, 128 Masekhta Baḥodesh, n. 46, 129 n. 50 Yitro 8 198 Song Rab. 1:1 251 Mek. de-Rabbi Ishmael, Song Rab. 1.1.10 186 Pisḥa, Bo 5 205 n. 59 Song Rab. 1.1.11 190, 197 Mek. de-Rabbi Ishmael, Song Rab. 1.2.1 186, 192, 196 Pisḥa, Bo 7 (22) 203 n. 53, 205 Song Rab. 1.5.1 205 n. 59 n. 59 Song Rab. 1:8 251 n. 26, 259 Mek. de-Rabbi Ishmael, Song Rab. 1.9.1 191 n. 23 Masekhta de-Pisḥa, Song Rab. 1.12.1 203 n. 53, 203 Bo 14 196 n. 33, 205 n. 62 n. 54, 207 n. 68 Mek. de-Rabbi Ishmael, Song Rab. 1:52 251 n. 27 Pisḥa, Bo (25) 205 n. 59 Song Rab. 2.1.1 205 n. 59 Mek. de-Rabbi Ishmael, Song Rab. 2.2.2 203 n. 53 Va-yeḥi, Beshalaḥ 1 Song Rab. 2.4.1 207 (87–88) 205 n. 59 Song Rab. 2.7.1 195 n. 32 Mek. de-Rabbi Ishmael, Song Rab. 2.9.1 193 Shirah, Beshalaḥ, 3 205 n. 62 Song Rab. 2.14.1 197 Mek. de-Rashbi Song Rab. 2.14.3–6 203–204, 208 19.17 204 n. 58, 205 Song Rab. 2.16.1 186, 189 Midr. Asseret Song Rab. 4.1.1 205 n. 59 ha-Dibrot 214 n. 7 Song Rab. 4.4.2 198 Midr. Pesikta Rabati Song Rab 4.4.9 198 n. 36 Eccl 12:11 250 Song Rab. 4.8.1 198 n. 36 Midr. Prov 16:10 222 n. 26 Song Rab. 4.9.1 198 Midr. Ps 137:2 222 n. 26 Song Rab. 4.12.1 198 n. 36 Midr. Song 2:1 202 Song Rab. 5:11 201 n. 46 Midr. Song 2:15–16 250 n. 24 Song Rab. 6.8.1 205 n. 61 Midr. Song 2:19 250 n. 24 Song Rab. 7.3.2 200 Midr. Song 2:30–31 250 n. 24 Song Rab. 8.13.1 195 n. 32 Midr. Song 5:10–16 185 Midr. Tanhuma, Other Rabbinic Works Hukat 8 256 n. 45 ’Abot R. Nat. 183 n. 1, 200 n. 41, Midr. Tannaim to 203 n. 54, 206 Deut 1:4 205 n. 60 n. 67 Midr. Tannaim to DEZ 1,25 227 n. 37 Deut 14:2 204 n. 58 DEZ 2,4 227 n. 37 Midr. Tannaim to Kallah Rab. 3,1 227 n. 37 Deut 15:9 205 n. 61 Kallah Rab. 4,19 227 n. 37 Midr. Tannaim to Kallah Rab. 4,31 223 n. 28 Deut 20.1 205 n. 60 Mek. de-Rabbi Yishmael Pesiq. Rab. 32 223 n. 28 on Exod 14:10 250 n. 24 PRE 1 224 n. 30 Mek. de-Rabbi Yishmael PRE 18/25 229 n. 43 on Exod 19:16 250 n. 24 PRE 25 (b. ‘Erub. 18b) 229 n. 42 Mek. de-Rabbi PRE 29 224 n. 29 Ishmael 4 251 n. 28 PRE 37 (b. Soṭah 41b) 229 n. 42 Mek. de-Rabbi Ishmael, PRE 38 (b. Ḥul. 4a) 229 n. 42 Masekhta de Shirata, PRE 40 (b. Sanh. 40a) 229 n. 42 Beshalaḥ 3 204 n. 58, 189 PRE 50 (b. Meg. 15a) 229 n. 42 n. 18, 205 S. ‘Olam Rab. 7 203 n. 50 index of primary sources 279

Shir ha–Shirim Shir ha–Shirim Zuta 1.1 194 Zuta 1.15 189 n. 19 Shir ha–Shirim Shir ha–Shirim Zuta 1:13 199 Zuta 11 196

Medieval Jewish Sources

Seder Eliyahu Comm. on Genesis 1 258 n. 52 PSEZ 18, 24f 222 n. 26 Comm. on Ps 104:1 258 n. 52 SER 1 221, 225 n. 33 Comm on Ps 104:9 259 n. 53 SER 5 222 n. 26 SER 10 221, 223 n. 27 Maimonides SER 11 223 n. 27 Guide of the Perplexed SER 15 221 I:2, 23–26 259 n. 55 SER 16 221 Guide of the Perplexed SER 17 223 n. 27 I:15, 40–41 259 n. 55 SER 17 223 n. 27 Guide of the Perplexed SER 18 222 n. 26 I:34, 77 253 n. 35 SER 26 222 n. 26 Guide of the Perplexed SER 31 225 n. 33 I:37 265 SER 53 223 n. 27 Guide of the Perplexed SER 83 223 n. 27 I:38 265 SER 86 223 n. 27 Guide of the Perplexed SER 91 222 n. 26 I:62, 152 253 n. 35 SEZ 1 223 n. 28, 225 Guide of the Perplexed SEZ 1, 169 222 n. 26 I:63, 164 255 n. 40 SEZ 1, 191 236 n.56 Guide of the Perplexed SEZ 2, 171 221 n. 24 2:25, 330 257 n. 46 SEZ 3, 175 222 n. 26 Guide of the Perplexed SEZ 4 227 n. 37 2:25, 330 257 n. 46 SEZ 4, 178 227 Guide of the Perplexed SEZ 6, 182–83 228 n. 39 III: 43 249 n. 19 SEZ 8, 185–86 228 n. 39 SEZ 9, 187–89 228 n. 39 Samuel Ibn Tibbon SEZ 14 235, 237 Comm. on Eccl., H § SEZ 14, 195–196 221 n. 24 17–23 256 n. 42 SEZ 14, 196–197 231 Comm. on Eccl., H § SEZ 15, 197 222 n. 26, 223 19/E § 17 266 n. 78 n. 28 Comm. on Eccl., H § SEZ 15 225 45–46/E § 41–51 263 n. 66 Comm. on Eccl., H § Rashi 80–187/E § 163–169 263 n. 67 Judg 3:19 125 n. 32 Comm. on Eccl., H § Song 1:1 196 n. 34 166/E § 149, 263 n. 67 Comm. on Eccl., H § Judah Hallevi 194–206/E § Kuzari 5:10 258 n. 49 176–187 263 n. 67 Comm. on Eccl., H § Abraham Ibn Ezra 215–217/E § Comm. on Genesis, 195–196 263 n. 67 Introduction 252 n. 30, 259 Comm. on Eccl., H § n. 53 243/162 258 n. 51 280 index of primary sources

Samuel Ibn Tibbon Cont. Comm. on Eccl., H § MYH § 117 258 n. 51 252/E § 220 258 n. 51 MYH § 135 258 n. 51 Comm. on Eccl., H § MYH § 161 258 n. 51 262/E § 229 258 n. 51 MYH § 187 267 n. 80 Comm. on Eccl., H § MYH § 211 265 n. 71 266/E § 239 258 n. 51 MYH § 225 258 n. 51 Comm. on Eccl., H § MYH § 229 266 n. 76 287/E § 243 258 n. 51 MYH § 232 256 n. 43 Comm. on Eccl., H § MYH § 234 266 n. 76 294/E § 262 258 n. 51 MYH § 255 256 n. 43 Comm. on Eccl., H § MYH § 263 258 n. 50 324–347/E § MYH § 264 258 n. 51 298–320 263 n. 67 MYH § 300 276 n. 76 Comm. on Eccl., H § MYH § 317 265 n. 75 332/E § 320 258 n. 51 MYH § 325 266 n. 77 Comm. on Eccl., H § MYH § 340–341 266 n. 76 335–336/E § MYH § 349 266 n. 76 309–310 255 n. 40 MYH § 356 266 n. 76 Comm. on Eccl., H§ MYH § 394 258 n. 51 347/E § 331 258 n. 51 MYH § 449 258 n. 51 Comm. on Eccl., H § MYH § 467 258 n. 51 348–349/E § MYH § 619 258 n. 51 321–322 263 n. 67 MYH § 686 258 n. 51 Comm. on Eccl., H § MYH § 710 256 n. 43 358/E § 358 258 n. 51 MYH § 762 256 n. 43 Comm. on Eccl., H § MYH § 766 256 n. 43 375–380/E § MYH § 781 256 n. 43 348–352 263 n. 67 MYH § 787 256 n. 43 Comm. on Eccl., H § MYH § 900 262 n. 64 387/E § 741 258 n. 51 MYH § 919 258 n. 51 Comm. on Eccl., H § MYH § 920 258 n. 51 446–450/E § MYH § 922 259 n. 53 424–428 263 n. 67 MYH § 926 258 n. 51 Comm. on Eccl., H § MYH § 947 258 n. 51 475/E § 465 266 n. 79 MYH § 995 262 n. 63 Comm. on Eccl., H § MYH § 1000 256 n. 43 491–492/E § MYH § 1045 256 n. 43 472–473 263 n. 67 MYH 6 1002–1003 259 n. 53 Comm. on Eccl., H § MYH § 1083 262n. 62 640–654/E § MYH § 1084 263 n. 65 620–634 263 n. 67 MYH § 1085 262 n. 63 Comm. on Eccl., E § MYH § 1086 256 n. 43 19, 20, 48, 279, 308, 338, MYH § 1089 262 n. 63 359, 471, 753, 756 256 n. 43 MYH § 1096 262 n. 63 MYH § 1116 262 n. 63 MYH § 11 259 n. 56 MYH § 1124–1128 262 n. 63 MYH § 32 261 n. 61 MYH § 1125–1128 262 n. 63 MYH § 61 258 n. 51 MYH § 1127 256 n. 43 MYH § 65 257 n. 47 MYH § 1167 262 n. 63 MYH § 66 258 n. 51 MYH § 1178 258 n. 51 MYH § 67 263 n. 67 MYH § 1192 262 n. 63 MYH § 81 258 n. 51 MYH § 1221–1222 258 n. 51 MYH § 111 258 n. 51 MYH § 1300–1307 255 n. 38 index of primary sources 281

Samuel Ibn Tibbon Cont. MYH § 1307 256 n. 43 Radaq MYH § 2192–2193 264 Judg 3:19 120 n. 14

Miscellaneous

Avicenna, Kitâb al-Shifâ, at-Tabiyyât 5, 76–79 260 n. 59